Textbook of Clinical Occupational and Environmental Medicine

Citation preview

© Harcourt Health Sciences Company 1994 © 2005, Elsevier Inc. All rights reserved. First edition 1994 Second edition 2005 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 or otherwise, without either the prior permission of the publishers or a licence permitting restricted copying in the United Kingdom issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Permissions may be sought directly from Elsevier’s Health Sciences Rights Department in Philadelphia, USA: phone: (+1) 215 238 7869, fax: (+1) 215 238 2239, e-mail: [email protected]. You may also complete your request on-line via the Elsevier homepage (http://www.elsevier.com), by selecting ‘Customer Support’ and then ‘Obtaining Permissions’. ISBN 0 7216 8974 4 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book is available from the Library of Congress

Notice Medical knowledge is constantly changing. Standard safety precautions must be followed, but as new research and clinical experience broaden our knowledge, changes in treatment and drug therapy may become necessary or appropriate. Readers are advised to check the most current product information provided by the manufacturer of each drug to be administered to verify the recommended dose, the method and duration of administration, and contraindications. It is the responsibility of the practitioner, relying on experience and knowledge of the patient, to determine dosages and the best treatment for each individual patient. Neither the Publisher nor the editors assume any liability for any injury and/or damage to persons or property arising from this publication. The Publisher

Printed in China Last digit is the print number 9 8 7 6 5 4 3 2 1

Contributors Michelle R Addorisio MD MPH

Daniel E Banks MD

University of Connecticut Health Center Farmington, CT USA

Professor of Medicine Professor and Chair Department of Internal Medicine Louisiana State University Health Sciences Center Shreveport, LA USA

Bruce H Alexander PhD Assistant Professor Division of Environmental Health Sciences School of Public Health University of Minnesota Minneapolis, MN USA

Thomas J Armstrong PhD Professor, Industrial and Operations Engineering School of Public Health The University of Michigan Ann Arbor, MI USA Michael D Attfield PhD Surveillance Branch Chief Division of Respiratory Disease Studies National Institure for Occupational Safety and Health Morgantown, WV USA Edward L Baker MD MPH Professor and Director North Carolina Institute for Public Health Chapel Hill, NC USA John R Balmes MD Professor of Medicine Division of Occupational and Environmental Medicine University of California San Francisco, CA Professor of Environmental Health Sciences University of California, Berkeley San Francisco, CA USA

Rebecca Bascom MD MPH Professor of Medicine Penn State College of Medicine Milton S Hershey Medical Center Hershey, PA USA

Shirley Bassiri MD Internal Medicine Residency Program Department of Medicine Columbia Presbyterian Medical Center New York, NY USA

Alan R Berger MD Professor of Neurology and Associate Chairman Department of Neurology University of Florida Jacksonville, FL USA

Patricia Blackwell MD MPH Medical Director Occupational Health Clinic Occupational Health and Prevention Services Centers for Disease Control and Prevention Atlanta, GA USA Paul Blanc MD MSPH Professor of Medicine Division of Occupational Medicine University of California, San Francisco San Francisco, CA USA

Carl Andrew Brodkin MD MPH Clinical Associate Professor of Medicine and Environmental and Occupational Health Sciences University of Washington Seattle, WA USA Sandy Bogucki MD PhD Associate Professor Section of Emergency Medicine Yale University School of Medicine New Haven, CT USA Chris Carlsten MD Senior Fellow, Occupational and Pulmonary Medicine Occupational & Environmental Medicine Program University of Washington Seattle, WA USA Tania Carreón MSc PhD Associate Fellow Division of Surveillance Hazard Evaluations and Field Studies National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention, Assistant Professor Department of Environmental Health University of Cincinnati Medical Center Cincinnati, OH USA Stephanie Carter MSPH CIH Research Assistant and Doctoral Candidate Department of Environmental and Occupational Health Sciences University of Washington Seattle, WA USA

x■CONTRIBUTORS

Moira Chan-Yeung MB FRCPC

William E Daniell MD MPH

David L Eaton PhD DABT FATS

FRCP Professor of Medicine Respiratory Division Department of Medicine University of British Columbia Vancouver, BC Canada

Associate Professor Department of Environmental and Occupational Health Sciences University of Washington Seattle, WA USA

Professor and Director Center for Ecogenetics and Environmental Health University of Washington Seattle, WA USA

Paul S Darby MD PhD MPH CIME Clinical Instructor Department of Family Medicine University of Washington Seattle, WA USA

Ellen A Eisen ScD Professor Department of Health and Environment University of Massachusetts at Lowell Lowell, MA USA

John M Dement PhD CIH Professor, Division of Occupational and Environmental Medicine Department of Community and Family Medicine Duke University Medical Center Durham, NC USA

Maadhava Ellaurie MBChB Associate Professor of Medicine Penn State College of Medicine Milton S Hershey Medical Center Hershey, PA USA

Harvey Checkoway PhD Professor Departments of Environmental and Occupational Health Sciences and Epidemiology University of Washington Seattle, WA USA

Martin G Cherniack MD MPH Professor of Medicine Division of Occupational and Environmental Medicine University of Connecticut Health Center, Director University of Connecticut Ergonomics Technology Center Farmington, CT USA H Gregg Claycamp MS PhD CHP Director, Scientific Support Staff Food and Drug Administration Center for Veterinary Medicine Rockville, MD USA David Eric Cohen MD Director, Occupational and Enviromental Dermatology Department of Dermatology New York University Medical Centre New York, NY USA

Paul A Demers PhD Associate Professor School of Occupational and Enviromental Hygiene University of British Columbia Vancouver, BC Canada

Jeffrey Derr MD MPH Occupational and Environmental Medicine Residency Program School of Public Health University of Illinois at Chicago Chicago, IL USA

James R Donovan Jr MD MS Professor Department of Environmental Health University of Cincinnati Cincinnati, OH USA

Mark R Cullen MD Professor of Medicine and Public Health, Director Occupational and Environmental Medicine Program Yale University School of Medicine New Haven, CT USA

Alan M Ducatman MD MS Professor and Chair Department of Community Medicine West Virginia University School of Medicine Morgantown, WV USA

Derek E Dunn (deceased) PhD National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Cincinnati, OH USA

Bradley A Evanoff MD MPH Associate Professor of Medicine Division of General Medical Sciences Washington University School of Medicine St Louis, MO USA Peter S Evans PhD Senior Research Scientist Institute for Environmental Health, Inc. Seattle, WA USA Karin D E Everett PhD Research Engineer Scientist Department of Biology University of Washington Seattle, WA USA

Nancy Fiedler PhD Associate Professor University of Medicine and Dentistry of New Jersey Robert Wood Johnson Medical School Environmental and Occupational Health Sciences Institute Piscataway, NJ USA

CONTRIBUTORS■xi

Lawrence J Fine MD DrPH Medical Advisor Office of Behavioral and Social Sciences Research Office of the Director, National Institutes of Health Bethesda, MD USA

Susan H Forster MD Associate Clinical Professor Department of Ophthalmology and Visual Science Yale University School of Medicine, Chief of Ophthalmology Yale University Health Services New Haven, CT USA

Michael L Fischman MD MPH Associate Clinical Professor and Assistant Chief Division of Occupational and Environmental Medicine Department of Medicine University of California San Francisco, CA USA

Mark W Frampton MD Professor of Medicine and Environmental Medicine University of Rochester School of Medicine and Dentistry Rochester, NY USA

Alfred Franzblau MD Jordan A Firestone MD PhD MPH Assistant Professor of Neurology, University of Washington, Staff, Harborview Medical Center Seattle, WA USA

Professor of Occupational Medicine University of Michigan School of Public Health Ann Arbor, MI USA

Mary Carol Fromes (deceased) MD Lora E Fleming MD PhD MPH MSc Associate Professor Department of Epidemiology and Public Health and Division of Marine Biology and Fisheries School of Medicine and Rosenstiel School of Marine and Atmospheric Sciences University of Miami Miami, FL USA

Brian G Forrester MD Assistant Professor University of Alabama School of Medicine Birmingham, AL USA

MPH University of Michigan School of Public Health Ann Arbor, MI USA

Howard Frumkin MD MPH DrPH FACP FACOEM Professor and Chair Department of Environmental and Occupational Health Rollins School of Public Health of Emory University, Professor of Medicine Emory Medical School Atlanta, GA USA David H Garabrant MD MPH

Linda S Forst MD MPH MS Associate Professor University of Illinois at Chicago School of Public Health Environmental and Occupational Health Sciences Chicago, IL USA

Professor of Occupational Medicine and Epidemiology University of Michigan School of Public Health Ann Arbor, MI USA

Julie L Gerberding MD MPH Director, Centers for Disease Control and Prevention, Administrator, Agency for Toxic Substances and Disease Registry Atlanta, GA USA

Prajakta Ghatpande MSc MS Research Scientist Institute for Environmental Health, Inc. Seattle, WA USA

Craig S Glazer MD Assistant Professor Department of Internal Medicine University of Texas Southwestern Medical Center at Dallas Dallas, TX USA

Michael Gochfeld MD PhD Professor of Environmental and Occupational Medicine Enviromental and Occupational Health Sciences Institute University of Medicine and Dentistry of New Jersey Robert Wood Johnson Medical School Piscataway, NJ USA Bernard D Goldstein MD Dean Graduate School of Public Health University of Pittsburgh Pittsburgh, PA USA Daniel A Goldstein MD Senior Science Fellow Director of Medical Toxicology Monsanto Company Regulatory Affairs St Louis, MO USA

Audrey R Gotsch PH CHES Interim Dean University of Medicine and Dentistry of New Jersey School of Public Health New Brunswick, NJ USA

James W Grosch PhD Research Psychologist National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Cincinnati, OH USA

xii■CONTRIBUTORS

Mridu Gulati MD Occupational and Environmental Medicine and Pulmonary Clinical Fellow Department of Internal Medicine, Yale University School of Medicine New Haven, CT USA

Mats Hagberg MD PhD Professor and Director Department of Occupational Medicine Sahlgrenska Academy at Göteborg University, Chief Physician and Director Department of Occupational and Environmental Medicine Sahlgrenska University Hospital Göteborg Sweden

William E Halperin MD DrPH Professor and Chair Department of Preventative Medicine and Community Health New Jersey Medical School University of Medicine and Dentistry of New Jersey Newark, NJ USA

Thomas E Hamilton MD PhD MPH Clinical Associate Professor of Endocrinology and Metabolism Division of Endocrinology and Metabolism, University of Washington Medical Center; Fred Hutchinson Cancer Research Center, Seattle, WA; Private Practice, Endocrinology and Metabolism Bellevue, WA USA

Philip Harber MD MPH Professor of Family Medicine, Chief, Division of Occupational and Environmental Medicine University of California, Los Angeles Los Angeles, CA USA

Amanda Hawes JD Partner Alexander, Hawes, and Audet San Jose, CA USA Frank J Hearl BSChE SMChE PE Senior Advisor National Institute for Occupational Safety and Health Washington, DC USA

Robin Herbert MD Assistant Professor of Community and Preventative Medicine Mount Sinai Medical Center New York, NY USA

Christina A Herrick MD PhD Assistant Professor of Dermatology Yale University School of Medicine New Haven, CT USA

Marilyn V Howarth MD Assistant Professor of Occupational and Emergency Medicine University of Pennsylvania School of Medicine, Director of Occupational and Environmental Consultation Services Hospital of the University of Pennsylvania Philadelphia, PA USA

Lea Hyvärinen MD PhD Honorary Professor, Rehabilitation Science, University of Dortmund Dortmund, Germany, Senior Lecturer, University of Oulu and University of Tampere, Helsinki Finland Louis F James MD MPH & TM

Robert F Herrick MS ScD Senior Lecturer Department of Environmental Health Harvard School of Public Health Boston, MA USA

FACPM Colonel, United States Air Force (retired) Medical Corps Senior Flight Surgeon, Former Acting Director, Hyperbaric Service, St Mary’s Hospital West Palm Beach, FL, Former Co-Director, Hyperbaric Service, Mariner’s Hospital Tavernier, FL USA

Jessica Herzstein MD MPH

Joel D Kaufman MD MPH

Global Medical Director Air Products & Chemicals Inc. Lexington, MA USA

Associate Professor and Director Occupational & Environmental Medicine Program Department of Occupational and Environmental Health Sciences University of Washington Seattle, WA USA

Michael J Hodgson MD MPH Director Occupational Safety and Health Program Department of Veterans Affairs Office of Public Health and Environmental Hazards Washington, DC USA

Christer Hogstedt MD PhD Professor and Research Director National Institute of Public Health Stockholm Sweden

Matthew C Keifer MD MPH Associate Professor Departments of Environmental and Occupational Health Sciences and Medicine University of Washington Seattle, WA USA

Karl T Kelsey MD MH Professor of Cancer Biology and Environmental Health Harvard School of Public Health Boston, MA USA

CONTRIBUTORS■xiii

Susan M Kennedy PhD Professor, School of Occupational and Environmental Hygiene, Director, Centre for Health and Environment Research University of British Columbia Vancouver, BC Canada

Kathleen Kreiss MD Field Studies Branch Chief Division of Respiratory Disease Studies National Institute for Occupational Safety and Health Morgantown, WV USA

Edwin M Kilbourne MD

Anthony D LaMontagne ScD MA

Chief Medical Officer National Center For Environmental Health/ Agency for Toxic Substances and Disease Registry/United States Department of Health and Human Services/Office of the Director, National Institutes of Health Atlanta, GA USA

MEd Associate Professor Centre for the Study of Health and Society The University of Melbourne Parkville Victoria Australia

Ulrike Luderer MD PhD MPH Assistant Professor of Medicine Center for Occupational & Environmental Health University of California, Irvine Irvine, CA USA Ingvar Lundberg MD

Howard M Kipen MD MPH Director and Professor of Occupational Health Environmental and Occupational Health Sciences Institute University of Medicine and Dentistry of New Jersey Robert Johnson Medical School Piscataway, NJ USA

Tord Kjellstrom BMed MEng PhD (Medicine) Professor of Public Health National Institute of Public Health Stockholm Sweden Jeffrey L Kohler PhD Director, Pittsburgh Research Laboratory National Institute for Occupational Safety and Health Pittsburgh Research Laboratory Pittsburgh, PA USA

Anne Krantz MD MPH Assistant Professor of Medicine, Rush Medical College, Chief, Section of Toxicology Division of Occupational Medicine John H Stroger Jr Hospital of Cook County Chicago, IL USA

Philip J Landrigan MD MSc Professor and Chairman Department of Community and Preventative Medicine Mount Sinai School of Medicine New York, NY USA

Professor National Institute for Working Life and Department of Public Health Sciences Karolinska Institute Stockholm Sweden

Gregory J Ma MSPH SM(AAM) Microbiology Supervisor King County Environmental Laboratory Seattle, WA USA

Senior Partner The Polyclinic Seattle, WA USA

Steven Markowitz MD Professor and Director Center for the Biology of Natural Systems Queens College City University of New York New York, NY USA

Carola Lidén MD

Carmen J Marsit PhD

Professor of Occupational and Environmental Dermatology Department of Medicine Karolinska Institute Stockholm Sweden

Research Fellow Harvard University School of Public Health Boston, MA USA

Stan Lee MD

James E Lockey MD MS Professor and Director Division of Occupational and Environmental Medicine University of Cincinnati Cincinnati, OH USA William T Longstreth Jr MD MPH Professor of Neurology Harborview Medical Center University of Washington Seattle, WA USA

Christopher J Martin MD MSc Residency Director and Assistant Professor Institute of Occupational and Environmental Health Department of Community Medicine West Virginia University School of Medicine Morgantown, WV USA Thomas P Matte MD MPH Medical Epidemiologist National Center for Environmental Health Centers for Disease Control and Prevention Atlanta, GA USA

xiv■CONTRIBUTORS

Donald R Mattison MD Adjunct Professor Department of Environmental Health Sciences Mailman School of Public Health Columbia University New York, NY USA Rob McConnell MD Associate Professor of Preventative Medicine, Keck School of Medicine University of Southern California Department of Preventative Medicine Los Angeles, CA USA James A Merchant MD PhD Dean, College of Public Health Professor of Occupational and Environmental Health University of Iowa College of Public Health Iowa City, IA USA

Gabrielle Morris MD Diving Medicine Physician Seattle and King County Department of Health, Occupational, Aviation and Diving Medicine Physician, US Healthworks, FAA Senior Aviation Medical Examiner, Seattle, WA USA

Linda Rae Murray MD MPH Chief Medical Officer – Primary Care Ambulatory and Community Health Network Cook County Bureau of Health Services Chicago, IL USA

James Nethercott MD (deceased) University of Maryland, Johns Hopkins Hospital Baltimore, MD USA

Gun Nise PhD Robert R Miksch PhD Chief Research Scientist Institute for Environmental Health, Inc. Seattle, WA USA Ben Hur P Mobo Jr MD MPH Instructor of Medicine Occupational and Environmental Medicine Program Yale University School of Medicine New Haven, CT USA

Sandra N Mohr MD MPH Associate Medical Director New York Life Insurance Company New York, NY USA Jacqueline M Moline MD MSc Assistant Professor Community and Preventative Medicine and Internal Medicine Mount Sinai School of Medicine New York, NY USA

Assistant Professor and Senior Occupational Hygienist Department of Public Health Sciences Division of Occupational Medicine Karolinska Institute Stockholm Sweden

Peter Orris MD MPH FACP FACOEM Professor and Director Occupational Health Services Institute University of Illinois School of Public Health, Professor of Internal and Preventative Medicine Rush University School of Medicine Cook County Hospital Chicago, IL USA

Adelisa L Panlilio MD MPH Medical Epidemiologist Centers for Disease Control and Prevention Atlanta, GA USA

Edward L Petsonk MD Senior Medical Officer Division of Respiratory Disease Studies National Institute for Occupational Safety and Health Morgantown, WV USA Michael Pulley MD PhD Assistant Professor of Neurology Department of Neurology University of Florida, Jacksonville Jacksonville, FL USA

Adrianna Quintero JD Attorney Environment and Health Outreach Program Director of Latino Outreach Natural Resources Defense Council San Francisco, CA USA Peter M Rabinowitz MD MPH Assistant Professor of Medicine Occupational and Environmental Medicine Program Yale University School of Medicine New Haven, CT USA Carrie A Redlich MD MPH Professor of Medicine Occupational and Environmental Medicine Program and Pulmonary Critical Care Section Yale University School of Medicine New Haven, CT USA

Thomas S Rees PhD Associate Professor of Otolaryngology – Head and Neck Surgery University of Washington, Director of Audiology Harborview Medical Center Seattle, WA USA

David M Rempel MD MPH Professor of Medicine University of California, San Francisco Richmond, CA USA

CONTRIBUTORS■xv

Stephen J Reynolds PhD CIH Professor of Industrial Hygiene Department of Environmental and Radiological Health Sciences Colorado State University Fort Collins, CO USA Caroline S Rhoads MD Associate Professor of Medicine Harborview Medical Center University of Washington Seattle, WA USA Frederick P Rivara MD MPH The George Adkins Professor of Pediatrics, Adjunct Professor of Epidemiology The Harborview Injury Prevention and Research Center University of Washington Seattle, WA USA Mark G Robson BS MS PhD MPH ATS Chairman Department of Environmental and Occupational Health University of Medicine and Dentistry of New Jersey School of Public Health Piscataway, NJ USA

Cecile S Rose MD MPH Associate Professor of Medicine Pulmonary and Occupational Medicine National Jewish Medical and Research Center Denver, CO USA Linda Rosenstock MD MPH Dean, School of Public Health Professor of Medicine and Environmental Health Sciences University of California, Los Angeles Los Angeles, CA USA

Mark A Rothstein JD Professor, Director of the Institute for Bioethics, Health Policy and Law, Herbert F Boehl Chair of Law and Medicine Institute for Bioethics, Health Policy and Law University of Louisville Louisville, KY USA Rachel Rubin MD MPH Division Chair and Assistant Professor of Medicine Occupational and Environmental Medicine Rush Medical College Stroger Hospital of Cook County Chicago, IL USA

Avima M Ruder PhD Senior Research Epidemiologist Division of Surveillance Hazard Evaluations and Field Studies National Institute for Occupational Safety and Health, CDC Cincinnati, OH USA Mark B Russi MD MPH Associate Professor of Medicine and Public Health, Yale University School of Medicine, Director, Occupational Health Yale-New Haven Hospital New Haven, CT USA

Steven L Sauter PhD Chief, Organizational Science and Human Factors Branch National Institute for Occupational Safety and Health/Centers for Disease Control and Prevention Cincinnati, OH USA E Neil Schachter MD Maurice Hexter Professor of Medicine and Community Medicine, Medical Director, Respiratory Care Department Mount Sinai Medical Center New York, NY Paul A Schulte PhD Director Education and Information Division National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Cincinnati, OH USA

David A Schwartz MD MPH Professor of Medicine and Genetics Duke University Medical Center Durham, NC USA Noah S Seixas PhD CIH Professor Department of Enviormental and Occupational Health Sciences University of Washington Seattle, WA USA

Mansour Samadpour PhD Director Institute for Environmental Health, Inc. Seattle, WA USA

Jonathan M Samet MD MS Professor and Chair Department of Epidemiology Johns Hopkins Bloomberg School of Public Health Baltimore, MD USA

Stuart L Shalat BS BA ScM ScD Associate Professor of Exposure Epidemiology Environmental and Occupational Health Sciences Institute Robert Wood Johnson Medical School Piscataway, NJ USA Elizabeth F Sherertz MD Contact Dermatitis Specialist The Skin Surgery Center Winston-Salem, NC USA

xvi■CONTRIBUTORS

Gina Solomon MD MPH Assistant Clinical Professor of Medicine, University of California, San Francisco, Senior Scientist, Natural Resources Defense Council San Francisco, CA USA

Akshay Sood MD MPH Assistant Professor of Medicine Division of Pulmonary and Critical Care Medicine Southern Illinois University School of Medicine Springfield, IL USA

Nancy L Sprince MD MPH Professor of Occupational and Environmental Health and Internal Medicine Department of Occupational and Environmental Health University of Iowa College of Public Health Iowa City, IA USA

Lawrence B Stein PhD Psychologist Private Practice Red Bank, NJ USA

Tim K Takaro MD MPH MS Clinical Assistant Professor Department of Environmental and Occupational Health Sciences University of Washington Seattle, WA USA Susan M Tarlo MBBS FRCP(C) Professor of Medicine and Public Health Sciences University of Toronto Toronto Western Hospital Toronto, ON Canada Peter S Thorne MS PhD Professor of Toxicology and Environmental Health Department of Occupational and Environmental Health The University of Iowa, The College of Public Health Iowa City, IA USA Mark J Utell MD Professor of Medicine and Environmental Medicine, Director, Pulmonary/Critical Care and Occupational Medicine Divisions University of Rochester Medical Center Rochester, NY USA

Daniel Wartenberg PhD Professor Department of Environmental and Community Medicine Environmental and Occupational Health Sciences Institute University of Medicine and Dentistry of New Jersey Robert Wood Johnson Medical School Piscataway, NJ USA

Kalman L Watsky MD Section Chief of Dermatology Hospital of Saint Raphael, Associate Clinical Professor of Dermatology Yale University School of Medicine New Haven, CT USA

Nargues A Weir MD FACCP Attending Physician Northern Virginia Pulmonary and Critical Care Associates Inc. Annandale, VA USA

Laura S Welch MD Medical Director The Center to Protect Workers’ Rights Silver Springs, MD USA

Frances J Storrs MD Professor Emerita Department of Dermatology Oregon Health and Science University Portland, OR USA

Gregory R Wagner MD Director, Division of Repiratory Disease Studies National Institute for Occupational Saftey and Health Morgantown, WV USA

Catharina Wesseling MD PhD

Niel Wald AB MD

Ellen Widess JD Senior Program Officer The Rosenberg Foundation San Francisco, CA USA

Coordinator, Health Section of Central American Institute for Studies on Toxic Substances Universidad Nacional Heredia Costa Rica

Jaime Szeinuk MD Assistant Professor of Community and Preventative Medicine Mount Sinai School of Medicine New York, NY USA

Oyebode A Taiwo MD MPH Assistant Professor of Medicine Occupational and Environmental Medicine Program Yale University School of Medicine, New Haven, CT USA

Professor of Environmental and Occupational Health Graduate School of Public Health University of Pittsburgh Pittsburgh, PA USA

Elizabeth M Ward PhD Director Surveillance Research Department of Epidemiology and Surveillance Research American Cancer Society Atlanta, GA USA

David A Youngblood MD MPH & TM FACPM Former Director Hyperbaric Oxygen Department Roper Hospital Charleston, SC, Haleiwa, HI USA

Preface to the First Edition

The field of occupational and environmental disease is rapidly evolving. Questions and concerns about the health consequences of exposures to a seemingly endless array of potential hazards in the workplace and elsewhere in the environment are increasingly raised by clinicians themselves or brought to their attention by patients or other interested parties. Yet most practitioners find themselves ill-equipped to recognize, diagnose and treat occupational and environmental diseases. Further, textbooks in the field have not traditionally been dedicated to helping clinicians meet this challenge by presenting clinically relevant information that is both comprehensive yet practical and easily accessible. The Textbook of Clinical Occupational and Environmental Medicine was conceived and written to address the needs of students, trainees, and clinicians who seek a resource to integrate occupational and environmental medicine into routine clinical practice. It evolved in part from our well-received, earlier effort—a concise and less encyclopedic text entitled Clinical Occupational Medicine. This textbook, although we hope it benefitted by our experiences with the first, is by intent and necessity markedly different in approach and scope. The role of workplace factors has been broadened to include environmental physical, chemical, and biologic agents that may have adverse effects on human health. Environmental diseases, although in general less well recognized and understood than those arising from the workplace, are integrated throughout the sections of the textbook. The textbook is divided into four major sections. The first, Principles and Practice, provides a broad overview of the specialized skills central to the successful practice of occupational and environmental medicine, recognizing the strong interrelationship in the field between scientific and ethical, legal, economic, and social issues. The second section encompasses three core disciplines that are necessary complements to the diagnosis, treatment, and prevention of occupational and environmental diseases: toxicology, epidemiology and industrial hygiene. The third and longest section provides an organ system approach that enables the clinician to consider potential occupational and environmental diseases as they most commonly present in an individual patient. The last section, strongly cross-referenced to the third, enables the reader to consider specific toxins or hazards. Each chapter in this section is organized by exposure type (e.g., radiation, biologic factors, metals) and presented to provide an understanding of the environmental and occupational settings where specific agents are likely to be encountered, their acute and chronic health effects, and approaches to treatment and prevention of exposure to them. The contributors to the textbook are well known and recognized experts from North America and Europe. The editors have worked closely with them to provide a consistent format throughout the textbook. We are extremely grateful for the spirit in which our contributors responded to this objective by accommodating to our evolving effort to achieve for the reader a coherent, comprehensive text. Long periods of silence on the part of some contributors were thankfully rewarded with remarkable chapters. One was preceded by a welcome fax from Greg Wagner, which—like a clear day in Seattle that may follow weeks of rain—served to brighten memories of the recent past. His doggerel is excerpted below

xviii■PREFACE As midnight came and the bell tolled off the printer the pages rolled Now sitting with the rising sun I know at least a draft is done! The editors would also like to acknowledge the outstanding secretarial and editorial assistance provided by Paula Sandler, Rebecca Hubbard, Anne Gienapp, Lanita Stewart, and Marjorie E. Marenberg. We also want to thank our colleagues at W.B. Saunders for encouraging us to proceed and at all turns strengthening the product, especially John Dyson, Ray Kersey, David Kilmer, Pat Morrison, and Carol DiBerardino. And finally, we would like to acknowledge the influence of Bernadino Ramazzini, the Italian physician credited as the father of occupational medicine, who taught us many things, but in his extraordinary 1713 treatise, Diseases of Workers, understood very well the occupational hazards of authors. He wrote The Author to His Book Since you itch and are burning to be published, First pay heed to an anxious father’s warning, Briefly here’s what the fates have destined for you, Since you bring them something novel, All the learned at once will run to greet you. Two short pages, I think are all they’ll read through, Then they’ll fling you to factories or by-streets. Where the poorest buy sausages and fish sauce. Yes, you’re fated to wrap up something greasy. Still, cheer up, for the same thing often happens Now to massive imposing legal Pandects. All-receivers they are; they wrap up mackerel, Or we screw them to hold the grocer’s pennorth, Pepper, maybe, or smelly seed of cumin. You must know you were born in grimy workshops Not in elegant mansions of our betters, Not in glittering courts where chief physicians Lay own laws for the cooks but sit down nowhere. Trust me, then, you will suffer this more lightly Than do books that can boast of prouder titles. If they read you and straightway send you packing Back to workshops, remember—you were born there. From Diseases of Workers, translated from the Latin text De Morbis Artificium of 1713 by Wilmer Care Wright Publishing Company, New York, New York 1964. LINDA ROSENSTOCK MARK R. CULLEN

Preface to the Second Edition

The Textbook of Clinical Occupational and Environmental Medicine was originally conceived and written to address the needs of students, trainees and clinicians, as a resource to integrate occupational and environmental medicine into routine clinical practice. We found the response to the first edition extremely rewarding – the text was critically acclaimed and sustained broad national and international distribution. It enjoyed use by many working in different capacities in the occupational and environmental field, not merely clinicians. In the decade since publication of the first edition, although many of the challenges in diagnosis, treatment and prevention remain, much in the world of occupational and environmental medicine has changed. Workers and others in the United States and abroad face new and emerging threats due to dramatic changes in globalization, technology and demographics. Even the merger of the two components, occupational and environmental – which a decade ago was still being questioned – is now all but taken for granted. It is against this background that the editors of the first edition have drawn upon the prodigious efforts of two colleagues for this second edition. As with the first edition, we have worked closely with our distinguished contributors to provide a consistent format throughout the textbook and we are extremely grateful for the spirit in which our contributors responded to this objective. We have also undertaken some changes: chapters have been deleted, many more added, and all others significantly updated. In addition to reorganizing the initial four sections, a new section has been added to provide an overview of major workforce sectors, recognizing that an appreciation of the overall health risk in a given setting, such as mining, involves more than an isolated understanding of individual exposures. Realignments in the publishing world, parallel to consolidations and acquisitions seen in other parts of the economy this past decade, brought us through a series of changes to our London-based publisher, Elsevier. We are grateful to our Elsevier colleagues, especially Rolla Couchman, Hilary Hewitt, Susan Stuart and Amy Head for facilitating this process, and for their support and constant professionalism. The editors would also like to acknowledge the outstanding support and editorial assistance provided by Linda Oliva and Gauri Balani. As editors, we have in common wondrous appreciation for the support of family and friends – with special recognition by Linda (Lee, Adam and Matthew), Mark (Michele, Zoe and Esme), Drew (Kayla and Naomi), and Carrie (Mara, Joshua, Evelyn and Norman). We also share a deep and abiding respect for workers throughout the world, from whom we have learned so much and to whom we dedicate this book. Linda Rosenstock Mark R Cullen Carl Andrew Brodkin Carrie A Redlich

Chapter 1 Introduction to Occupational and Environmental Medicine Mark R Cullen, Linda Rosenstock, Edwin M Kilbourne Work and economic development are fundamental to the human condition, and have contributed to enormous advancement in wellbeing and health over the centuries. Unfortunately, virtually every aspect of work entails risk of harm as well, both because of the physical nature of the activities involved and the intimate relationship in which workers are placed to natural and man-made hazards in the environment. Moreover, the activities of production themselves and the products generated by work have introduced into the broader environment innumerable hazards of everyday life: in the air we breathe, the food and water we consume, and the material goods we use and ultimately must dispose of. Occupational and environmental medicine (OEM) is an evolving medical specialty that seeks to identify and modify the adverse effects of these hazards on the health of individuals and larger populations. The focus of OEM differs from many other specialties, which may encompass interest in particular agents (e.g., infectious disease), organ systems (e.g., pulmonary medicine), mechanisms of injury (e.g., immunology) or curative approaches (e.g., surgery). In these more traditional disciplines, the focus of attention is on factors that an individual patient or his/her clinician, or both, can modify to prevent, ameliorate or cure disease or the consequence of injury. In OEM practice, although individuals are of paramount importance, equally important are external factors beyond the direct control of individuals or physicians. The same external factors affecting one person often threaten the health of others. There is a natural tension between the clinical health issues relating to the individual patient and broader public health issues, which transcend the traditional doctor–patient relationship. These broader issues fall very centrally in the domain of OEM practice. The focus of this text, however, is the clinical side, placed in the context of these broader perspectives. Historically the field of OEM has developed out of two formerly distinct disciplines: occupational medicine and environmental medicine. Though common clinical and public health perspectives, scientific underpinnings, training needs and practice approaches have fostered this ‘merger’ during the past several decades, there remain differences in the two components by dint of their distinct population foci, divergent societal approaches and distinctive histories. We shall begin here by describing the central principles of disease that irrevocably bind the two; in the sections that follow, unique issues relating to the scope and practice will be introduced.

PRINCIPLES OF OCCUPATIONAL AND ENVIRONMENTAL DISEASE 1. The clinical and pathologic expression of most environmentally caused diseases are indistinguishable from those of nonenvironmental origin There is a widely held belief among medical practitioners outside of OEM that diseases of occupational or environmental origin are both rare and distinctive. In reality, diseases caused by work and ambient environment are neither rare nor often distinctive in their clinical presentations and laboratory findings. Most occupational diseases, such as occupational cancers, not only resemble diseases caused by other factors but are otherwise indistinguishable except by careful documentation of a history of a relevant exposure. Other occupational and environmental diseases, like asthma or dermatitis, may be distinguished clinically only through obtaining an exposure history, with or without specialized testing, which presupposes high suspicion for the diagnosis. Only a minority of occupational diseases, such as pesticide and heavy metal poisonings, are sufficiently distinctive that they are likely to be uniquely identified by routine laboratory testing procedures.

2. Many diseases of occupational or environmental origin are multifactorial, with non-environmental factors playing a contributory role The majority of chronic diseases and even a fair number of acute ones are multifactorial in origin. Coronary artery disease, for example, cannot be simply attributed to hypertension alone in a patient who also smokes. In fact, the discovery of one cause not only does not preclude the possibility of a second, it often makes the effect of an environmental exposure more likely. For example, it has been well established that asbestos exposed workers who smoke have a far higher likelihood of lung cancer than individuals exposed to cigarettes or asbestos singly. Similarly, alcohol consumption is known to potentiate the effects of some environmental hepatotoxins by causing hepatocellular disease. The most important consequence of this principle in practice is that the potential role of an environmental toxin is not reduced by the presence of another pathogenic factor; in fact, it may be increased. This holds true for

4 Introduction to Occupational and Environmental Medicine common types of clinical complaints that result from exposure to the environment, such as irritation and sensitization of the skin and respiratory tract. Such problems are too often ascribed to causes such as smoking or viral infection, leaving remediable occupational and environmental causes undetected.

3. The effects of occupational and environmental exposures occur after a biologically predictable latent interval following exposure Agents or chemicals capable of causing direct and acute injury to the body will typically exert their effects either immediately or soon after exposure. In these cases, because the onset of disease occurs early, possible causal connections are relatively easily identified. On the other hand, the effects of agents that act by sensitizing the immune system, such as those that cause dermatitis or asthma, more often are exhibited only after a period of months to years of exposure. Other substances initiate insidious disease processes that may become clinically apparent only after a latent interval of many years. For example, carcinogens may not cause cancer until years after the individual’s first exposure. Importantly, there is no uniform relation between these late outcomes and any early effects. For example, leukemia may occur in a person exposed to external ionizing radiation at levels far below that which would cause acute radiation sickness or other demonstrable health effect. Indeed, individuals unaffected by early effects may be at higher risk for later effects because they tolerate doses of higher intensity and duration than those who do suffer acute effects, and consequently remove themselves from further exposure, or are removed.

4. The dose of an exposure to a noxious agent is the strongest predictor of the likelihood and type of effect Although this principle is elucidated in detail in Chapter 5, it is important to recognize that toxins, like drugs, have clear relationships between dose of exposure and subsequent effect, and proportion of exposed individuals affected. Although each host differs from others, knowledge of these relationships and estimation of the amount of each are key to diagnostic decision-making. In general, higher exposures confer a higher likelihood of being affected (dose–response relationship) and of more serious effects (dose–effect relationship). As shown in Figure 1.1, three distinct patterns can be discerned. For direct acting toxins, such as heavy metals, organic solvents, or pesticides (Fig. 1.1a), there is for each individual a threshold dose below which there is no demonstrable effect. As the dose increases, the severity of effect increases up to a level that ultimately, at least theoretically, would be fatal. In addition, as the dose increases, the proportion of the exposed population adversely affected also increases. Other harmful agents act by eliciting an immunologic or other hypersensitivity response (Fig. 1.1b). With these agents, such as those that cause asthma, dermatitis, and

allergic alveolitis, many persons experience no untoward effect regardless of dose, presumably based on genetic or other host factors. Increasing dose, however, increases the likelihood of sensitization in those who are susceptible. Once sensitization occurs, however, the severity of reactions is typically independent of subsequent exposure dose and may occur at a very low level. Finally, there are agents that interact with genetic material to cause mutations or initiate cancers (Fig. 1.1c). With these agents, the administration of even the smallest dose confers a finite chance of a mutation. The risk of disease at the lowest end of the dose–response curve may be only theoretical or unmeasurable (e.g., the risk of lung cancer from smoking one cigarette). In this situation, as with agents inducing hypersensitivity, the dose does not affect the severity of disease once it is present; only the probability of disease increases as the dose of exposure increases. Although these relationships are different, the importance of dose in arriving at a correct diagnosis, providing treatment, and preventing disease remains crucial in every case. As discussed more fully in Chapter 3, successful evaluation and management of the patient suspected of having a disorder of environmental origin or who has a risk for such a disorder depend on the clinician’s ability to assess the patient’s recent and past exposure dose, at least qualitatively. Primary prevention, on the other hand, depends on minimizing exposure dose in the first place.

5. People differ substantially in their responses to noxious exposures Humans sometimes differ remarkably in their responses to environmental exposures. These differences may be due to a wide range of factors, including genetic differences in metabolism, age, gender or size, co-exposures to environmental substances that may interact with agents of interest, coexisting morbid conditions, or complex behavioral factors. The major principles that underlie this variability are discussed in Chapter 11. For present purposes, it is important to recognize such variability, which frequently obscures the relationship between environmental exposures and health effects. For example, a health problem in one among many individuals exposed to a harmful agent may suggest an alternative explanation for the effect when in reality only that single individual was susceptible to that dose of exposure. Similarly, the fact that many workers have functioned without adverse consequence around a chemical or process may wrongly suggest to an employer or susceptible coworker that the environment is equally safe for all. From the practitioner’s perspective, it is important to recognize that although the pattern of occurrence of illness in a population may be a vital clue and should always be sought, the absence of an obvious pattern or a confusing pattern may be a function of host variability within the population.

OCCUPATIONAL MEDICINE In developed countries the vast majority of adults – men and women – are active in the formal economy, inter-

Occupational Medicine 5

Proportion affected (%)

100

Susceptible population

Proportion affected (%)

100

0

0

Threshold

B

Dose

Proportion affected (%)

100

Severity

Lethal

Clinical

Subclinical

No effect

0 Threshold

A

Dose

C

Dose

Figure 1.1: Schematic dose–response relations for environmental hazards. (a) For direct-acting agents, there is a threshold for each individual, followed by increasing severity with rising dose. Similarly, a rising fraction of the exposed population is affected as dose rises; eventually everyone is affected. (b) Indirect-acting toxins affect only susceptible individuals. The dose determines what proportion of these individuals are affected. Note that dose may not determine the severity of the reactions, which may be more related to host factors. (c) Carcinogens are believed to cause disease as a linear function of dose. The shape of the dose–response curve at very low doses (dotted line) is difficult to determine directly from studies; it is often assumed to be linear, but this assumption is controversial for some carcinogens.

rupted occasionally for reproduction, medical disabilities and economic dislocations such as layoffs. Virtually every form of work has attendant hazards – physical, chemical, biologic and social; these are discussed in Sections 2 and 4 of this text. Although some work environments are complex, involving many different hazards, and the potential human effects of many hazards remain incompletely studied, the range of possibilities is nonetheless circumscribed and, in theory at least, can be reconstructed and characterized – even quantified – in the effort to assess risk for injury and illness, and evaluate health problems that do arise. For complex historical and social reasons, however, attention to this relationship has been very limited until the past few decades, even in developed countries (Chapter 2). In developing countries, where far higher fractions of the adult population are either unemployed or engaged in informal work, characterization of the health effects of work is far less complete (Chapter 10). Despite significant recent advances in worker health promotion, there remains as yet no reliable estimate of the magnitude of the medical and economic burden posed by occupational diseases and injuries. Efforts to determine the nature and extent of occupational diseases are more elusive than for injuries; the latter, in general, involving more physically apparent outcomes with more easily traced causes.

Chapter 31 provides an overview of data sources and information about the distribution of both fatal and non-fatal occupational injuries. In this section we review some of the problems defining the extent of occupational disease and injury in developed countries, focusing on what is known and not known in the United States. Readers can find more detailed information about the incidence and prevalence of specific disease entities in the organ system chapters of Section 3 of this text. Two main factors contribute to the lack of reliable estimates about the extent of occupational diseases: clinical under-recognition that such diseases are in fact occupational and inadequacy of public surveillance systems to capture and summarize those that may be recognized clinically. Under-recognition, in turn, is based on a number of factors, including: 1. inadequate knowledge about the health effects of many hazards among employers, workers and health practitioners; 2. the absence of pathognomonic or even relatively specific findings for many occupational diseases; 3. latency between exposure and disease onset for most chronic conditions; 4. multifactorial causation for many occupational diseases, including those due to both occupational and

6 Introduction to Occupational and Environmental Medicine environmental factors and non-environmental ones; and 5. marked variations in individual susceptibility to most occupational hazards. Compounding these complexities in disease recognition are the inadequate attention devoted to occupational diseases in medical education in the United States and the historical isolation that largely excluded attention to occupational factors and conditions from mainstream medical care. Beyond diagnostic difficulties lie fragmented and often biased reporting systems. Moreover, even if reporting systems functioned better, there are a number of significant disincentives for the physician to report an occupational disease. These include: burdensome paperwork and bureaucratic interactions; lack of payment, underpayment, and payment delays if cases are contested by employers; and the wish to avoid litigious involvement.

The burden of occupational injury and disease: sources and limitations of data The following section provides a brief overview of some of the state and national data sources that have been used in deriving estimates of the nature and extent of the occupational disease burden in the United States. Although data systems in other developed countries differ, the fundamental limitations are common to all to varying degrees. The situation in developing economies is far more intractable, as discussed in Chapter 10.

Annual survey of occupational injuries and illnesses In the United States the Bureau of Labor Statistics (BLS) program of OSHA, mandated by the 1970 OSH Act, maintains a national employer based system to monitor jobrelated illnesses and injuries and summarizes these data in annual surveys.1 The BLS data are obtained by annual compilation of required injury and illness reporting logs from all eligible employers, with substantial penalties invoked for non-compliance. Many shortcomings of this reporting system limit its reliability and completeness including exclusion of many workers (the self-employed; small businesses and farms; and federal, state, and local agency employees); under-reporting of most occupational diseases (especially chronic, multifactorial and latent ones); and the disincentives inherent in any system linking reporting rates to regulatory priorities, however otherwise rational such links might appear. Needless to say, the quality of the aggregate data can be no better than the quality of the data input. Nonetheless, the most recently available data do provide a broad and staggering picture of the magnitude of the occupational health problem in the US, limitations notwithstanding. In the year 2000 almost 5.3 million occupational injuries were self-reported at covered workplaces alone; 1.6 million of these resulted in lost work time. In addition, 362,000 occupational illnesses were reported, including over 200,000 cases of musculoskeletal disorders attributed by employers to work, over 40,000

skin conditions, and almost 80,000 respiratory cases, poisonings, etc – almost all acute in nature.2 In a separate reporting function, 5,344 deaths were reported at work due to occupational causes.

Workers’ compensation reports (see also Chapter 57.2) Workers’ compensation records, both state and federal, provide one of the most widely available sources of data about occupational diseases and injuries. There is ample support for the belief that these databases systematically underestimate the extent of occupational illnesses and distort the distribution of types.3 For example, a recent study conducted in the northeastern United States found that as many as 90% of self-recognized musculoskeletal disorders had not been reported to the workers’ compensation system.4 Workers’ compensation systems also vary as to their inclusiveness of coverage; agricultural workers, for example, are excluded from workers’ compensation benefits in many states. In addition to problems of incomplete worker coverage and failures to report diseases – especially chronic diseases – to this system (whether because of patient or physician inclination not to do so), many of these systems report only those claims that are accepted; widespread experience suggests many are contested by insurance carriers in virtually every jurisdiction. The reverse also occurs: poorly documented claims are sometimes compensated. Nonetheless, some inferences can be derived from these data which form part of the incomplete whole picture.3

Morbidity and mortality from chronic conditions In the United States, some states record the individual’s usual occupation on death certificates; a smaller number of states include industry as well. These data and accompanying information about cause of death and contributory causes (so-called ‘multiple causes of death’) have proved useful in estimating excess mortality in different jobs and industries. Since the late 1960s, the National Center for Health Statistics (NCHS) has provided annual summaries of all conditions on death certificates from all states, including information about occupation and industry from the several states which code this; since 1989 more extensive information about industry and occupation has been coded.5 These data allow exploration not only of the underlying cause of death but also about other conditions that may have contributed to death. Still, there are many limitations to using death certificates in general and multiple causes of death in particular, including: under-and over-reporting of causes by physicians unknowledgeable about occupational diseases; very limited and poor quality information about job and industry from which to infer exposures; and all those factors leading to under-recognition of occupational diseases alluded to earlier. Nonetheless, manipulations with these data have led to various estimates of occupational disease mortality in the

Occupational Medicine 7 US, ranging from 40,000 to 65,000 deaths from chronic occupational disease per year in the US.6 Almost 10,000 of these are directly attributable to asbestos.7 The majority of cases – cancers, and chronic disease of the respiratory tract, heart, liver and kidneys – have been imputed by comparing proportional death rates in various industries, adjusting as best possible for confounding factors such as tobacco and socioeconomic class. For this reason, caution is suggested in the interpretation of these data. Canadian estimates of incidence and mortality, derived from similar databases, suggest comparable experience throughout North America, at least, adjusting for differences in the size of the workforces.8

National Hospital Discharge Survey The National Hospital Discharge Survey (NHDS), conducted annually by the NCHS, provides data on hospital stays in non-governmental hospitals.9 Information is provided regionally. Data on occupations and industries are not available, but the system does allow capture of diagnostic events that may be occupational in origin, e.g., carpal tunnel syndrome or asthma, or other so-called ‘sentinel health events’ – common disorders that have been linked to occupational causes in some significant fraction of all new cases.10 These data, despite significant limitations, have been used in helping to derive estimates of the extent of certain occupational diseases in the United States, such as coal workers’ pneumoconiosis, whose prevalence has General diagnosis category Respiratory (5814 of 6021 diagnoses)*

Musculoskeletal (973 of 1755 diagnoses)

Sensory organ disorders (909 of 965 diagnoses) Symptom-defined and miscellaneous syndromes (443 of 626 diagnoses) Chemical poisoning syndromes (313 of 458 diagnoses) Skin (183 of 254 diagnoses) Psych/Neuro (165 of 395 diagnoses)

resulted in a drop from a peak of 24,000 hospitalizations in 1984 to 11,000 in 1996, and asbestosis, whose prevalence based on these data has increased during the same time period from 6000 a year to 13,000.7

Outpatient data Occupational diagnostic referral clinics developed in the US in the 1970s and have become widespread in medical centers in the 1980s and 1990s.11 These clinics have as their primary goal the identification and management of work-related diseases not readily handled by primary sources of worker care. Although workplace diversity and identified conditions of evaluated patients are limited by local or regional referral practices and patterns, and the availability of such clinics remains extremely limited, data from these clinics provide some insight into common diseases and exposures. Tables 1.1 and 1.2 give an overview of common diagnoses and exposures from 24 clinics, members of the Association of Occupational and Environmental Clinics (AOEC) who submitted case reports to the AOEC database describing patients with work-related diseases or injuries12 who were diagnosed between January 1991 and December 2000. While not necessarily representative of all patients with work-related conditions, these case reports provide insight into the types of occupational conditions being treated by occupational medicine specialists, as well as into the types of exposures that are causing or exacerbating these diseases.

Specific diagnosis

# of cases

% of cases

Asbestosis/parenchymal disease only

3631

40.1

Asthma (691 cases) and reactive airways dysfunction syndrome (156 cases) Asbestos-related pleural disease only Upper respiratory irritation, chronic or NOS Silicosis Parenchymal and pleural disease

847

9.4

642 425 155 114

7.1 4.7 1.7 1.3

318 192

3.5 2.1

144 133 98 88

1.6 1.5 1.1 1.0

909

9.7

Multiple chemical sensitivity/acquired chemical intolerance Sick building syndrome Headache (chemical, migraine, tension, or NOS) Toxic effect of lead Toxic effects of solvents

166 156 121 181 132

1.8 1.7 1.3 2.0 1.5

Dermatitis, all

183

2.0

Toxic encephalopathy

165

1.8

Carpal tunnel syndrome or median nerve neuropathy Tendinitis / tenosynovitis / bursitis of the forearm, wrist, hand, or fingers Epicondylitis Low back strain / sprain / tears Low back pain / radiculopathy / muscle spasm Upper extremity cumulative trauma disorder / musculoskeletal pain Noise-induced hearing loss

*Numbers in parentheses represent number of cases with specific diagnoses listed in table, out of total number of diagnoses in category. NOS = not otherwise specified.

Table 1.1 Diagnoses made in at least 1% of the occupational cases, 1991 to 2000 (n = 9044 cases; 10,882 diagnoses)

8 Introduction to Occupational and Environmental Medicine Exposure Asbestos Noise Repetitive motion Keyboard use Indoor air pollutants Solvents, NOS Inorganic lead Lifting Crystalline silica 1,1,1-Trichloroethane Tetrachloroethylene Fall, NOS Acute trauma, NOS Dust, NOS Smoke, NOS Tuberculosis Paint Chemicals, NOS Vibration, NOS Isocyanates, NOS Coal Latex, natural rubber Welding, NOS Cutting oils Formaldehyde Ergonomic factors, NOS Mold Toluene Hydrocarbons, NOS Xylene Pesticides, NOS Ammonia solution Glutaraldehyde Epoxy resins Methyl ethyl ketone Struck by/against object

Percent of 5641 occupational cases 49.9 16.0 8.3 7.7 7.7 4.8 3.3 3.1 3.0 2.7 2.5 2.5 2.1 1.9 1.6 1.5 1.4 1.4 1.3 1.3 1.2 1.2 1.1 1.0 0.9 0.9 0.9 0.8 0.7 0.7 0.6 0.6 0.6 0.5 0.5 0.5

*Exposures contributing to at least 0.5% of the cases are included in table; some cases have more than one contributing exposure. NOS = not otherwise specified

Table 1.2 Most frequent contributing exposures* associated with occupational conditions, AOEC cases, 1991–2000 (n = 9044 cases; 11,239 exposures)

Over a 10-year period, over 9000 patients were diagnosed with work-related diseases or injuries. The AOEC case report records up to three diagnoses and three exposures; therefore, the numbers of diagnoses and exposures exceed the number of patients. As seen from Table 1.1, asbestosis, diagnosed in 41% of all patients with workrelated conditions, was by far the most common disease seen in these AOEC clinics; asbestos-related pleural disease was also common. This burden of disease from past asbestos exposures should diminish over time, as asbestos exposures are now well controlled in the US. Noiseinduced hearing loss was seen in almost 10% of patients with occupational disease, almost all diagnosed within the patient subgroup with asbestos-related disorders. The AOEC clinics also treat numerous patients with diseases and injuries caused by more recent exposures. Chief among these are respiratory conditions and soft-tissue musculoskeletal disorders. Asthma or RADS (reactive airways dysfunction syndrome) was diagnosed in 9.4% of occupational cases, and upper respiratory irritation in 4.7%. Carpal tunnel syndrome, upper extremity tendinitis/

tenosynovitis/bursitis, epicondylitis, and low back problems together accounted for almost 10% of occupational cases. Table 1.1 also lists several other commonly diagnosed conditions, including dermatitis, lead poisoning, multiple chemical sensitivity, toxic encephalopathy, and sick building syndrome. Table 1.2 details the occupational exposures most frequently related to patients’ injuries or illnesses. Leading the list of associated exposures were: asbestos (49.9%); noise (16%); repetitive motion (8.3%); keyboard use (7.7%); and indoor air pollutants (7.7%). As can be seen from the remaining exposures on the list, occupational diseases and injuries are being caused by very diverse chemical, physical and biologic exposures.

Occupational disease surveillance systems Both the federal government in the US and some states have developed surveillance systems to track incident cases of all occupational diseases or specific entities. Five states, for example, undertook reporting of occupational asthma under a program sponsored by the National Institute for Occupational Safety and Health: 1100 cases were reported between 1993 and 1995.13 Twenty-eight states require reporting by laboratories of blood lead levels for adults as a way of tracking OSHA-mandated blood testing of exposed workers. Results suggest numbers of cases of lead poisoning still occur: Washington State reported almost 2800 levels greater than 25 micrograms/dL between 1993 and 2001;14 Massachusetts reported 547 during a recent 5-year period.15 Connecticut requires reporting of all workrelated diseases, collecting between 1500 to 2000 reports per year since 1996, mostly musculoskeletal disorders, elevated lead levels, skin and lung conditions.16 Despite these and emerging efforts both nationally and regionally, it is noteworthy that no comprehensive system exists to capture occupational diseases and non-acute injuries, so all estimates of incidence and mortality must be viewed with caution as likely underestimates of the true burden.

Estimates of the prevalence of occupational conditions in the population Because incidence data are limited, attempts have been made to estimate the prevalence of occupational effects in the population based on broader measures of the health of the population. The boldest attempt thus far was the Occupational Health Supplement to the National Health Information Survey (NHIS), a periodic populational sample survey conducted every several years by the National Center for Health Statistics. Unfortunately the supplement has only been done once – in 1988 – limiting any inference about changes. Moreover, since the survey is self-administered, only conditions of which respondents may be aware, such as pain or skin rash, are amenable to study. Interpretation is further

Environmental Medicine 9 limited by potential bias in respondents’ perceptions about what conditions are and are not related to their work. Nonetheless, analyses of the 1988 survey have resulted in some evidence that the proportion of the population with work-related musculoskeletal disorders such as low back pain, carpal tunnel syndrome and dermatitis may dwarf estimates from BLS and the other sources described above,17 with literally millions of workers affected by, and to a disturbing extent disabled by conditions perceived to have occurred on the job. While it would be presumptuous to apply these estimates as verified, the extent of positive responses regarding a small number of self-reported conditions further underscores both the large scope of the persisting occupational disease burden in our society, and ultimately the need for more valid, timely and longitudinal data to control it.

ENVIRONMENTAL MEDICINE Since the birth of occupational medicine and continuing through very recent times, interest in the environmental determinants of non-infectious disease has centered overwhelmingly around the workplace. Substantial concern regarding the potential adverse effects on health of nonoccupational environmental exposures has developed only relatively recently. The beginning of this trend is difficult to pinpoint, but in the United States, it has become particularly evident over the last three decades. Today, the non-occupational environment rivals or exceeds the occupational environment as a source of health concerns among the public at large. Hardly a day goes by without news media coverage of one or more potential environmental health problems. At the national and international level, prominent environmental health issues currently include all of the following and more: ■ increased cancer risk from radon in indoor air; ■ neurologic dysfunction from exposure to lead in house dust and drinking water; ■ respiratory and cardiovascular mortality from particulate and other ambient air pollutants; ■ still-to-be-clarified consequences of the accumulation of measurable body burdens of biopersistent halogenated organic compounds (e.g., dioxins and polychlorinated biphenyls) in large segments of the population; ■ possible neurotoxic, carcinogenic, and other effects of substances added to foods; and ■ potentially devastating health consequences that may arise from global warming and depletion of stratospheric ozone. Quantifying the precise levels of risk attributable to environmental exposures in a given population presents difficult methodologic problems. Nevertheless, in one major study, an estimated 75–80% of cancers in the United States were judged to be avoidable, and largely due to environmental factors.18 Tobacco smoke, both actively and passively inhaled, may account for approximately 30% of all cancers in the United States. But other air pollutants are important as well. Over 1000 deaths per year occur in the United States as a result of unintentional exposure

to carbon monoxide, and the prevalence of asthmatic symptoms may double in children as a result of living close to a heavily polluting industrial facility. As many as 10,000 to 20,000 lung cancer deaths occur annually as a result of exposure to environmental radon.19 A substantial proportion of preschool children in the United States, particularly those living in older housing, are at risk of developing measurable intellectual dysfunction owing to environmental lead. At the local level, there is concern about the potential danger to people located close to hazardous waste sites and the possible effects of water or air pollution related to nearby industry. At the level of the individual, the practitioner may see patients whose illness or personal health concerns relate either to one of the major issues described earlier or to more unusual exposures or health problems that are specific to that individual.

Purview of environmental medicine By comparison with occupational health, non-occupational environmental medicine presents both the researcher and the health practitioner with an unusually thorny set of issues and problems. Full consensus has yet to be reached even regarding the boundaries of the new medical discipline, but one could reasonably define nonoccupational environmental medicine as that medical specialty involving the prevention, diagnosis, therapy, and study of disease and injury due to external influences but unrelated to the patient’s workplace. Nevertheless, the potential problems accompanying such a broad definition of the field should be recognized. There is substantial overlap with extant and established specialties, such as infectious disease medicine and pulmonary medicine. Given the paucity of practitioners of this component until very recently, it would seem most appropriate for practitioners of non-occupational environmental medicine to concentrate both clinical and research activities on those problems and issues not already covered by other specialists. Some might prefer a more limited (but potentially viable) definition that would exclude essentially all infectious agents from the purview of non-occupational environmental medicine. This limitation in scope would have the added benefit of focusing the specialty area on the effects and potential effects of chemical and physical environmental agents, which are not dealt with comprehensively in any other specialty area except occupational medicine. Whether certain chemical exposures related to lifestyle (e.g., tobacco smoking and ingestion of dietary lipids) should be excluded from the consideration as problems in non-occupational environmental medicine can also be debated. Despite potential overlap with the fields of allergy and clinical immunology, the potential contribution of the environmental medicine practitioner to the amelioration of health problems due to non-infectious biologic agents is large. There is clearly some overlap between non-occupational environmental medicine and the developing specialty of clinical toxicology. However, clinical toxicologists tend to

10 Introduction to Occupational and Environmental Medicine deal with sudden, generally well-characterized, chemical exposures for which the resulting clinical picture is often acute, rapidly terminating in death or resolution of clinical abnormalities. They deal less with chronic health effects related to cumulative exposure to relatively low doses of toxicants, whereas this area is a specific focus of environmental medicine. The practice of clinical toxicology is directed toward treating persons for clinical symptoms that are already apparent. Clinical toxicology is less oriented toward risk communication and counseling aimed at preventing adverse health outcomes over the long term. Finally, clinical toxicology does not deal with the physical environmental hazards that form an important focus of research and practice in environmental medicine. Even if one defined the field in the most restrictive terms possible, the scope of environmental medicine would remain enormous. Potential chemical hazards are particularly numerous with potential human exposure outside of work to more than 50,000 registered with the US Environmental Protection Agency under the Toxic Substances Control Act (see Chapter 57.1) There are fewer physical than chemical environmental agents of concern. This set of etiologic agents requires very different and diverse bases of scientific knowledge to understand the processes involved in the etiology and pathogenesis of the documented and potential health effects produced by such diverse physical agents as ionizing radiation, electromagnetic fields, hot and cold temperatures, high and low atmospheric pressures, and mechanical force.

Differences from occupational medicine Whereas occupational medicine is a well-established specialty, with a substantial (if still inadequate) number of practitioners focused on reasonably well-defined populations and hazards (earlier cautionary comments notwithstanding), there are relatively few physicians whose practices involve a substantial component of nonoccupational environmental medicine. Moreover, many of those who do practice non-occupational environmental medicine do so only as part of a broader practice involving a recognized specialty or subspecialty (e.g., allergy/ immunology, otolaryngology, neurology, or pulmonary medicine).

Information bases For both the practitioner and the researcher, the non-occupational environment presents challenges, many of which are either absent from or present to a lesser extent in occupational medicine. Of these challenges, the frequent absence of quantitative and even certain qualitative information on exposure is particularly troublesome. The worker at a given site is subject to a particular set of exposures that are determined by the industrial process used there. If he or she becomes ill, a relatively rapid determination can, at least in theory, be made regarding whether a workplace exposure could plausibly account for the

illness. Such a determination could be based on the known adverse effects of the specific materials and processes used and the number, duration, and circumstances of contacts the patient has had with them. It is often possible (and always desirable) to conduct environmental measurements of exposures of concern in the workplace to provide quantitative data on both average and peak exposures (see later). Moreover, because a number of workers may share the same exposures, the presence of a pattern of symptoms among workers with common duties can alert one to the possibility that a workplace environmental exposure may be the cause. Workers themselves may have at least some familiarity with the potential adverse consequences of specific exposures in their industry, and, therefore, they may be more likely to report them to the physician. By contrast, outside of the workplace, peoples’ lifestyles and exposures are extremely varied. Patients rarely maintain an encyclopedic listing of the particular products they use or of other exposures that may be relevant to their complaints or illness. Often exposures affect only a single individual or family, substantially lessening the possibility that an association of a disease with a particular exposure could be identified by means of an analysis of the pattern of health effects seen among several persons. The patient may be unable to reproduce or even accurately report the typical pattern of use of potentially etiologic materials, substantially complicating any effort to quantify exposure. Moreover, particular activities (especially hobbies and avocations likely to involve repeated exposures) may or may not lead to exposures recognized by the patient as possibly relevant to symptoms or illness; the patient may fail to consider that his or her non-workplace exposures might be responsible for a health problem and, therefore, may fail to report them to the physician. A careful history that comprehensively reviews possible causative exposures is, thus, particularly important in non-occupational environmental medicine (see Chapter 3). Comprehensive occupational medical coverage of employees often includes a program of medical surveillance of workers. Such programs are oriented toward the exposure(s) of particular concern in the industry that the program is designed to monitor. Biologic samples (e.g., urine or serum) may be tested periodically to determine whether or not any undue toxicant absorption has occurred. In addition, a worker may receive directed, periodic medical examinations, the purpose of which is the early detection of health effects from the particular substances or processes to which that worker is exposed, hopefully at a stage when those effects can be either mitigated or reversed. If the worker is seen because of a complaint or concern, the surveillance data serve as baseline information that facilitate the evaluation. Conversely, physicians dealing with patients who are ill or fear they may be ill from environmental exposures incurred outside the workplace are unlikely to have data equivalent to the surveillance data to which occupational physicians may have access. Although testing of biologic specimens from the patient may still be informative, the perspective provided by prior baseline testing is almost

Environmental Medicine 11 always absent, as are data from previous examinations directed toward documenting information relevant to the particular environmental exposure(s) or health effect(s) of concern. The practitioner of occupational medicine also may have the benefit of information from periodic environmental monitoring (e.g., air sampling) of substances that may threaten health, typically performed by industrial hygienists or other trained personnel. Such measurements may greatly aid in interpreting patients’ complaints or the validity of their health concerns. Even if no measurements were taken before the fact, many industrial processes are performed under conditions that are sufficiently well characterized that they may be reproduced with reasonable precision at a time when the monitoring can be conducted. In contrast, it is rare for environmental measurements relevant to a specific individual to have been taken outside the workplace. Moreover, the patient with a non-occupational, environmental exposure is relatively unlikely to have maintained records on such items as the quantity of a potentially problematic substance used. As an individual, he or she is also much more likely to have varied his or her use of these substances over time in ways that are unlikely to be documented and about which the patient’s own memory may be poor. These factors substantially limit the possibility of replicating the exact circumstances of exposure in order to make environmental measurements that may be ideal for clinical evaluation (see Chapter 3).

Scientific investigation of new problems The complete spectrum of environmentally caused illnesses and their specific environmental antecedents has not yet been fully described. Thus, when one or more patients appear with symptoms that are either novel or unexpected for the set of environmental exposures experienced, it is reasonable to consider the possibility that the illness actually is caused by one or another of the exposures and that the particular disease–exposure link simply has not been described previously. This possibility may be easier to investigate in the workplace than outside it; reasons include specific awareness of the possibility of occupational disease in many populations of workers and the frequent occurrence in the workplace of qualitatively and quantitatively similar exposures among multiple, socioeconomically comparable but unrelated individuals, substantially increasing the statistical power of epidemiologic studies aimed at elucidating the cause of a new syndrome. On the other hand, there may be situations in which investigations of both new and previously described environmental health effects are problematic in the workplace and more easily completed in the community at large. Such situations may arise, for example, when employees feel substantial pressure either not to report or to downplay their report of symptoms that could reflect negatively on the employer. Alternatively, there may be peer pressure or pressure from employee advocacy groups (e.g., unions)

to report symptoms that the employee might otherwise ignore. Either circumstance can complicate the conduct of an epidemiologic study.

Host populations The special sensitivities of children and the elderly (i.e., those whose ages are outside the limits of those who typically make up the workforce) to certain environmental agents are more relevant to non-occupational than to occupational medicine (see Chapter 11). Thus, to a greater degree than the specialist in occupational medicine, the non-occupational environmental physician must take such potential differences in sensitivity into account. Children appear more sensitive than adults to a wide variety of environmental exposures. For example, the developing nervous system of a child appears to be substantially more vulnerable to the toxic effects of lead than that of an adult. Babies tolerate excess nitrate in drinking water far less well than adults. Other potent neurotoxins, such as pesticides (organophosphates and organochlorines) and saxitoxins (the agents of paralytic shellfish poisoning), produce severe illness or death at lower doses in children than in adults. In addition, models for assessing cancer risk generally predict a quantitatively greater risk to children from a given level of exposure to a carcinogen than for adults. Elderly persons also exhibit different responses to some environmental exposures than do younger adults. The elderly are far more likely to develop clinically apparent illness when exposed to extremely hot or cold conditions than are younger adults. Excretory functions involving renal and hepatic function may be diminished in the elderly, making them more likely to have adverse reactions to certain chemical exposures. In addition, chronic diseases and use of medicines for their treatment are more prevalent among the elderly than among other populations groups and may increase susceptibility to certain environmental exposures. Ambient air pollution, for example, results in excess respiratory and cardiovascular mortality, primarily among such persons. On the other hand, the importance of some environmental exposures may be diminished among the elderly. For example, assessment of exposures to certain carcinogens and counseling in this regard may not be as relevant to elderly persons, particularly if life expectancy is less than the anticipated latent period of the carcinogen. In addition, the adverse effects on reproduction of some environmental agents may be relatively unimportant to patients no longer in their reproductive years. Thus, to an unusually large extent, physicians involved in non-occupational, environmental medicine must take basic differences in susceptibility into account in the assessment, counseling, and treatment of specific patients, and individual differences in susceptibility demand heavier emphasis in non-occupational, environmental medicine research. A corollary of this statement is that recommended limits for occupational exposure to chemical and physical environmental agents can by no means be automatically

12 Introduction to Occupational and Environmental Medicine construed as appropriate or applicable to situations of exposure outside the workplace.

Social aspects (see also Chapter 9) The administrative, regulatory, and economic contexts in which occupational medicine and non-occupational environmental medicine are practiced also create important differences between the two areas. Ethical issues similar to those related to potential conflicts of interest of occupational physicians employed by the same companies whose workers they care for are far less likely to arise among physicians dealing with environmental medicine outside the workplace. Moreover, non-occupational environmental specialists may escape much of the burden of the complex, time consuming, and red tape laden workers’ compensation system. Although they do not necessarily have to confront certain ethical dilemmas faced by occupational physicians, physicians involved in non-occupational environmental medicine – particularly consultants involved in epidemiologic evaluations – share with their colleagues in occupational medicine another set of complex social issues. Environmental physicians are frequently called on by the media – in public meetings, courts of law, or other public forums – to offer expert opinions regarding whether or not specific cases of disease or injury were caused by a specific environmental exposure. Often such opinions must be offered on the basis of few objective data or prior scientific studies and in the context of heated feelings and firmly established, preconceived notions on the part of those affected and their supporters. The expert may be confronted by equally daunting pressure from the party or parties responsible for the environmental exposure, who risk substantial economic losses from an expert opinion supporting a disease exposure link.

Established versus hypothetical hazards Environmental medicine deals largely with questions of exposure and risk. In this regard it is useful to differentiate clear-cut hazards – those known to cause effects at occupational dose levels, from potential hazards – those known to cause human effects only at much higher dose levels, or those never demonstrated directly to cause harm in humans. Clear-cut environmental hazards can be designated as such on the basis of epidemiologic studies that show a cause-and-effect relationship between exposure to the hazard and the development of a particular illness or injury (usually, but not invariably, done in the occupational setting). For patients exposed to such environmental hazards, the physician’s course of action is relatively clear. The exposed patient should be counseled regarding the nature of the risk, along with the likelihood – based on estimated dose and host factors – of possible health outcomes. When appropriate, the patient should be evaluated for the presence of the disease for which he or she may be at increased risk. If the disease is present, appropriate treatment should be instituted.

More problematically, much of the public’s current concerns regarding the environment relate to exposures for which harmful effects in human populations have not yet been demonstrated. In general, such potential environmental hazards have been identified as possibly dangerous on the basis of their chemical or physical similarities to clear-cut hazards and/or their apparently harmful effects on animals or in in-vitro systems. Counseling patients about risk in such circumstances is substantially more complicated than for the former class, because the issue of whether or not the risk actually exists at all has to be considered. Quantitative estimates of the extent of possible risk may be lacking entirely. When such quantitative data exist at all, they are typically derived from animal studies or extrapolated from other data on similar (but not identical) compounds. Under these circumstances, some of the important caveats regarding the validity of risk assessment techniques (see Chapter 60) must be incorporated into the patient’s counseling on risk related to the environmental exposure.

Data from community-based studies Nowhere can the occupational roots of much of nonoccupational environmental medicine be seen as clearly as in a review of currently recognized environmental hazards. Most such hazards have received their strongest and first scientific support for a causal relationship with human illness on the basis of studies performed on exposed workers. Well-known examples of exposures whose health effects were elucidated in this way include the linking of asbestos to asbestosis, mesothelioma, and carcinoma of the lung; vinyl chloride to angiosarcoma of the liver; and mercury exposure to adverse effects on the nervous system. It is primarily on the basis of these occupational epidemiologic data that practitioners of nonoccupational, environmental medicine are aware of the potential risks such exposures pose to persons having nonworkplace exposure. Occupational studies are frequently more useful sources of data than studies of the same agents outside the workplace, because exposure in the workplace is often more intense, more prolonged, more regular, and more easily quantifiable than are comparable exposures outside the workplace. Thus, in a study population of a given size, the numbers of cases of the health outcome of interest are likely to be substantially greater in the occupational than in the non-occupational situation. Moreover, because exposure is generally more easily quantified in occupational studies, the extent of risk is more easily quantified, and the existence and shape of the dose–response curve can be estimated with greater precision. Nevertheless, data from studies of environmental illness in non-occupational settings are important and continue to be needed to supply information that is missing or unobtainable from occupational studies alone. For example, occupational studies do little to address questions regarding the existence and extent of effects of specific environmental agents on children and others outside the

Environmental Medicine 13 range of demographic parameters in which workers usually fall. Moreover, what might happen in the home cannot easily be extrapolated from results of studies conducted under industrial conditions. For example, children’s exposure to mercury (and consequent health effects) related to ‘off-gassing’ from household interior latex paint could not easily have been studied in the workplace. Diseases involving severe hypersensitivity to substances arising from industrial operations can sometimes be studied more satisfactorily outside than inside the workplace, because either the workers may be relatively few in comparison to the numbers of persons in nearby areas who are exposed to the substances emanating from the workplace or workers may systematically eliminate themselves from jobs involving contact with substances to which they are hypersensitive and that, therefore, cause them severe discomfort. Thus, non-occupational rather than occupational studies led to the discovery that severe (sometimes fatal) asthma attacks can be elicited in sensitized persons by soybean dust released during the course of industrial operations involving the bean.20

Risk assessment and environmental medicine (see also Chapter 59) Here, the term risk assessment refers to the use of available data to develop a quantitative estimate of the probability that a given health outcome may result from a specific exposure experienced at a particular point in one’s life, for a particular length of time, by a particular route, and at a particular dose level. Although risk assessments may be based on epidemiologic studies (i.e., those involving observation of human beings), they more frequently involve extrapolations based on data obtained from toxicity testing in animals. Most often, the health outcome of concern is the development of malignant tumors. The test animals are almost always given far higher doses of the test compound than the dose levels for which one wishes to estimate risk, and the test generally lasts a period of about 2 years, the approximate lifespan of small rodents most often used as test animals. Malignant tumors or other health outcomes are assessed histopathologically or by other appropriate techniques. The data obtained from these studies are incorporated into mathematical models that attempt to estimate the maximum dose or concentration of the test chemical to which a human being could be exposed over the course of an approximately 70-year lifetime, while incurring an excess risk – i.e., incremental risk over background – of the health outcome of concern that is less than a particular fraction, frequently 1 per 1,000,000 or 1 per 100,000. Despite widespread use in regulatory and public health decision making, full consensus has not been reached regarding many issues directly related to the validity of various risk assessment approaches. Some of the most problematic issues relate to: ■ extrapolating findings from one species to another; ■ adjusting incidence figures from relatively short-lived rodents to relatively long-lived human beings;

■ calculating equivalent dosages across species; ■ dealing with contradictory results obtained from studies

of the effects of the same exposure on different animal species; and ■ determining the proper mathematical function for extrapolating high-dose study exposures to the levels of exposure obtained from the environment, which are often several orders of magnitude lower. Risk assessments based on human (epidemiologic) data are not limited by the problems of extrapolating data from animal models but do suffer from the other problems. Chief among these is quantification of dose. Many human studies of potential use in risk assessment involve retrospective examination of data collected over many years, often from times when the exposure of concern was not thought to be particularly harmful (which is logical, if one considers that large numbers of human beings would have been unlikely to have exposed themselves unnecessarily to a known substantial danger). The attention paid to documenting dose was, therefore, often correspondingly small, and the exposure data are, therefore, frequently imprecise. Human studies yielding positive results also may involve relatively high levels of exposure and, therefore, may suffer from the same problem as animal studies with regard to choice of an appropriate mathematical model for extrapolating data to dose levels more typical of home or community exposure. All of the problems mentioned earlier lead to uncertainty in the risk estimate. Yet, the results of risk assessments are frequently presented as one or more specific numbers, with no mention of the underlying qualitative and quantitative uncertainties that must inevitably affect the interpretation of such estimates. However they are derived and with whatever problems, the results of risk assessment are widely applied in formulating environmental regulations regarding the permissible levels of contaminating substances in air, water, soil, and food. The results of risk assessments also guide decisions on risk management made by environmental public health officials (e.g., in determining the extent to which spilled or unintentionally released chemicals must be removed or ‘cleaned up’ before usual activities in an area can be resumed). Although the techniques of risk assessment play an important role in the approach to environmental regulation and environmental public health, their appropriate use in physicians’ contacts with individual patients remains unclear. It may, in fact, be useful for a patient to be told that a particular exposure to which he or she may be subjected, when incurred constantly over the course of a lifetime, is estimated to lead to approximately a 1 in 10,000 lifetime additional risk of cancer. However, any counseling of the patient in this regard should be accompanied by full disclosure of the types of uncertainties mentioned earlier, all of which potentially affect the validity of the risk estimate. Moreover, to the greatest extent possible, any quantitative risk estimate used in counseling a patient should be set in the context of risks the patient faces frequently and, therefore, is likely to understand. For

14 Introduction to Occupational and Environmental Medicine example, it may be useful to mention to the patient that 1 in 100 is the order of magnitude of risk for dying over a lifetime in a plane or car crash; by comparison, 1 in 1,000,000 approximates the lifetime risk of being killed by lightning. Recommending behaviors for risk reduction is complicated by a number of factors, not the least of which is the varying tolerance of particular individuals for different levels of risk. In addition, the risk assessment information commonly available is based on assumptions of constant exposure over a lifetime. Comparable figures for acute exposures that may be of concern to individual patients may be lacking. Other specific features of individual patients’ situations (e.g., very young or very advanced age) may result in substantially greater or less theoretical risk for a given patient than that indicated by the estimate used by public health personnel for protecting the general population’s health. Currently, the health practitioner does not have easy access to individualized risk estimates fitting the individual circumstances of particular patients as are now readily available for other clinical risks such as cholesterol or body mass index. The best one can do is to emphasize the caveats associated with any quantitative estimate of risk given to a particular patient.

entirely or might have been recognized by them only much later on. In addition, because environmental medicine is a developing area in which previously unrecognized disease exposure associations may be anticipated from time to time, alert clinicians can play an important role in identifying new public health problems. In the early 1980s, clinicians in Barcelona, Spain, pointed out the sudden appearance of clusters of cases of exacerbation of asthma that, from time to time, flooded the city’s emergency rooms. That report provided the stimulus for an investigation that eventually led to the discovery that dust arising from ships unloading soybeans in the nearby port was the instigating agent.20 This discovery, in turn, allowed appropriate preventive measures to be taken. Similarly, the eosinophilia-myalgia syndrome was first reported to public health authorities by a group of clinicians who recognized a cluster of patients in New Mexico with similar, peculiar illnesses and a common history of ingestion of L-tryptophan. Subsequent investigation led to the discovery of over 1500 cases nationwide, a clear-cut association with L-tryptophan from a particular manufacturer, recall of implicated products, and an end to the epidemic.21 Fostering recognition of such outbreaks is one of the greatest potential roles of practitioners in this emerging field.

Clinical and public health practitioners and investigators

References

Close partnership between clinicians and public health practitioners is particularly important in the field of environmental medicine. The need for such collaboration arises, in part, from the fact that although a clinician may effectively treat the deleterious effects on health produced by an environmental exposure, he or she is rarely in a position to change the environmental factors that produced the illness, by contrast with occupational medicine where that relationship may be very much more tenable. For example, it would be senseless to provide chelation therapy for a child with lead toxicity and effectively reduce his or her lead level but then return the child to the same environment that caused the illness in the first place. In this situation, contact with the responsible public health authorities can lead to action that will be of immediate benefit to the patient. Public health officials also may be a good source of data for health practitioners attempting to counsel patients in a reasonable way about recent highly publicized environmental health concerns. Such episodes of particular concern can be national (e.g., the escalating concern over radon accumulation in buildings and the episode of concern in 1989 regarding residues of daminozide [Alar] and its breakdown products in apples) or local (e.g., concerns of persons living near hazardous waste sites). A close association with clinicians can provide important information to public health professionals as well. There is currently little formal surveillance for environmentally caused disease. Informal reports from clinicians can alert those working in public health to the presence of a problem that otherwise might have escaped notice

1. The BLS website is: data.bls.gov. 2. Bureau of Labor Statistics. US Department of Labor News, December 18, 2001. USDOL-47. 2001. 3. Goldsmith DF. Uses of workers’ compensation data in epidemiologic research. Occup Med State Art Revs 1998; 13:389–415. 4. Morse T, Dillon C, Warren N, Levenstein C, Warren A. The economic and social consequences of work-related musculoskeletal disorders: the Connecticut Upper Extremity Project (CUSP). Int J Eviron Health 1998; 4:209–16. 5. The website for these data is: www.cdc.gov/nchs/products/ catalogues/subject/mortmed.html. 6. Herbert R, Landrigan PJ. Work-related death: a continuing epidemic. Am J Pub Health 2000; 90:541–5. 7. Division of Respiratory Disease Studies, NIOSH. Work-related Lung Disease Surveillance Report 1999. DHHS (NIOSH) Number 2000-105. Publications Dissemination, Cincinnati: NIOSH, 2000. 8. Krant A. Estimates of the extent of morbidity and mortality due to occupational diseases in Canada. Am J Ind Med 1994; 25:267–78. 9. The website for these data is: www.cdc.gov/nchc/about/major/hdasd/nhdsdes.html. 10. Rutstein DD, Mullan RJ. Sentinel health events (occupational): a basis for physician recognition and public health surveillance. Am J Pub Health 1983;73:1054-62 11. Rosenstock L, Daniell W, Barnhart S. The 10-year experience of an academically affiliated occupational and environmental medicine clinic. West J Med 1992; 157:425–9. 12. Data provided by Kathy Hunting, Association of Occupational and Environmental Clinics, June 2002. 13. NIOSH. Worker Health Chartbook 2000. DHHS (NIOSH) Number 2000-127. Publications Dissemination. Cincinnati: NIOSH, 2000. 14. Whittaker SG, Curwick CC. Surveillance for occupational lead poisoning, State of Washington 1993–2001: incorporating data from May 15 1993 through June 30, 2001. Technical Report 44-3-2001. Safety and Health Assessment and Research

Environmental Medicine 15 for Prevention. Olympia, WA: Washington State Department of Labor and Industries 2001. 15. Tumpowski C, Rabin R, Davis L. Lead at Work. Elevated blood lead levels in Massachusetts workers. Occupational Health Surveillance Program. Massachusetts: Dept of Public Health, 1998. 16. Morse T. Occupational Disease in Connecticut 2001. Occupational Disease Surveillance Program. Connecticut: Department of Labor and Department of Health, 2001. 17. Lalich NR, Sestito JP. Occupational health surveillance contributions from the National Health Interview Survey. Am J Ind Med 1997; 31:1–3.

18. Doll R, Peto R. The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. J Nat Cancer Inst 1981; 66:1191–308. 19. Darby S, Hill D, Doll R. Radon: a likely carcinogen at all exposures. Ann Oncol 2001; 12:1341–51. 20. Anto JM, Sunyer J, Rodriguez-Roisin R, Suarez-Cervera M, Vazquez L. Community outbreaks of asthma associated with inhalation of soybean dust. New Engl J Med 1989; 320:1097–102. 21. Varga J, Uitto J, Jimenez SA. The cause and pathogenesis of the eosinophilia-myalgia syndrome. Ann Intern Med 1992; 116:140–7.

Chapter 2 Occupational and Environmental Medicine: the Historical Perspective Paul Blanc The history of occupational and environmental medicine is composed of multiple threads continually interwoven over time into a single historical cord. One thread is formed by the key clinicians and researchers who have contributed to the development of this discipline. Most brief summaries of occupational and environmental medical history take this as their limited focus. Yet as important as these biographical elements are in the story of occupational and environmental medicine, other factors have come into play that make the evolution of this discipline distinct from that of other branches of medicine. First, advances in technology have played a driving role in occupational and environmental medicine that is unparalleled in other fields of health. It is true that advances in diagnostic and therapeutic modalities, from the microscope to the laser, demonstrate the powerful impact that technologic innovation can have on medical practice as a whole. But despite the role that such inventions have played in clinical care, the underlying pathologic processes of concern to practitioners have not changed because of them. Simply put, the microscope did not create illnesses due to new strains of bacteria. In contrast, technologic change continually introduces new occupational and environmental hazards, leading to evolving patterns of established diseases as well as inducing entirely novel conditions never experienced before in human history. Radiation poisoning is an obvious example, but by no means an isolated one. Second, the history of occupational and environment medicine reflects the impact of larger social movements outside the narrow confines of medicine. This is not to argue that other branches of medicine are immune to such forces. The course of modern medicine still reflects the impact of the French revolution, transmitted down through the influential work of French medical scientists working at the end of the 18th and in the first half of the 19th centuries.1 So too, the Flexner Report and all that it brought with it for American medicine had a larger sociopolitical context.2 Nevertheless, occupational and environmental medicine, more than any other health discipline in the last 200 years, has tended to wax and wane as a consequence of societal forces. The hygienic movement of the 19th century (particularly in Great Britain), which was linked in turn to wider social reforms, is a case in point. The First World War marked another social–political confluence of forces distinctly impacting occupational and environmen-

tal medicine. More recently, the social movements of the late 1960s in the US coincided with the establishment of OSHA, NIOSH, and the EPA, all of which profoundly affected the field. The goal of this chapter is to place the story of occupational and environmental medicine within the context of these three disparate, yet inter-related themes: the key historical figures that contributed to the development of the field, the technologic changes that have led to an evershifting burden of disease, and the sociopolitical forces that helped set the priorities of the discipline. An overview of this chronology is summarized in Table 2.1. Finally, key resources in occupational and environmental medical history will be summarized in order to direct interested readers further beyond the scope of this brief overview.

OCCUPATIONAL AND ENVIRONMENTAL MEDICINE IN THE ANCIENT WORLD The remote history of occupational health is fragmentary and largely conjectural.3–5 In terms of occupationally related illness and injury, as the medical historian Henry Sigerist many years ago pointed out eloquently, the most salient determinant of work-associated morbidity in the ancient world was the widespread use of slave labor.5 In a 1936 address to the New York Academy of Medicine, Dr Sigerist remarked: ‘Labor in ancient civilization was primarily slave labor. The pyramids were built by state slaves whose lives had no value whatever, whom every war would replace. We still can see the Egyptian workers laboring under the whip as represented on wall paintings and in reliefs. The lot of the city worker was hardly any better and we can still perceive their voice of rebellion . . . We admire the graceful Greek bronze statuettes that fill our museums but we do not think of the copper miners providing material for these works of art, or the coal miners digging for coal to make the bronze, working ten hours in narrow galleries suffocated by heat and smoke. They were prisoners of war or convicts as a rule.’5 As noted by Sigerist, the adverse occupational safety and health impacts of slave labor in the ancient world were most dramatic in mining operations. This industry was also one of the first to manifest the adverse impact on occupational safety and health of technologic advancements in

18 Occupational and Environmental Medicine: the Historical Perspective Historical period

Notable factors

Key documentation

Ancient–Classical

Slave labor Mining and metallurgy Agriculture and crafts

Scattered references in classical medical and non-medical writing

7th–17th Centuries

Early technological change Armaments and metallurgy Alchemical experiments

Ellenbog, Agricola, Paracelsus Chinese technological texts

1650–1800

Large-scale trades Renaissance technologies Coal fuel power pollution

Ramazzini and his translators Evelyn on air pollution

1800–1900

Industrial Revolution Novel exposures to toxins Germ theory predominates

Thackrah and Patissier Annales d’Hygiene Medicine Legale Hirt, Simon, Arlidge, Proust

1900–1950

The New Public Health First and Second World Wars Progressive movements Air pollution emergencies

Oliver, Hamilton, Legge Teleky, Underhill, Drinker Various governmental reports

1950–1970

Nuclear threats, Cold War Petrochemical industry OSHA, NIOSH, EPA

Carson, Brodeur, Berman, Sellars

Table 2.1 A brief chronology of occupational and environmental health

metallurgy6 (which increased the demand for a variety of ores) and mining methods (especially pumping systems that allowed mining to delve deeper than ever before).7 Such mining operations were important economic forces in Egypt and in Greece, but took place on a truly massive scale in the Roman Empire, particularly in the Iberian peninsula. Rosen’s superb History of Miner’s Diseases provides the most detailed history of this subject.7 Allusions by non-medical writers of the classical period to various labor hazards, including risks to miners, constitute some of the earliest written documentation of occupational safety and health problems. Many of these comments are terse references, such as those that have been noted in the epigrams of Martial or the satires of Juvenal.3,4 One of the most notable references is the widely cited description of mining by Pliny the Elder (Natural History XXXIII, 40) in which he writes: ‘Persons employed in the manufactories in preparing minium (red lead) protect the face with masks of loose bladder-skin, in order to avoid inhaling the dust, which is largely pernicious.’5 There are other occasional references to working conditions of artisans or laborers in ancient Western writings, including the Egyptian papyri Eber and Sallier.4,5 Similar examples of fragmentary descriptions of working conditions and their inherent risks may also exist in the nonWestern early written tradition (especially Chinese medical and technologic writing), but this literature, insofar as it might touch on occupational morbidity, has not been systematically reviewed in the English language. The concept of ‘environmental’ health, in the modern sense of the term, is even more difficult to track in early history. A recent review of ancient Roman law makes it clear that statutory and other legal remedies for water and air pollution were indeed undertaken as early as the 1st century AD (for example, a legal opinion by the Roman

Aristo that a cheese maker should not emit smoke into the building above it).8 Talmudic commentary provides further indication that legal concepts of environmental and even occupational protection have a surprisingly long history.9 Classical medical writing (as opposed to general works such as histories or satires) largely ignored occupation in relation to health. Certain vocations were addressed, for example, horseback riding (Hippocrates associated it with impotence and sterility; Aristotle with augmented sex drive).3 This trend continued through Galen, although he may have been the first physician who recorded a personal brush with a significant occupational health hazard. According to one history of the subject: ‘In the course of one of his numerous voyages, Galen spent some time on the Island of Cyprus. While there he had visited a mine where copper sulfate was recovered. Unaware of the danger, he himself was nearly overcome by the fumes in the mine. He records that the workmen who carried out a vitriolic liquid ran from the mine with all speed with each load to avoid perishing in the midst of the labors.’3

THE 7TH CENTURY THROUGH THE MID-17TH CENTURY In the thousand years following the end of the Classical period, the foundations were set upon which modern Western medical science would later be built. Occupational health, as a medical concern, was marginal (at best) throughout most of this long period. Environmental health was given fairly short shrift too, being little discussed beyond a few generalities about climate and good air. This can be contrasted, by comparison, to the subject of diet and health, a topic addressed by many medical writers over these centuries.

1650 to 1800 19 During this time there were few, if any, social forces that would have worked to bring such concerns more to the forefront. There were, however, technologic innovations that began to have an impact through changing patterns of morbidity and mortality. These innovations, at first sporadic and isolated, became increasing important as the Middle Ages drew to a close. One of the most straightforward examples is provided by technologic developments in armaments, especially the introduction of firearms. The traumatic risks of soldiering, essentially an occupational hazard, were well documented by the famous 16th century military surgeon Ambroise Pare, who gave special emphasis to the novel (at that time) phenomenon of gunshot wounds.10 From the same period, the introduction of new sea-faring technologies (not only the sailing ships themselves, but as importantly, innovative navigational equipment) led to severe outbreaks of scurvy as an occupational disease of seamen.11 Even earlier than this, mining and metal working technologies became the engine driving the evolution of occupational health as a distinct medical issue.7 As was noted previously, this was already true, although to a much lesser extent, in the Classical period. Lead was certainly very widely used during the Roman period and has even been the subject of speculation as to its potential role as a major environmental toxin in that time period.12 One of the first, clear-cut medical descriptions of lead colic was noted by the 7th century Byzantine physician Paul of Aegina.13 In the Middle Ages, lead, arsenic, and mercury were all employed in the Western pharmacopeia and were the subject of study and misadventure by generations of alchemists. But above and beyond that, the appreciation of novel metals and alloys, and new ways of refining others, brought about the introduction of greater risks and even entirely new hazards to miners and metal workers.7 The first publication devoted entirely to the subject of occupational health was Ullrich Ellenbog’s fittingly titled, seven-page pamphlet, On the Poisonous Wicked Fumes and Smokes of Metals (Von den gifftigen besen Tempffen und Reuchen der Metal).14 First printed in 1524, it was originally written in 1473 and warns the metal worker of the health hazards of metal, coal, and acid fumes. Between 1500 and 1650, an increasing number of writers began to address the subject of miners’ and metal workers’ health. The first book-length treatment of the subject was that of Paracelsus (Von der Bergsucht under anderen Bergkrankheiten), written in 1533 but only first published posthumously in 1567.15 Agricola’s De Re Metallica (1556), although mainly devoted to the technology of mining and smelting, contains a notable section on occupational health in Book VI. Agricola writes: ‘It remains for me to speak of the ailments and accidents of miners, and of the methods by which we can guard against these, for we should always devote more care to maintaining our health, that we may freely perform our bodily functions, than to making profits. Of the illnesses, some affect the joints, others attack the lungs, some the eyes, and finally some are fatal to men.’16

In addition to Paracelsus and Agricola, during this period authors of note on occupational exposures include Pansa, Ursinus, and Stockhausen.7 Early technologic writing, for example, that of Ercker17 or Birunguccio,18 is also of interest in relation to likely occupational exposures. Some of the medical writing is straightforward, describing hazards, such as asphyxia, that are clear cut and readily discernible in modern occupational terms. Other writing is terribly obscure (especially Paracelsus).15 Even Agricola, amidst his lucid descriptions of hazards such as stagnant air, falls from ladders, and cave-ins, warns that in some mines (‘although in very few’), there are demons which can be put to flight by prayers and fasting.16 During this time, another technologic change in Europe that began to impact health was manifest in the environmental rather that the occupational arena. This was the introduction of fossil coal (as opposed to wood, peat, or charcoal) as a fuel.19 At first, this practice was limited to Britain, with a particularly important industrial application in firing lime kilns. As early as the 13th century, a commission had been established in London to investigate the problem of pollution from coal burning. In 1306, a proclamation was issued banning the practice, albeit to little effect.20 After a brief plateau, the coal market (and with it coal mining) increased logarithmically during the 16th and 17th centuries in Britain.22 Although fossil coal was employed in China for fuel long before its use in Europe, there has been no systematic review of the history of its use in China and its potential public health impact from an historical perspective. This parallels a general paucity of information on health impacts of early technologic innovations that originated in Asia, although the history of this technology as it pertains to chemistry and metal working is quite rich.21–24

1650 TO 1800 In 1700, after some years of preparatory work on the subject, Bernadino Ramazzini published the first edition of his landmark treatise, De Morbis Artificum (‘The Diseases of Artificers’, usually translated in modern terms as ‘The Diseases of Workers’).25 This work is remarkable for its innovations. First, in simply approaching the subject matter as a topic worthy of a complete treatise, rather than by narrowly focusing on a single occupational group or exposure, Ramazzini fundamentally changed the way in which all medical writing would subsequently address occupational illness and injury. Second, and no less importantly, Ramazzini explicitly put occupational risk factors on the agenda in terms of the medical differential diagnosis. In the preface to De Morbis Artificum, Ramazzini directly addresses the reader with the following, often cited (deservedly so) admonition: ‘The divine Hippocrates informs us, that when a physician visits a patient, he ought to inquire into many things, by putting questions to the patients and bystanders . . . to which I would presume to add one interrogation more; namely, what trade he is of. But I find it very seldom minded in the common course of practice, or if the physician knows

20 Occupational and Environmental Medicine: the Historical Perspective it without asking he takes little notice of it. Though at the same time a just regard to that, would be of great service in facilitating a cure.’26 Ramazzini’s text was divided into 42 chapters, although the number of occupations covered was even greater (for example, a single chapter covers oil pressers, tanners, lutestring makers, candle-makers and cheese-makers). Even this was not sufficiently comprehensive for the author. In 1713, he published a second edition of the work, adding to it a supplement of an additional 12 chapters to address occupations omitted from the original (including printers, carpenters, and sailors, among others).27 The work practices that Ramazzini documents provide a fairly exhaustive summary of manufacturing technology as it existed in Europe at the turn of the 17th to 18th centuries, just before the advent of the Industrial Revolution that would, by century’s end, begin to fundamentally transform the workplace and the risks it entailed. Indeed, there is very little ‘recent’ technologic innovation driving Ramazzini’s observations. Most of the occupations he describes, even those that entail metallurgy and mining, employed materials and methods well established for at least several generations. Other work practices described are virtually unchanged since Roman times, for example, the hazards of fullers (cleaners of cloth). Ironically, one of the few novel syndromes related to an emerging industry was that of nicotine toxicity among tobacco workers, especially for those involved in tobacco grinding.27 Although rapidly changing technology leading to novel illnesses is not a major factor in De Morbis Artificum, the work clearly reflects the impact on Ramazzini of the larger social, cultural, and political forces of his day. In broad historical terms, he was very much a product of the late Italian Renaissance. More specifically, he spent most of his career in Modena during the years leading up to the publication of the De Morbis Artificum under the patronage of the Ducal House of Este, one of the great liberal benefactors of the age.27 This spirit imbues Ramazzini’s work, which emphasizes the lot of the common worker with empathy and compassion on almost every page. Perhaps the most striking passage demonstrating Ramazzini’s sociopolitical context is his chapter on the ‘Diseases of Jews,’ in which he ascribes these disease to the occupations into which the Jews were forced by the restrictions of the period. As odd as the chapter title strikes the modern reader, its actual content underscores Ramazzini’s decidedly liberal approach to his subject. It would be hard to overestimate the impact that Ramazzini’s work had on the discipline of occupational medicine throughout the 18th century and well into the 19th. The work was translated from Latin into English within five years of its initial publication (1705); a new English edition in 1746 included the 1713 appendix.26,27 More important were two other 18th century annotated translations with significant additions by the translators, one in French (1777)28 and one in German (1780–83).29 The latter is particularly noteworthy, more than doubling Ramazzini’s original text. Beyond these translations, Ramazzini was widely cited by his contemporaries and

the generations that followed. His most famous student was the great pathologist Morgagni, whose classic text, The Seats and Causes of Diseases, is not only meticulous in documenting the occupations of many of its cases, but also directly discusses possible work-related contributions to the pathology that Morgagni observed in several instances.30 The key medical writers in the 18th century who addressed occupational subjects (although none with as all-encompassing a text) acknowledged their predecessor. One example is Tissot, whose text on the diseases of ‘men of letters’ was first published in 1768 and widely circulated.31 Ramazzini’s work on occupational diseases was even the subject of several 18th century doctoral theses,32–34 the most well known of which was the work of a Swede, Nicholas Skragge.35 Skragge wrote his thesis under the tutelage of Linneaus, whose lecture notes document that he gave considerable attention to Ramazzini.36 Ramazzini’s presence within 18th century occupational and environmental medicine is so dominant that it obscures another pivotal contribution of this time, John Evelyn’s Fumifugium; or, The inconvenience of the aer and smoak of London dissipated.37 Initially published in 1661, it is the first tract (26 pages only) specifically focused on air pollution and its potential dangers. Although Evelyn’s dates (1620–1706) overlap almost entirely with those of Ramazzini (1633–1714), the two could not be further apart. Evelyn, neither a medical practitioner nor a scientist, was a Royalist, a great diarist of the age (as was his contemporary, Pepys), and a colorful polemicist. In Fumifugium, railing against the coal smoke polluters of his day, he writes: ‘Whilst these are belching forth their sooty jaws, the City of London resembles the face of Mount Aetna, the Court of Vulcan, Stromboli, or the Suburbs of Hell than an Assembly of Rationale Creatures . . .’37 Although it might be tempting to dismiss Evelyn as something of an Enlightenment crack-pot, Fumifugium is important in the history of air pollution specifically and environmental health generally. In this context, the unsigned, contemporary preface to a 1772 re-printing of the book is particularly relevant: ‘Our Author [Evelyn] expresses himself with proper warmth and indignation against the absurd policy of allowing brewers, dyers, soap-boilers, and lime-burners to intermix their noisome works against the dwelling-houses in the city and suburbs; but since his time we have a great increase of glass-houses, foundries, and sugar-bakers, to add to the black catalogue; at the head of which must be placed the fire-engines of the water-works at London Bridge and York Buildings, which (while they are working) leave the astonished spectator at a loss to determine whether they do not tend to poison and destroy some of the inhabitants by their smoke and stanch than supply with their water.’38 At the beginning of the 18th century, Evelyn stood nearly alone in addressing air pollution in any modern sense of the word.39 By century’s end, a series of discoveries had led to an understanding of the precise chemical composition of air. This, in turn, provided a new scientific basis to consider how the contents of air, on a chemical

The 19th Century 21 basis, might act upon the body. The same chemical-analytic attention had not yet been given to water pollution, but was beginning to be applied to food contamination. For example, George Baker’s classic ecologic analysis of the regional association between the practice of using lead in apple presses and the ‘Devonshire colic’ was combined with laboratory experiments demonstrating the lead content in various cider preparations.6 Throughout the second half of the 18th century, the potential for systematic application of scientific discoveries to industrial manufacturing was gaining increasing attention. The Diderot’s Encylopedia, subtitled an ‘Analytical Dictionary of the Science, Arts, and Trades’, published its first volume in 1751.41 The lavish illustrations of the Encyclopedia, detailing the mechanical state of the art in industry at the time, heralded the technologic transformations of manufacturing about to occur.

THE 19TH CENTURY By 1800, 100 years after Ramazzini, the occupational and environmental health equation had been fundamentally altered. First, the Industrial Revolution was in full swing. Second, the French Revolution redefined the rights of the people, and, by extension, those of the worker. Third, medical research was beginning to link toxicology to pathophysiology in ways directly relevant to occupational and environmental as a scientific discipline. Even after the first flush of the Industrial Revolution, technical innovation (chemical and mechanical) continued to remain a dominant force in occupational health and safety throughout the century. Far more than simply a series of manufacturing changes in the cotton textile industry,42 it was introducing both entirely novel exposures as well as increasing old exposures to new levels of intensity.43 Some key examples of novel exposures introduced in this period linked to outbreaks of entirely new syndromes include: chlorine gas inhalation,44 work in high barometric pressure environments,45 the introduction of white phosphorus (‘lucifer’) matches,46 carbon disulfidebased rubber vulcanization,47 aniline and other synthetics in textile dyeing,48 and, just at the end of the century, ionizing radiation in medical radiography.49 One of the most important established exposures that was greatly magnified through new industrial processes was that of silica dust, especially through the introduction steam power-driven grinding wheels.50,51 Lead is another prime example of a classic exposure with greatly increased exposure levels in the 19th century, particularly in the pottery, glass, and metal-working industries.52 Even zinc oxide exposure, which was trivial in 18th century brass manufacturing, burst on the scene with new methods of making the alloy, and with that new technology a novel occupational syndrome – metal fume fever.53 During the 19th century, social forces came to bear on occupational health in important new ways. Despite the revolutionary tradition on the Continent (or perhaps in part because of it), occupational health had its clearest political manifestations in Great Britain, where it was part

of a larger hygienic movement tied closely to a broad agenda of social reform. The key hygienic document of the first half of the 19th century is the famous Chadwick Report on the Sanitary Condition of the Labouring Population.54 It addresses specific occupational exposures only briefly, although they are included. To a greater extent, the governmental sanitary reports of John Simon underscore the extent to which occupational safety and health was very much on the public health agenda.55 Even Florence Nightingale’s Notes on Nursing, which she intentionally adapted for working men and women, addresses occupational health hazards in general terms.56 The most important English language occupational text of the 19th century was that of Charles Thackrah, The effects of the principal arts, trades, and professions, and of civic states and habits of living. First published in 1831, Thackrah greatly expanded the work in a second edition in 1832.57,58 He was planning a third edition when he died of tuberculosis (most likely occupationally acquired) in 1833.59 It is the first text to address the myriad occupational health risks brought about by the new technologies of the Industrial Revolution, especially the health of textile workers. This was a subject of particular interest to Thackrah because he practiced in Leeds, a textile factory center. Thackrah’s work had an immediate influence on other hygienists of the period, especially in Great Britain60 and in the US.61,62 It also had an impact on political reformers. Michael Sadler, rising to speak on pending social legislation in Parliament, is reported to have had a copy of Thackrah’s book in his hand and to have quoted from it extensively.59 Invoking occupational health as a motivation for more broadly protective legislation was not without precedent: one of the first pieces of child labor legislation in Britain (1778) was for chimney sweeps63 among whom scrotal cancer had been reported by Percival Pott in 1775.64 British legislative reforms of working conditions were introduced in a series of landmark statutes promulgated in the second half of the 19th century. As a part of these enabling acts, governmental medical positions as ‘factory inspectors’ were established. These positions, in turn, provided the training ground for several generations of physicians specializing in occupational disease.65,66 In France, occupational medicine diverged from the discipline that was evolving in Great Britain. Industrial conditions differed between the two countries in significant ways. Although there were some relatively large manufacturing sectors (for example, textiles), production to a much greater degree than in Britain remained in smallerscale workshops, especially in Paris.67 Medicine in France was distinct as well, characterized by a greater emphasis on systematic analysis and characterization of pathophysiologic processes.1 This was also manifest in occupational medicine. Patissier in France played a role similar to Thackrah’s in Britain. In 1822, Patissier wrote the first original French text on occupational medicine.68 Often cited as an annotated translation of Ramazzini, it goes far beyond that. Patissier broke down occupational exposures into

22 Occupational and Environmental Medicine: the Historical Perspective classes of risk, categorized hierarchically consistent with French regulations. Most of the later French writers on occupational health in the 19th century followed Patissier’s somewhat didactic schema. Another major text of the early 19th century, Tanquerel des Planches’ Lead Poisoning, demonstrates even more strikingly the emerging new ‘science’ of occupational medicine in France during this period.69 In more than 1000 pages, it meticulously details over 1200 cases of lead poisoning at the famous Charite Hospital. It remains to this day the largest case series of its kind. The emerging science of toxicology, dominated by researchers ranging from Orfila70 to Bernard,71 also had a major impact on occupational medicine in France. Indeed, the leading scientific journal of the 19th century for occupational medicine content was the toxicologically oriented Annales d’Hygiene et de Medicine Legale. Many of its key occupational articles were also reprinted as separate bound volumes, providing a major source of scholarship on the subject. In the latter part of the 19th century, German occupational medicine became ascendant, particularly through a series of scientific and medical publications. Of these, the massive textbook of Hirt is the most noteworthy example.72 At the same time, the first workers compensation insurance schemes were coming into being, starting with the German ‘Accident Insurance Law or Trades and Industry’ of 1884.73 Beyond Great Britain, France, and Germany, occupational medicine as a discipline had little presence in the 19th century. In the United States this was especially the case, although clinical reports and scholarly articles on the subject did appear sporadically.74 Moreover, there was a subtle, but general decline in the field of occupational health as the century progressed, even in its former centers of strength in Great Britain and France. There were notable textbooks produced in this period, such as those by Aldridge 75 and by Proust (Marcel’s father, who was one of the first to apply the term ‘byssinosis’ to illness in cotton textile workers),76 but these are something of the exception proving the rule. The explanation for this decline is not clear cut. Technologic innovation was still introducing new hazards and novel syndromes during this period (the association of aplastic anemia with benzene was first published in 1897, for example).77 In part, the political climate may have been less conducive. Clearly, social reform was no longer at the forefront. More importantly, however, occupational disease and, even more dramatically, ‘environmental’ illness were falling out of scientific favor among the very public health experts who had once touted these concepts. The reason may derive from germ theory and its role in preventive medicine. It became increasing clear that the major epidemic diseases (acute and indolent), especially those that attacked the poor and working classes, were infectious in nature and attributable to specific, transmissible microorganisms. With these insights came a major paradigm shift which became known as the ‘new’ public health.

Diffuse concerns over sanitary conditions were replaced with a focused emphasis on controlling contagion or on the innate resistance or vulnerability of the potential host.39,78 Although the discipline of occupational disease was overshadowed to the extent that it could not be illuminated by the new science of microbiology, environmental health concerns beyond bacterial pathogen and vector control were eclipsed almost entirely by the 19th century’s end. Any discussion of air quality in terms of chemical or particulate pollution as a direct or proximate cause of disease was likely to be ignored as little more than miasmic superstition. So too, water- and food-borne illness prevention was approached as if wholly accounted for by microbiological factors.

1900–1970 The 20th century began with a few positive signs of renewed vigor in the discipline of occupational medicine. The first meeting of the International Association for Labor Legislation, the forerunner of the ILO, took place in 1901.73 In the UK, a major multi-authored, state-of-the-art textbook appeared in 1902, edited by Thomas Oliver (later to be knighted for his work in the field).79 Workers’ compensation legislation spread beyond Germany and Great Britain. During this time, agitation for worker protection and compensation for injury in the US was taking place within the context the larger progressive reform movement,80–82 as well as in response to greater labor militancy, particularly among mine workers.83–85 In 1909 the first state workers compensation law was passed in the United States.73 Even environmental health, from the standpoint of industrial point-source air and water pollution, began to receive some renewed attention, albeit sparingly. In Britain, alkali works were known to be a major producer of air contaminants since early in the 19th century. Various frustrated attempts at control finally culminated in the Alkali Works Regulation Acts in 1906.73 In Britain and the US as well, smokestack control began to be evaluated, although negative impacts on property were often emphasized over adverse human health effects.86,87 Nonetheless, change was slow. Despite the modest advances noted above, occupational and environmental health was still dominated by the contagion-control agenda of the broader public health movement. In the US, the field was even further constrained by an emerging corporate influence, a history documented extensively in Christopher Sellers’ scholarly work, Hazards on the Job.88 The First World War over-turned the status quo in occupational and environmental medicine. The industrial expansion of armaments production demanded by modern warfare, particularly in the manufacture of munitions and airplanes, led to outbreaks of chemically related illness for which contagion control had no role.89,90 Governmental investigation and intervention led to revamping existing units or establishing new ones such as the US Public Health Service Bureau of Industrial Hygiene.73,88 Research into the

1900–1970 23 causes and prevention of ‘industrial fatigue,’ which was subsumed within the discipline of occupational health, was also born out of the same First World War driven needs.91 More dramatically, the introduction of chemical gas warfare on a mass scale put to rest completely any lingering dogmatic tenet that germs alone accounted for the only large-scale public health threat worth bothering with.44,92,93 Ironically, it was research on mines and caissons that provided a key underpinning for the British war gas effort through the work of JS Haldane.45 In the years following the First World War, and particularly during the 1930s, renewed attention to occupational illness and injury was infused with a new political awareness. This coincided with the progressive labor movement of the period and drew from its growing strength. In the United States, the apotheosis of this revitalization was Alice Hamilton. Her work in occupational health began at Hull House and she continued to be socially and politically engaged throughout her long career.89,94,95 As with past practitioners, Hamilton’s occupational medicine concerns were driven both by old hazards made worse by new technology, such as silica spread by air-powered tools in quarries, and novel problems altogether, such as the outbreak of neurological disease among the workers in the new rayon industry (as with 19th century rubber workers, the disease was caused by carbon disulfide).89 Dr Hamilton stands out in this period as a great figure, but she was not alone. In Britain, Thomas Legge stepped down from governmental appointment as the Senior Medical Inspector of Factories in protest over Britain reneging on an international agreement to control lead and took a job as an advisor to the Trade Union Congress (an equivalent of the CIO).90 In Spain, a young industrial surgeon named Josep Trueta developed a new system for treating open limb trauma with closed casts. He went on to become a leading Republican military physician in the Spanish Civil War; his closed-cast technique went on to become standard practice for the care of war wounds, saving many lives.96 The degree to which explicit occupational health concerns permeated the non-medical literature of the period is also indicative a wider recognition these problems. In the US, Theodore Dreiser’s Tragic America, a nonfiction critique of the Depression era, specifically catalogues a number of occupational illnesses, including asbestosrelated lung disease.97 In fiction, one of the short stories that first brought success to Albert Maltz, later a major Hollywood screenwriter, was an eerie tale featuring a worker dying of acute silicosis.98 This was one of the first print notices of the Gauley Bridge disaster, later also treated in a proletarian novel.99,100 Written in a slightly earlier period, but only published widely later, Franz Kafka’s stories of the bureaucratic apparatus were imbued with the first-hand experiences that he gained as a successful operative of the Workmen’s Accident Insurance Institute of the Kingdom of Bohemia.101 In the UK, too, non-medical writing also turned to matters of workplace safety and health. The most remarkable novelis-

tic example may be the following, from A.J. Cronin’s The Citadel: ‘He went through the literature on the subject. Its paucity astounded him. Few investigators seemed to have concerned themselves greatly with the pulmonary occupational disease. Zenker had introduced the high-sounding term, pneumokoniosis, embracing three forms of fibrosis of the lung due to dust inhalation. Anthracosis, of course, the black infiltration of the lung met with in coal miners had long been known and was held by Goldman and Trotter in England to be harmless. There were a few treatises on the prevalence of lung trouble in makers of millstones, particularly the French millstones, and in knife and axe grinders … “grinder’s rot” … and stone cutters. There was evidence, mostly conflicting, from South Africa upon … gold miner’s phthisis, which was undoubtedly due to dust inhalation. It was recorded also that workers in flax and in cotton and grain shovelers were subject to chronic changes in the lungs. But beyond that, nothing.’ 102 The Second World War and its immediate aftermath had an effect on occupational medicine, although not as transforming an event as the First World War. Certain sequelae are of note. Women entered the industrial workforce in large numbers, leading to renewed attention to their occupational health hazards103 (although this subject had been of some concern since the 19th century).104–106 The forced exile of large numbers of physicians and biomedical researchers, which crippled medicine in Germany and Austria generally,107 had its impact on the disciplines of occupational medicine and toxicology as well. Ludwig Teleky, for example, one of the leading figures in German occupational medicine, was forced to flee to the US.73 Industrial health-related medical research did not cease in Germany, however, particularly in aerospace medicine, where it was directly incorporated into the war effort. The German human research program on low barometric pressure environments, carried out on concentration camp victims, led to several Nuremberg indictments.108 As in the previous World War, chemical warfare research again was given high priority on both sides, driving toxicologic science to discoveries that would later be to the detriment of occupational health, most saliently through the development of organophosphates.92 Paralleling the legacy of chemical warfare research, the birth of the nuclear weapons industry was also to have long lasting health consequences for a new workforce. It would be difficult to argue, however, that this created significant change within the discipline of occupational medicine, with much of the professional oversight of this issue divested to a new specialization of ‘radiation safety.’ Nonetheless, the threat of atomic warfare, and the reality of ongoing, aboveground testing of nuclear weapons, did mean that airborne pollution was once again destined to be on the public health agenda. Even as the nuclear age was lending pressure to a reordering of public health priorities, other events supervened to accelerate this process. In Donora, Pennsylvania, from October 27th to 31st 1948, the US experienced its first air pollution disaster.39,109,110 Four years later, from the

24 Occupational and Environmental Medicine: the Historical Perspective 5th to 9th of December 1952, a larger and even more deadly episode occurred in England, the so-called ‘killerfog’ of London.111 This was not the very first outbreak of such a crisis. In December of 1930, a similar event had taken place in the Meuse Valley in Belgium. The event was not ignored by scientists, on either side of the Atlantic.112,113 Yet data from the event were limited and subject to conflicting interpretation. Moreover, there was a tendency to dismiss the episode as something of an anomaly. For example, Philip Drinker of Harvard, in one of the first scholarly reviews of air pollution in the US scientific literature, wrote in 1939 that: ‘Naturally, we want to know whether such an accident could occur in industrial America. Our stacks emit the same gases as did the Belgian, but fortunately, so meteorologists tell us, we have no districts in which there is even a reasonable chance of such a catastrophe taking place.’112 In the US, Donora did more than simply prove the weather man wrong. It established air pollution control as a newly recognized and critical need.39 Since there were no personnel trained specifically in this area, the early work in the field fell to the discipline of occupational health, largely to engineers in industrial hygiene. By 1957, the US Pubic Health Service had organized a separate Air Pollution Division, which began to develop criteria documents to address specific pollutants and their potential control.110 In this period there was no matching regulatory movement in occupational safety and health promotion in the US. Not only worker’s compensation, but exposure standard-setting, too, was also a matter of state-by-state control. The only national guidelines available were the threshold limit values of a non-governmental organization, the American Conference of Governmental Industrial Hygienists.84 Nationally and internationally, this was a time of stagnation, if not retrenchment, in occupational medicine. The ILO, which had predated the League of Nations, joined with it, and then survived its demise, continued on with its work based in Geneva. The United Nations headquartered the World Health Organization there as well, but little effective collaboration emerged from this geographic proximity. In Great Britain, the long line of occupational medicine leaders stretching back to the 19th century hygienists and on down through Thomas Oliver and Thomas Legge appeared to have died off, although in 1955 Hunter did publish the first of what was to become many editions of a dominant text in the field.114 On the Continent, the picture was even bleaker, with the sole exception of Italy. There, the major institute for occupational health research, the Milan University Clinica del Lavoro, had survived the fascist period remarkably intact and its director, Emilio Vigliani, took a leadership role in rebuilding the discipline.115 Cold War politics were not conducive to occupational safety and health promotion. In the United Sates certainly, in the aftermath of the McCarthy witch-hunts, any collaboration between academics and organized labor would have been suspect, at best. Outside of academic circles,

occupational medicine practice was corporate based and dominated by the Industrial Medical Association (IMA), whose membership between 1948 and 1959 more than doubled to 4000.116 The IMA’s presidents over this time included the medical directors of Caterpillar Tractor, Inland Steel, Ford Motor, and New England Telephone and Telegraph.116 This period underscores, once again, the particular responsiveness of this discipline to larger political and cultural forces. The first major sign of the ice breaking was not in occupational medicine, but rather in environmental health, in this instance primarily in relation to water and soil, rather than air pollution. Rachel Carson’s seminal Silent Spring, was published in 1962.117 Carson was explicitly concerned with emerging and novel manifestations of environmental damage resulting from a new technology: petrochemical synthetics, particularly chlorinated hydrocarbons. We look back on Silent Spring as a landmark publication, but its importance is not merely in retrospect. Immediately upon its release, its impact was widespread and powerful. Strong conservation organizations pre-dated Silent Spring, but their evolution into an environmental movement is difficult to imagine without it. The links between the environment, particularly non-human health effects, and traditional concerns of occupational disease were not straightforward to the general public, but were not lost on industry. Only recently has data emerged documenting the degree to which chemical manufacturers and their trade groups viewed with alarm the potential political and especially regulatory implications of Rachel Carson’s work.118 The decade in the US that began with Silent Spring ended with the establishment of the Occupational Safety and Health Administration, the National Institute for Occupational Safety and Health, and the Environmental Protection Agency. These regulatory advances, made possible by series of enabling legislative acts, did not occur in a vacuum, culturally, politically, or scientifically.

RECENT HISTORY As the preceding synopsis makes clear, the history of occupational and environmental medicine teaches us that this discipline has a long and complex past. Understanding that past, and the forces that have helped shape it, can better inform our understanding of the issues that face us going forward in an ever-changing world. It is not only convenient, but perhaps prudent as well, to use a cut-off of 1970 for an overview of the historical perspective in occupational and environmental medicine. Nonetheless, it is safe to assume that the interplay of technologic change and social forces continues to exert a substantial effect on the course of the discipline. In the decades of the 1970s and 1980s, the United States assumed a critical leadership role internationally in key environmental areas such as removal of lead from gasoline, sulfur-containing coal emission reductions, and water pollution remediation. The regulatory evolution of occupational health protection in the US was less robust, although promulgation of a new OSHA lead standard in

Additional Resources 25 the 1970s did represent a major advance. Paralleling these trends, environmental medicine has grown to take on an increasingly prominent role within the discipline, particularly in the US. Over the past 40 years, multiple exposure-related outbreaks of occupational or environmental disease have occurred worldwide. Some of the most notable episodes include vinyl chloride-caused angiosarcoma of the liver, kepone-induced neuropathy, methyl mercury-related teratogenesis (Minimata disease), and dibromodichlorpropane(DBCP)-induced male sterility. Although definitive historical assessments of these events have yet to be written, the importance of technologic and social factors is abundantly clear in the causes, identification, and struggle to control these outbreaks. The same confluence of forces is no less relevant to ongoing issues faced by occupational and environmental medicine.

ADDITIONAL RESOURCES A complete history of occupational and environmental health is far beyond the limitations of a single chapter in a general text. There are, however, many additional resources that the interested reader can pursue in order to gain supplemental information. In recent years, a number of excellent historical analyses addressing various aspects of occupational health have been published. These include Hazards on the Job (Sellers),88 Deadly Dust (Rosner and Markowitz),119 Hawk’s Nest Incident (Cherniack),99 The Bends (Phillips),45 Workers’ Health, and Workers’ Democracy: The Western Miners’ Struggle, 1891–1925 (Derickson),85 and Occupation and Disease (Denbe).120 In addition to these works, there are noteworthy earlier histories. The two most important books that helped define the field of occupational health history are Rosen’s The History of Miners’ Diseases (1943)7 and Teleky’s Factory and Mine Hygiene (1948).73 In the 1950s, the scholarly work of Meiklejohn was also groundbreaking.59,66,121 From a somewhat later period, noteworthy works with considerable historical material include Death on the Job (Berman)84 and Expendable Americans (Brodeur).122 In terms of 20th century memoirs by occupational medicine practitioners, the most important is Alice Hamilton’s Exploring the Dangerous Trades89 (and of related autobiographical interest, Sicherman’s Alice Hamilton: A Life in Letters).95 Another American occupational health memoir of the same period is that of McCord, A Blind Hog’s Acorns.123 Some of the key library resources in occupational and environmental medicine also should be noted. The greatest single focused collection in the field of occupational medicine was undoubtedly that of Alfred Whittaker, which was later acquired by the Blocker History of Medicine Collections at the University of Texas, Galveston.124 This collection includes the only extant copy of Ellenbog’s On the Poisonous Wicked Fumes and Smokes of Metals (the earliest printed work in occupational medicine).14 Another small, but important collection in the history of occupational medicine was donated by Robert Legge to the University of California San Francisco.

Occupational health titles are well represented in most of the great medical history collections, although some are particularly noteworthy in this regard, in particular that of the Wellcome Medical Trust (London). In addition to a large number of important printed items, it also holds other material, including the occupational medicine archives of Donald Hunter. The mining collection of Herbert Hoover, housed at Claremont College, contains a number of rare items especially relevant to mining safety and health.125 A major resource on microfilm is available through the Goldmiths’-Kress Library of Economic Literature, which reproduces a 60,000 document collection held at the University of London and Harvard University and includes many items relevant to occupational history prior to 1850.126 Finally, the medical bibliography of Garrison and Morton provides an excellent citation listing of many of the key texts (including journal articles) that constitute landmarks in the history of the discipline.127

References 1. Foucault M. The birth of the clinic (translated by AH Sheridan Smith). New York: Vintage Books, 1973. 2. Starr P. The social transformation of American medicine. New York: Basic Books, 1982. 3. Goldwater LJ. From Hippocrates to Ramazzini: early history of industrial medicine. Ann Med History New Series 1936; 8:27–35. 4. Legge RT. The history of industrial medicine and occupational diseases. Ind Med 1936; 5:300–14. 5. Sigerist HE. The Wesley M. Carpenter Lecture. Historical background of industrial and occupational diseases. Bull New York Acad Med 1936; 12(11):597–609. 6. Humphrey JW, Oleson JP, Sherwood AN. Greek and Roman technology: a source book. London: Routledge, 1998. 7. Rosen G. The history of miners’ diseases. New York: Schuman, 1943. 8. DiPorto A, Gagliardi L. Prohibitions concerning polluting discharges in Roman law. In: Grieco A, Iavicoli S, Berlinguer G, eds. Contributions to the history of occupational and environmental prevention. Amsterdam: Elsevier, 1999;211–21. 9. Chuwers P, Neumark Y. Worker health and environmental protection in Biblical and Talmudic sources. In: Grieco A, Iavicoli S, Berliguer G, eds. Proceedings, First International Conference on the History of Occupational and Environmental Protection. Rome: ISPESL, National Institute for Occupational Safety and Prevention, 1998;135. 10. Pare A. The apologie and treatise of Ambroise Pare containing the voyages made into divers places with many of his writings upon surgery. Edited with and introduction by Geoffrey Keynes, Chicago: University of Chicago Press, 1952. 11. McCord CP. Scurvy as an occupational disease: IV. Scurvy and the nations’ men-of-war. J Occup Med 1971; 13:441–7. 12. Wedeen RP. Poison in the pot: the legacy of lead. Carbondale: Southern Illinois University Press, 1984. 13. Adams F. The medical works of Paulus Aegineta. Vol 1. London: J Welsh, 1834. 14. Ellenbog U. On the poisonous evil vapors. Lancet 1932; 1:230-1. 15. Sigerist HE, ed. Four treatises of Theophrastus von Hohenheim called Paracelsus. Baltimore: Johns Hopkins Press, 1941. 16. Agricola G. De re metallica. Translated by Herbert Clark Hoover and Lou Henry Hoover. New York: Dover, 1950. 17. Sisco AG, Smith CS. Lazarus Ercker’s treatise on ores and assaying. Chicago: University of Chicago Press, 1951.

26 Occupational and Environmental Medicine: the Historical Perspective 18. Birunguccio V. The pirotechnia. Sisco AG and Gnudi MT, translators. New York: American Institute of Mining and Metallurgical Engineers, 1942. 19. Nef JU. Coal mining and utilization. In: Singer C, Holymard EJ, Hall AR, Williams TI, eds. A history of technology. Vol III. New York: Oxford University Press, 1957;72–88. 20. Brimblecombe P. The big smoke. London: Methuen, 1987. 21. Li Ch’iao-p’ing. The chemical arts of old China. Easton Pennsylvania: Journal of Chemical Education, 1984. 22. Needham J. The development of iron and steel technology in China. London: Newcomen Society, 1958. 23. Singer C. East and west in retrospect. In: Singer C, Holymard EJ, Hall AR, Williams TI, eds. A history of technology. Vol II. New York: Oxford University Press, 1956;753–76. 24. Ying-Hsing S, T’ien-Kung K’ai-Wu. Chinese technology in the seventeenth century. University Park and London: The Pennsylvania State University Press, 1966. 25. Ramazzini B. De morbis artificum diatriba. Modena: Capponi, 1700. 26. Ramazzini B. Treatise on the diseases of tradesmen. London: Thomas Osborne, 1746. 27. Ramazzini B. De morbis artificum Bernardini Ramazzini diatriba. Disease of workers. The Latin text of 1713 revised, with translation and notes by Wilmer Cave Wright. Chicago: University of Chicago Press, 1940. 28. Ramazzini B. Essai sur maladies des artisans. M. Fourcroy, translator. Paris: Moutard, 1777. 29. Ramazzini B. Abhandlung von den krankheiten der kunstler un handweker neu bearbeit und vermehret von Johann Christian Gottlieb Ackermann. Stendal: D.C. Franzen und J.C. Gross, 1780, 1783. 30. Morgagni G. The seats and causes of diseases. Translated by Benjamin Alexander. (Facsimile of the 1769 edition). Birmingham: Classics of Medicine Library, 1983. 31. Tissot SA. An essay on the disorders of people of fashion; and a treatise on the diseases incident to literary and sedentary persons. Edinburgh: A Donaldson, 1772. 32. de Begontini A. Bernadino Ramazzini da morbis artificum prosequito. Vienna: Typis Geroldianis, 1778. 33. Giesl JF. Bernadino Ramazzini da morbis artificum prosequito. Vienna: Typis Geroldianis, 1778. 34. Tralles JW. De Praeservandis artificum et opificum morbis. Magdeberg: Johan Christian Hendel, 1745. 35. Skragge N. Disertationem medicam, qua morbi artificum leviter adumbrantur. Uppsala: Uppsala University, 1765. 36. Lindfors AO. Linnes dietetik. Uppsala: Akadmeiska Boktryckeriet, 1907;64–74. 37. Evelyn J. Fumifugium: or the inconvenience of the aer and smoake of London dissipated. Dorchester, Dorset and London: National Society for Clean Air, 1961. 38. Evelyn J. Fumifugium: or the inconvenience of the aer and smoake of London dissipated, 2nd edn. London: B White, 1772. 39. Blanc PD, Nadel JA. Clearing the air: the links between occupational and environmental air pollution control. Public Health Rev 1994; 22:251–270. 40. Baker G. An essay concerning the cause of the endemial colic of Devonshire, which was read in the theatre of the College of Physicians, in London, on the twenty-ninth day of June, 1767, 2nd edn. London: Payne and Foss, 1814. 41. Gillispie CC, ed. A Diderot pictorial encyclopedia of trades and industry. Manufacturing and the technical arts in plates selected from ‘L’Encyclopedie, ou dictionnaire raisonne des sciences, des arts et des metiers’ of Denis Diderot. New York: Dover Publications, 1987. 42. Baines E. History of the cotton manufacture in Great Britain. London: H Fisher, R Fisher, and P Jackson, 1835. 43. Clow A, Clow NL. The chemical industry: interaction with the industrial revolution. In: Singer C, Holmyard EJ, Hall AR, Williams TI, eds. A history of technology. Oxford: Clarendon Press, 1958; 230–56.

44. Underhill FP. The lethal war gases: physiology and experimental treatment. New Haven: Yale University Press, 1920. 45. Phillips JL. The bends. New Haven: Yale University Press, 1998. 46. Geist L. Die regeneration des unterkiefers nach totaler necrose durch phosphordampfe. Erlagen: Verlag von Ferdinand Enke, 1852. 47. Delpech A. Industrie au caoutchouc souffle. Researches sur l’intoxication speciale que determine le sulfure de carbone. Annales d’ Hygiene Public et de Medicine Legale (Series 2) 1863; 19:65–183. 48. Richardson BW. Health and occupation. London: Society for Promoting Christian Knowledge, 1879. 49. Walsh D. Deep tissue traumatism from roentgen ray exposure. Br Med J 1897; ii:272. 50. Greenhow EH. On chronic bronchitis escpecially as connected with gout, emphysema and diseases of the heart, being clinical lectures delivered at Middlesex Hospital. London: Longmans, 1869. 51. Holland CG. Diseases of the lungs from mechanical causes. London: John Churchill, 1843. 52. Oliver T. Lead poisoning in its acute and chronic forms. The Goulstonian Lectures, delivered in the Royal College of Physicians, March, 1891. Edinburgh & London: Young J Pentland, 1891. 53. Blanc PD. Metal fume fever from a historical perspective. In: Grieco A, Iavicoli S, Berlinguer G, eds. Contributions to the history of occupational and environmental prevention. Amsterdam: Elsevier, 1999; 211–21. 54. Chadwick E. Report on the sanitary condition of the labouring population of Great Britain. London: W Clowes for HM Stationery Office, 1843. 55. Simon J. Public health reports (2 vols). London: The Sanitary Institute, 1887. 56. Nightingale F. Florence Nightingale’s notes on nursing. Edited with an introduction, notes and guide to identification by Victor Skretkowicz. London: Scutari Press, 1992. 57. Thackrah CT. The effects of the principal arts, trades, and professions, and of civic states and habits of living. London: Longman, 1831. 58. Thackrah CT. The effects of arts, trades, and professions, and of civic states and habits of living, on health and longevity, 2nd edn. London: Longman, 1832. 59. Meiklejohn A. The life and times of Charles Turner Thackrah. Edinburgh: E & S Livingstone, 1957. 60. Noble D. Facts and observations relative to the influence of manufactures upon health and life. London: John Churchill, 1843. 61. Lee CA. On the effects of arts, trades, and professions, as well as habits of living, on health and longevity. Family Magazine (New York) 1840–41; 8:175–7,212–5,270–2,302–5. 62. McCready BW. On the influence of trades, professions and occupations in the United States in the production of disease – being the prize dissertation for 1837. Transactions of the Medical Society of the State of New York 1836–37; 3:91–150. 63. House of Commons, Great Britain. A copy of the report presented to the House of Commons by the committee appointed to examine the several petitions, which have been presented to the House, against the employment of boys in sweeping of chimneys. London: House of Commons, 1817. 64. Fleming LE, Ducatman AM, Shalat SL. Disease clusters: a central and ongoing role in occupational health. J Occup Med 1991; 33:818–25. 65. Holdsworth C. Dr. John Thomas Arlidge and Victorian occupational medicine. Med Hist 1998; 42:458–75. 66. Meiklejohn A. Industrial health – meeting the challenge. Br J Ind Med 1959; 16:1–10. 67. Chevalier L. Laboring classes and dangerous classes in Paris during the first half of the nineteenth century (translated by Frank Jellinek). New York: Howard Fertig, 1973.

Additional Resources 27 68. Patissier P. Traite des maladies des artisans. Paris: Chez J.-B. Bailliers, 1822. 69. Tanquerel des Planches L. Traite des maladies de plomb ou saturnines. Paris: Ferra, 1839. 70. Orfila MP. Traite des poisons tires des regnes mineral, vegetal et animal, ou Toxicologie generale, consideree sous les rapports de la physiologie, de la pathologie et de la medecine legale, 2nd edn. Paris: Crochard, 1818. 71. Bernard C. Lecons sur les effets des substances toxiques. Paris: J.B. Baillier, 1857. 72. Hirt L. Die krankheiten der arbeiter. Breslau: F. Hirt, 1871–1878. 73. Teleky L. History of factory and mine hygiene. New York: Columbia University, 1948. 74. McCord CP. Occupational health publications in the United States prior to 1900. Ind Med Surg Med 1955; 24:363–8. 75. Arlidge JT. The hygiene, diseases, and mortality of occupations. London: Percival and Company, 1892. 76. Proust A. Traite d’hygiene publique et privee. Paris: G. Masson, 1877. 77. Santenson CG. Ueber chronische vergiftungen mit steinkohlentheerbenzin; vier todesfalle. Archiv fur Hygiene 1897; 31:336–76. 78. Sellers C. The Public Health Services Office of Industrial Hygiene and the transformation of industrial medicine. Bull Hist Med 1991; 65:42–73. 79. Oliver T, ed. Dangerous trades. The historical, social, and legal aspects of industrial occupations as affecting health, by a number of experts. London: John Murray, 1902. 80. Eastman E. Work accidents and the law. The Pittsburgh Survey. New York: Russell Sage Foundation (Charities Publishing Committee), 1910. 81. Kober GM. Industrial and personal hygiene. A report of the Commission on Social Betterment. Washington DC: The President’s Homes Commission, 1908. 82. Overlock MG. The working people, their health and how to protect it. Boston: Health Book Publishing Co, 1911. 83. Andrews JB. Labor problems and labor legislation. New York: American Association for Labor Legislation, 1919;65–92. 84. Berman D. Death on the job. New York: Monthly Review Press, 1978. 85. Derickson A. Workers’ health and workers’ democracy: the western miners’ struggle, 1891–1925. Ithaca, New York: Cornell University Press, 1988. 86. Haywood JK. Injury to vegetation and animal life by smelter wastes. U.S. Department of Agriculture Bulletin 113. Washington DC: US Government Printing Office, 1910. 87. Royal Sanitary Institute. Addresses, papers, and discussions at conference on smoke abatement, London, Dec. 12th–15th, 1905. London: Royal Sanitary Institute, 1906. 88. Sellers C. Hazards on the job: from industrial disease to environmental health science. Chapel Hill: University of North Carolina Press, 1997. 89. Hamilton A. Exploring the dangerous trades. Boston: Little, Brown, 1943. 90. Legge TM. Industrial maladies. London: Humphrey Milford Oxford Press (Oxford Medical Publications), 1934. 91. Brown AB. The machine and the worker. London: Ivor Nicholson and Watson Ltd, 1934. 92. Blanc PD. The legacy of war gas. Am J Med 1999; 106:689–90. 93. Fauntleroy AM. Report on the medico-military aspects of the European war. Washington DC: US Government Printing Office, 1915. 94. Grant MP. Alice Hamilton. Pioneer doctor in industrial America. London: Abelard-Schuman, 1967. 95. Sicherman B. Alice Hamilton, a life in letters. Cambridge: Harvard University Press, 1984. 96. Trueta J. Treatment of war wounds and fractures with special reference to the closed method as used in the war in Spain. London: Hamish Hamilton, 1939. 97. Dreiser T. Tragic America. New York: Horace Liveright, 1931;19,196.

98. Maltz A. The way things are. New York: International Publishers, 1938. 99. Cherniack M. The hawk’s nest incident: America’s worst industrial disaster. New Haven: Yale University Press, 1986. 100. Skidmore H. Hawk’s nest. New York: Doubleday Doran and Co., 1941. 101. Pawel E. The nightmare of reason. A life of Franz Kafka. New York: Farrar Strauss Giroux, 1984;181–9. 102. Cronin AJ. The citadel. London: Victor Gollancz, 1937;209. 103. Mettert MT. The occurrence and prevention of occupational diseases among women. US Department of Labor, Bulletin of the Women’s Bureau No. 184. Washington DC: US Government Printing Office, 1941. 104. Ames A. Sex in industry: a plea for the working girl. Boston: James R. Osgood and Company, 1875. 105. Hamilton A. Women workers and industrial poisons. Bulletin of the Women’s Bureau No. 57. Washington DC: US Government Printing Office, 1926. 106. Hutchins G. Women who work. International Pamphlets No. 27. New York: International Publishers, 1932. 107. Ernst E. A leading medical school seriously damaged: Vienna 1938. Ann Intern Med 1995; 122:789–92. 108. West JB. Highlife: a history of high-altitude physiology and medicine. New York: Oxford University Press, 1998;246–53,427. 109. Schrenk HH. Causes, constituents and physical effects of smog involved in specific dramatic episodes. Arch Ind Hyg Occup Med 1950; 1:189-94. 110. Whittenberger JL. Health effects of air pollution: some historical notes. Environ Health Persp 1989; 81:129–30. 111. Logan WPD. Mortality in the London fog incident, 1952. Lancet 1953; 1:336–8. 112. Drinker P. Atmospheric pollution. Ind Engineer Chem 1939; 31(11):1316–20. 113. Roholm K. The fog disaster in the Meuse Valley, 1930: a fluorine intoxication. J Ind Hyg Toxicol 1937; 19:126–37. 114. Hunter D. The diseases of occupations. London: English Universities Press, 1955. 115. Grieco A, Chiappino G, Alessio L et al. La scomparsa del Prof. Encrio Vigliani. Med del Lavoro 1992; 83:4–17. 116. Selleck HB, Whittaker AH. Occupational health in America. Detroit: Wayne State University Press, 1962. 117. Carson RL. Silent spring. Cambridge: Houghton Mifflin, 1962. 118. Cushman JH Jr. After ‘Silent Spring,’ industry spin on all it brewed. New York Times, March 26, 2001;A14. 119. Rosner D, Markowitz G. Deadly dust: silicosis and the politics of occupational disease in twentieth-century America. Princeton, New Jersey: Princeton University Press, 1991. 120. Denbe A. Occupation and disease: how social factors affect the conception of work-related disorders. New Haven: Yale University Press, 1996. 121. Meiklejohn A. John Darwall, M.D. (1796–1833) and ‘Diseases of Artisans’. Br J Ind Med 1956; 13:142–51. 122. Brouder P. Expendable Americans. New York: Viking Press, 1974. 123. McCord CP. A blind hog’s acorns: vignettes of the maladies of workers. Chicago: Cloud Inc, 1945. 124. Moody Medical Library. The Truman G. Blocker, Jr. History of Medicine Collections: books and manuscripts. Galveston: University of Texas Medical Branch, 1986. 125. Claremont College. Hoover collection on mining and metallurgy. Claremont, California: Arcon Press, 1980. 126. Goldsmiths’-Kress Library. Goldmsiths’-Kress library of economic literature: a consolidated guide to segment I of the microfilm collection. (5 vol.). Woodbridge: Research Publications, Inc., 1976. 127. Morton L. Morton’s medical bibliography: an annotated check-list of texts illustrating the history of medicine (Garrison and Morton), 5th edn. Brookfield: Gower, 1991.

Chapter 3 Approach to the Patient Mark R Cullen, Linda Rosenstock As noted in the Introduction, OEM includes both clinical and public health aspects. In this chapter, principles of clinical practice are explored that distinguish the field from the core disciplines of clinical medicine – internal medicine, orthopedics, dermatology, neurology, etc – with which the reader may be more familiar. One very obvious distinction is that clinical OEM includes two very different components. The bread and butter of the field has come to be known as ‘primary care’ occupational medicine (there is no counterpart on the environmental side) and includes activities that have existed, primarily within heavy industry, long before they were glamorized by this designation or became the subject of scientific scrutiny. Primary care components include, among other things, the conduct of preplacement physical examinations; care of minor injuries and immediately recognized adverse effects of over-exposure in the workplace; return to work examinations and medical screening. Historically these services were deemed sufficiently straightforward that any practitioner of medicine would be qualified to perform them without additional or specific OEM training. Changes in public perception, regulations, rapidly advancing knowledge and legal/economic considerations have rendered each aspect worthy of formal training, now incorporated into the residency requirements for specialty training in OEM in the US.

PRIMARY CARE OCCUPATIONAL MEDICINE Preplacement evaluation Preplacement evaluation has superseded the previous construct of pre-employment examination because of societal discrimination concerns; in the US and many other countries, individuals may not be assessed medically prior to being offered provisional employment as a matter of law. Once a job offer has been made, an exam may be conducted – and for jobs entailing any substantial risks undoubtedly should – to determine that the employee is physically and mentally qualified to perform the essential functions of the job safely. The legal aspects of this are discussed in Chapter 57.3. From a medical perspective, the examination must be tailored to the specific physical requirements and hazards of the job; general considerations, such as pre-existing health conditions or disabilities, are irrelevant except insofar as they would interfere with the job. For this reason, primary care occupational medicine practice requires extensive knowledge about the work setting and job requirements for

each position for which any exam may be performed. This knowledge must be sufficiently detailed to address any challenge that an employee has been discriminated against for any reason other than ability to perform the essentials of the job. While it is perfectly acceptable for the examiner – who does not enjoy a traditional doctor–patient relationship with such individuals – to inquire about or evaluate other health issues or behaviors (about which there may be concern from a preventive health perspective) he or she may do so only insofar as three things are borne in mind: such investigations must be voluntary, and the employee aware of that; the information must be maintained confidentially and not used to make employment decisions; and similar investigations must be requested of all new employees, irrespective of the positions for which they have been hired (including managers!).

Management of minor injuries and responses to work hazards The routine management of minor injuries and acute responses to hazards at work falls into the purview of outpatient emergency medicine or surgery, and is beyond the scope of this text. However, as with preplacement, there are important occupational health components. First, it is crucial that the underlying causes of the illness or injury be investigated in order to prevent further injuries from occurring. This may or may not be the responsibility of the primary care provider per se, but it is her or his obligation to make sure that all available information from the evaluation relevant to the event is provided to those at the workplace or company who are responsible. Second, since most employers are now eager to limit lost work time and generally prefer early return to work, even if in restricted roles, the provider must be cognizant of the demands of the patient’s job, to determine whether or not the injury or illness precludes it, unless there are overarching medical reasons for lost time such as the need for hospitalization, work-limiting medications or bedrest. Pressures from employee or his/her supervisor for premature or inappropriate return to work must be resisted, as must the temptation to prescribe time off for convenience or social reasons. Close communication with both the employee and the manager are essential in every case.

Evaluation for return to work Evaluation for return to work is an extension of the above activities, and requires review of medical reports regarding

30 Approach to the Patient any disabling illness – work related or otherwise – and then reiterating the preplacement approach in light of the new information. Even more crucially in this context, information without direct bearing on the patient’s specific job is not relevant, and must not be communicated in any way to the employer.

present or that the patient may be at high risk for subsequent occurrence; and 4. assessment of disability and/or potential for work rehabilitation. In the sections that follow we will concern ourselves with the first three scenarios; assessment of impairment and disability is the subject of Chapter 8.

Medical screening Finally there is medical screening. This activity involves the collection of information from the patient – it may be a questionnaire, a hearing test, a blood test etc – to assess either level of exposure to a hazard or a possible preclinical health effect. The primary purpose is to protect the individual worker, although results are often scrutinized for other purposes, such as part of a surveillance program to prevent adverse workplace effects more generally (discussed more fully in Chapter 4.3). The first level of interpretation, however, pertains to the wellbeing of the individual tested. In this way, results may demonstrate an over-exposure or a work-connected effect. In either case the practitioner is obliged to inform the worker explicitly of that fact, and assure appropriate response such as reduction of exposure, removal, compensation and/or treatments as necessary. Often, however, abnormalities detected on screening exams are not job related in origin, nor directly affect the ability of someone already doing the job to continue doing it (although occasionally they might, and in that situation must be addressed). When non-work connected abnormalities are suspected, two things become essential: the provider must inform the employee of the abnormal finding and her/his opinion regarding its origins, and the provider must share responsibility that the findings are appropriately followed up.

TERTIARY CARE OCCUPATIONAL AND ENVIRONMENTAL MEDICINE Unfortunately, despite all of the preventive aspects of those charged with making the workplace and outside environment free of excess hazard, some adverse consequences of exposure continue to occur; many more are suspected. Evaluation of such patients constitutes the ‘tertiary’ aspect of OEM practice, one that prior to a few decades ago could not be said to exist in organized form in the US or most other parts of the world. It is to that aspect that the rest of this chapter is devoted. Patients, physicians, and third parties typically prompt diagnostic consultation with an OEM specialist for one of four reasons: 1. the patient (or referring party) suspects that symptoms, signs, or laboratory abnormalities may be due to some environmental or occupational factor; 2. a disorder has been diagnosed where the cause may not be evident: the question arises as to whether it may be due to an environmental or occupational factor; 3. exposure to a suspected harmful agent has raised concern that early manifestations of disease may be

CLINICAL EVALUATIONS IN OEM In addition to the usual methods of clinical diagnosis, three tools are special to OEM practice: the occupational and environmental history, the environmental evaluation, and the use of specialized tests to establish causal associations. Despite variations in applicability, the principles are common to every case.

Occupational and environmental history and evaluation The occupational health history is fundamental to the assessment of the work-relatedness of health problems; as such, even in abbreviated form it should become a routine component of every comprehensive health history, not exclusive to OEM referral care.1,2 The environmental history complements the occupational health history by probing for the presence of non-occupational factors and their possible role in the disease process. The occupational and environmental history has multiple purposes. 1. To increase awareness of occupational and environmental factors. It is more the exception than the rule that clues to the potential role of these factors emerge from the physical examination or routine laboratory testing. Unless this history is specifically elicited or otherwise offered by the patient, the opportunity will be lost to consider occupationally and environmentally related disease or risk. 2. To make accurate medical diagnoses. Failure to obtain the history in the setting where occupational and environmental factors have played a role inevitably results in at least a partial misdiagnosis. For example, if fatty liver disease is correctly diagnosed but is attributed to alcohol over-consumption when solvent exposures have also played a role, then the diagnosis of alcoholic fatty liver disease is not correct and important treatment interventions may be overlooked. 3. To prevent the development of occupational and environmental disease. By using the occupational and environmental health history as a screening tool, identification of exposures to potentially hazardous factors can result in the reduction or elimination of these exposures. This factor may be beneficial in the setting where exposures cause diseases of long latency, as well as those responsible for acute and recurrent conditions. Identifying past asbestos exposure, for example, may render counseling about smoking cessation more effective when this counseling is

Clinical Evaluations in OEM 31

4.

5.

6.

7.

8.

provided in the setting of education about the synergistic effect of exposure to both carcinogens. In the case of exposures causing acute conditions – for example, pulmonary allergens – interventions to decrease exposure are likely to reduce the person’s risk of subsequently developing hypersensitivity. To prevent the aggravation of underlying medical conditions by occupational and environmental factors. The smoker with chronic bronchitis who is exposed to respiratory irritants in the workplace will, regardless of effectiveness of smoking cessation interventions, benefit by reducing his or her exposure to identified respiratory irritants. Similarly, in addition to optimizing glucose control in the individual with diabetes mellitus, avoidance of exposure to agents that may cause peripheral neuropathy is also an important intervention, because the individual predisposed to a peripheral neuropathy of any cause may be at increased risk for damage from environmental peripheral neurotoxins. To identify potential workplace hazards. In addition to using the occupational and environmental history as a screening tool to identify and ameliorate the risk of exposure to hazards, the history can help identify factors that would otherwise not be suspected as injurious. For example, the worker who presents with a persistent nocturnal cough may have been exposed to an irritating or sensitizing agent, initiating bronchial hyper-responsiveness manifested as cough. To detect new associations between exposures and disease. The field of occupational and environmental medicine is rapidly evolving. As more interest and attention are paid to environmentally induced illnesses, more is learned about the nature and extent of adverse effects of specific agents. Perhaps no other field offers the potential to uncover, through the evaluation of an individual patient, a previously unknown association between exposure and disease. Examples include adding to the list of now over 200 agents known to induce specific asthmatic responses, identifying new neurologic syndromes as chronic sequelae of past intoxications, and discovering new renal and hepatic toxins. To establish the basis of compensation for occupational and environmental disease. Whether for workers’ compensation for occupational disease or liability claims, the physician plays a key role in determining the likelihood that an environmental exposure has caused a given medical condition. The patient’s history of exposure, its onset, intensity, and duration – sometimes alone or in conjunction with other available exposure information – is fundamental to this assessment. To establish rapport with patients. This last benefit of the occupational and environmental health history in many ways is a secondary and unexpected bonus to the original objectives. In our experience, it is remarkable how often encouraging an otherwise taciturn person to describe the details of his or her job

facilitates a more relaxed and congenial medical evaluation. Demonstration of a physician’s interest in activities fundamentally important to the patient can lessen anxieties attendant with first time physician–patient encounters.

Components of the occupational and environmental history The occupational and environmental history can be obtained in several ways. One approach is to integrate a series of key questions directly into the routine health history. Another approach is to incorporate a screening history with all new visits, selectively updating this procedure as indicated. In either approach, however, the occupational and environmental history has two main components: the employment and exposure history and the occupationally and environmentally related health history. The first component contains information about current and past jobs as well as non-occupational environmental exposures. The health history component uses questions to elicit information about health problems and symptoms in relation to specific exposures and work settings, and about the existence of symptoms or illnesses in coworkers, household members, or community residents. A sample history form that can be self-administered and maintained as part of the patient’s database is shown in Figure 3.1. This form can serve as a screening tool; clinical judgment will determine when it is appropriate to take a more comprehensive history. Because in some clinical settings even this shortened form may not be readily administered, we are often asked what few questions should be asked of all new patients. A survey of members of the United States Association of Occupational and Environmental Medicine Clinics (AOEC) found the following three questions essential. 1. Please describe your job. 2. Have you ever worked with any health hazard, such as asbestos, chemicals, noise, or repetitive motion? 3. Do you have any health problems that you believe may be related to work? The following section describes the elements of the two core components of the occupational and environmental health history in greater detail, as would be appropriate to OEM tertiary evaluation.

Work and exposure history This component of the occupational and environmental health history extends beyond what might be routinely obtained in the clinical setting. Nonetheless, the objectives of the work history are similar to many other aspects of social history, particularly those that include identifying in an individual risk factors that indicate the need for prevention or intervention strategies. The following discussion is relevant to information about the current or most recent job or, in some instances, a previous job of concern regarding the problem under evaluation. For example, if a patient is being evaluated for

32 Approach to the Patient

Years From To I. Work and exposure history A. Current employment Questions 1–7 refer to your current or most recent job. 1. Job title

Job title

Exposures

Job #2 Job #3 Job #4 Job #5 Job #6 Job #7 Job #8 Wartime employment

2. Type of industry 3. Name of employer 4. Year job began Still working? Yes No

C. Other exposures If no, year job ended

5. Briefly describe this job, noting any part that you feel may be hazardous to your health.

6. Do you wear protective equipment on this job? Yes No If yes, check equipment used: Gloves Mask respirator Hearing protection

Air supply respirator Coveralls or aprons Safety glasses

7. In this job, are you exposed to any of the following? If yes, mark those to which you are exposed: Fumes and dust Solvents Vibration Emotional stress

Elements and metals Other chemicals Excess heat/cold Noise

1. Does anyone in your household work at a job that you suspect involves exposures that may be brought home from work (e.g., asbestos fibers on clothes)? Yes No 2. Are there any industries in the area in which you live that may pollute your environment? Yes No 3. Do you have any hobbies that expose you to chemicals, metals, or other substances? Yes No 4. Have you ever smoked cigarettes? ("No" means less than 20 packs of cigarettes in your entire life.) Yes No If yes, please answer the following: a. Do you now smoke cigarettes (that is, as of 1 month ago)? Yes No b. How many years have you smoked? c. Of the entire time you have smoked, about how many cigarettes per day do or did you smoke on the average?

Other

B. Employment history It is important that we know all the jobs you have had. Job #1 is your current or most recent job. Beginning with the job before this one–– Job #2––please fill in as much of the information requested as you can remember, and continue to do so until all previous jobs have been listed. Include any military service you have had. If you need additional space, use the back of this form.

II. General health history* 1. Is there any particular hazard or part of your job that you think relates to your problems? Yes No 2. Do any of your coworkers have problems or complaints similar to yours? Yes No * For each positive response to review of systems, ask whether symptoms are better, worse, or no different in association with work

Figure 3.1: A sample of a screening occupational and environmental history form, which can be self-administered and serve as the basis for a more comprehensive history.

suspected occupational asthma, the key to a successful diagnosis is to focus on the job when the patient first began having symptoms. In addition to asking for the patient’s job title (or occupation), it is important to ascertain the specific nature of the job. This information can be obtained by asking additional questions about the industry. For example, a painter in a shipyard is subject to different exposures from a painter in a residential setting. Hence, the key questions are: ‘What product or service does your employer make or produce?’ and ‘What aspect do you do on your job?’ If the job is already familiar to the physician, then the question may be modified; for example, one may ask ‘Is there anything you do now that is different from past jobs where you’ve been an electrician?’

It may be helpful to ask the patient to describe a typical work day. The screening history inquires about the use of personal protective equipment (PPE). Although the provision of good protective equipment should mitigate risks of exposure, it must be kept in mind that those who use protective equipment are often at increased risk for work-related illnesses. The protective equipment serves as a clue that hazardous materials are present, and exposure at least possible. For this reason a candid appraisal of the equipment’s actual use is invaluable as well. The patient should be asked directly about potentially hazardous exposures that are present at work, whether of biologic, chemical, physical, or psychologic origin. A

Clinical Evaluations in OEM 33 checklist approach (as illustrated in Fig. 3.2) can be used to direct this inquiry. If a patient gives a positive response within any category, then further information can be obtained about specific exposures. For an overview of the patient’s occupational and environmental history and for an evaluation of those conditions of long latency, such as cancer or pneumoconiosis, the history must include information about past jobs and exposures and potential important exposures in the non-work environment. An abbreviated history of all employment is provided in the sample history form (Fig. 3.1). Sometimes an individual omits information about employment during military service; therefore, this information, which may indicate that the individual was subject to other toxic exposures, should be specifically sought.

A. Current or most recent job (paid work) List of exposures

Yes

If yes check one Low Med High

Many workers are well informed about specific exposures. In other instances, however, the exposure history requires further investigation to identify specific constituents of products and exposure levels. Consideration of the exposure dose is important in identifying, preventing, and managing occupational diseases; the history is an important first step in establishing the level of exposure. Although it is by no means precise, the patient’s assessment of relative levels of exposure (i.e., low, medium, high) for specific agents can be valuable. One way of eliciting this information is shown in Figure 3.2, a portion of a comprehensive occupational and environmental history that can also be self-administered. Here, the patient is given an opportunity to report potential exposure to various widely prevailent hazards that appear or have appeared in the current or any past job. For physical

B. Any previous job

Yes

If yes check one

C. Any activity outside paid work

Yes

Low Med High

Example Asbestos 1. Fumes and dust Asbestos Plastic fumes Welding fumes Fumes (other) Glass (e.g. fiberglass) Silica (e.g. sand) Plaster Wood (specify type(s) if known:

)

Other (specify if known:

)

2. Elements and metals Aluminum Arsenic Cadmium Chromium Copper Lead Mercury Nickel Zinc Other (specify if known:

)

Figure 3.2: Sample of detailed self-report form for specific occupational and environmental exposures.

If yes check one Low Med High

34 Approach to the Patient

A. Current or most recent job (paid work) List of exposures

Yes

If yes check one Low Med High

B. Any previous job

Yes

If yes check one Low Med High

C. Any activity outside paid work

Yes

If yes check one Low Med High

3. Solvents Alcohols (e.g. methyl, wood) Benzene Toluene, xylene, naphtha Paint, varnish, degreasers Tri-, tetrachloroethylene Other (specify if known:

)

4. Other chemicals Acids Alkali (caustics) Ammonia Herbicides and pesticides Dyes

Figure 3.2 (cont’d): Sample of detailed self-report form for specific occupational and environmental exposures.

hazards it is often best to have the patient compare levels to common comparisons, such as ‘loud as a subway’. Sometimes acting out an activity or task may be useful for demonstrating potential risk, especially for musculoskeletal effects. As noted, the nature and use of protective equipment may provide an additional clue about levels of exposure, as would information about the general cleanliness of the workplace and the adequacy of ventilation. Finally, for some exposures that have acute as well as chronic effects, such as organic solvents, the description of immediate symptoms in relation to exposure – e.g., acute intoxication or headache – may provide evidence that excessive exposure levels have been encountered; the absence of these symptoms would suggest more modest exposure levels.

Environmental history The non-occupational environment, while generally less hazardous in terms of the number and dose of chemical, physical, and biologic hazards, is paradoxically more complex to query. Often the patient has a particular focus or concern, and detailed attention can be directed to that,

for example the consequence of a leaky furnace, or the installation of a new carpet in the home. However, in every OEM referral examination at least a survey of possible environmental contaminants should be made, covering at least a question about outdoor air, environment inside the home (e.g. heat sources, chemical use) hobbies and avocations, sources of toxic pollution in the community, drinking water and diet. Occasionally, patients are referred with cryptogenic signs or symptoms, such as recurrent hives or respiratory complaint, or an elevated body burden of some heavy metal, the source of which is obscure, rendering it necessary to probe all of these aspects in some detail. The home is the source of most referrals and concerns. Wells may be contaminated with organic hydrocarbons, metals or pesticides. Exposure occurs during consumption or, in the case of volatile chemicals, from the air during bathing. Most people will be unaware of whether there are contaminants in their water, unless these have already been tested for a reason. From a history-taking perspective, the major issue is whether they use well or city water; if the former, whether there is reason to suspect contamination. Rarely the water will have an odor or abnormal appearance, but it’s not a very discriminating question. An unre-

Clinical Evaluations in OEM 35 lated water concern is old plumbing, from which lead and copper may leech after water stands overnight. Boilers, furnaces, fireplaces and stoves provide opportunities for carbon monoxide, gas, particles and fuel exhaust exposure if flues or other devices leak or are clogged. Carpets, a source of irritation when installed, harbor molds and mildew (as do ceiling and wall board) after exposure to moisture. Furnishings, computers, pets, cleaning materials or services and other merchandise may introduce chemical hazards, allergens or, occasionally, infectious risks. Radon gas occurs primarily in homes built on hard rock, but can only be evaluated by measurement – there are no other clues to be gleaned from the patient. Many, if not all of these hazards are enhanced by indoor tobacco use. Questions about outdoor sources of contamination should focus on recognized point sources, since major air pollutants such as ozone and sulfate and oxides of nitrogen are more regionally distributed: the patient is unlikely to know what levels are, though these can be obtained from regulatory authorities. Of more usual concern are local polluters, which may include neighbors using chemicals on lawns and trees that patients find noxious or disconcerting. Many people engage in hobbies or avocations at home or in their garage, involving hazardous risks. These should be specifically queried, since they may involve exposures to very dangerous materials at levels at or even above those seen in comparable occupations, usually without appropriate controls. Examples may include gardening, painting cars, sanding (leaded) paints from old furniture, building models with glues and solvents, home renovations or virtually anything which for another patient would be an occupation. Not rarely, people do at home something similar to their work, so that this line of questioning is crucial for screening all patients. Finally, it is worth asking every patient at least briefly about their diet. The most important issues revolve around imported or unusual foods, a heavy portion of the diet coming from a single source (e.g. swordfish, an important source of environmental mercury exposure) or the use of non-commercially sold dietary supplements. Cookware of distant origin may also be a source of metal or other hazards. These factors are discussed at length in Chapter 53.

Health in relation to work and environment The traditional part of the health history, including the chief complaint and review of systems, needs to be appropriately expanded to assess a possible relationship between occupational and environmental exposures and health problems. In each of the chapters in Section 3 of this text, specific questions pertinent to disorders under discussion are highlighted. But a few questions should be asked of all patients; others are specific to particular conditions. If nothing else, patients should be asked whether they feel their health problems are occupationally or environmentally related. In many instances, the first suspicion about an occupational or environmental disorder arises from the patient’s concern about the effects of an exposure.

Although this suspicion may well prove to be unfounded, such concerns should always be taken seriously. The presence of similar symptoms or complaints among coworkers or others exposed may be an important clue. Particularly for agents acting as direct toxins (see Chapters 1 and 5), the presence of symptoms among others similarly exposed may indicate that a common exposure is responsible and that the exposure levels are generally above typical ‘thresholds’. Even for substances producing sensitization, such as chemical asthmagens, finding other persons in the workplace with similar symptoms may help identify the offending agent. The report of the presence or absence of symptoms in coworkers or others in an ostensibly contaminated environment should, however, be interpreted with caution. Sometimes an exposure affects only a single person, due to unique opportunities for exposure, an idiosyncratic reaction, or differing host susceptibilities. Conversely, ‘epidemics’ may seem to be occurring where common complaints are exhibited, whose causes turn out to be quite unrelated. An important component of this part of the history, particularly for symptoms reflecting acute and recurrent conditions such as dermatitis or asthma, is the relationship between symptoms and time of work shifts or specific exposures. In the occupational setting, patients should be asked whether anything different at work preceded the onset of their symptoms, such as handling a new task, new product, or new job assignment. Patterns of symptoms in relation to time at work may provide helpful hints to both the diagnosis and the etiologic agent. Several patterns are described below; these patterns may occur alone or in combination in an individual patient. Inquiry about timing of symptoms in relation to nonworkplace environmental exposure is essentially no different from inquiry about workplace exposures, and questioning should proceed along the general lines described below.

Change in symptoms during the work day For a number of substances that induce their effect as direct acting toxins, such as solvents or non-specific dusts and respiratory irritants, the patient may arrive at work free of symptoms only to experience their onset in a predictable fashion after arriving at work. A person with building-related illness (Chapter 50), for example, may report the onset of headache and dizziness within 1 to 2 hours after arriving at work; abatement of these symptoms occurs within a few hours after leaving work. For agents causing immediate hypersensitization responses, such as flour in bakers sensitized to it, the patient often describes the onset of symptoms consistent with asthma (whether exhibited as cough, chest tightness, shortness of breath, or wheeze) within minutes of exposure. Common upper respiratory and mucosal symptoms, including coryza, eye discomfort, and itching, precede or occur concurrently. Symptoms may not occur similarly on all work days, and they may vary depending on the level of exposure (e.g., when ventilation is on or off, climatic conditions,

36 Approach to the Patient specific job responsibilities) and other host factors (e.g., extent of recent exposure, medication use).

Change in symptoms over the work week For several agents, of which cotton dust exposure is the classic example, there may be a higher level of symptom intensity on first returning to work after several days away (‘Monday morning fever’), although symptoms (and concomitant pulmonary function decline) may worsen gradually as the week passes. In some instances, symptoms may be apparent only at the beginning of the work week. In metal fume fever, for example, the symptoms of this flu-like illness are most likely to occur on a Monday or Tuesday, with the individual exhibiting loss of the tolerance acquired during the previous work week. On the other hand, workers exposed to nitrates may get headaches both in the beginning of the work week and on weekends. Weekend flare-ups are associated with a withdrawal syndrome from these potent vasodilators, with associated vasospastic coronary or cerebral events occurring most commonly on weekends.

Change in symptoms on weekends and on vacations A number of work-related syndromes result from exposures that have immediate or early effects, such that associated symptoms resolve within hours or days and occasionally with longer periods away from work. The effects of overexposure to solvents are examples of this type of temporal change. Because some solvents have longer half-lives than others, it may take days for the acute effects of solvent intoxication (characterized by headaches, lightheadedness, dyspepsia, and irritability) to resolve. For individuals who are chronically exposed to these agents, these effects may persist longer but should gradually resolve unless permanent sequelae, such as chronic encephalopathy, have supervened. Trials of removal of the individual to assess the effect of withdrawal from exposure have an important diagnostic role in OEM. Examples include: removing a worker because of exposure to potential hepatotoxins to determine whether or not several weeks or months away from work may result in resolution of dysfunction; removing a student with respiratory symptoms from a problem classroom; removing a worker with carpal tunnel syndrome from exposure to repetitive motion; and removal of the patient with contact dermatitis from consumption of suspect water. Occupationrelated trials must be conducted in close cooperation with employers, to prevent unwanted economic sequelae or misunderstanding about the reason for work absence (see Chapter 9).

Onset of symptoms away from exposure A few agents are known to have unique patterns of inducing effects in relation to time of exposure. Some occupational asthma-inducing agents, for example, cause as the most common pattern of sensitization a delayed reaction from 4 to 12 hours after exposure, often initially exhibited as nocturnal asthma. The diisocyanates (TDI, MDI, HDI)

and Western red cedar are two well-studied agents associated with this pattern, which is characteristic of exposure to low molecular weight compounds. Often, the patient presenting with new onset asthma from these causes describes first awareness of wheezing or cough on nights following days at work; over time, with the emergence of non-specific bronchial hyper-reactivity, this clear-cut association with time at work may be lost. As mentioned earlier, some agents such as nitrates may not exert their effects until their levels are lowered, so that symptomatic vasoconstriction may occur on removal of the individual from exposure. Some solvents, notably trichloroethylene, may cause a reaction similar to that of the drug Antabuse; affected individuals experience a flushing response when consuming even modest amounts of ethanol, even hours after solvent exposure has ceased.

Other experiences with work-related events The last component of the modified general health history is to inquire about previously diagnosed work-related injuries or environmental illnesses, including any experience with workers’ compensation. In addition to completing the historical database, this information may be helpful in recognizing previous hazardous exposures and the patient’s clinical and emotional response to these.

The validity of the occupational and environmental health history As noted previously, the occupational and environmental health history can be obtained by using a self-reported questionnaire or by interview. In either case, the question of the accuracy of exposure information provided by the patient is often raised, particularly if the information is likely to be used in settings that may have adversarial connotations such as workers’ compensation or litigation. Several investigators have studied the reliability and validity of this part of the health history. In the occupational and environmental history, self-reported exposure information has been evaluated in comparison with other measures, including personnel records; outcome measures, such as vital status, chest x-ray studies, and cancer registries; and information obtained by interviews with individuals knowledgeable about workplace assessments. Studies have found varying degrees of association between self-reported information and data obtained from other sources.3 Results are nonetheless reassuring when the occupational and environmental history is the main or only source of exposure data. Wherever possible, however, self-reported exposures should be supplemented by other information, discussed in more detail in the next section. The need for additional data about the nature and extent of specific exposures varies on a case by case basis. In practice, the information obtained directly from the patient often is sufficient to raise suspicion, but adequate for diagnosis only when corroborating information cannot be obtained after appropriate effort.

Diagnostic Decision Making 37

Strategies for further evaluating occupational and environmental exposure Unfortunately for the practitioner, for many reasons the history by itself is insufficient for an accurate diagnosis. These reasons include: lack of specificity about the identity of hazards; inadequate information about exposure levels; recall biases (greater attention to exposures that were at the time bothersome or otherwise perceived as being harmful); and other biases, e.g., patients fearful of possible job loss may under-report exposures, whereas litigants may exaggerate the intensities of exposures and purported effects. For these reasons, an essential component of the work-up is obtaining additional exposure information whenever possible. This additional information serves several purposes: to learn the specific identity of chemical or physical hazards to which the patient has been exposed; to establish information about dose; and to corroborate or modify the information that has been obtained directly from the patient. This includes both environmental and prior medical information that may clarify perceptions or reports of the relationship between the two. The section that follows is a summary of avenues that may be used to obtain environmental and related health information. Strategies for obtaining this information are variable, and issues about confidentiality must always be considered (see Chapter 9). A further discussion of some of these sources is provided in Chapter 4.1.

Desk Reference. In fact, for many of these MSDS, the most useful information is the telephone number listed to call for additional help. Employers or others working directly for them (e.g., physicians) may be able to provide more useful information, including the results of past workplace assessments, although other biases may be at play in such informational requests.

Health and regulatory agencies Often, a workplace or environmental hazard has been inspected by an agency with regulatory authority. The results are generally available to physicians and, if so, may be an excellent source of information. Nonetheless, one limitation is that workplace inspections are generally conducted by industrial hygiene or safety personnel for the sole purpose of ascertaining whether or not there has been adherence to various regulations; the regulations may not reflect the possible harm that can occur at levels lower than the ‘acceptable limits.’

Unions and community groups Although they are not in the same position as current or past employers, who have direct access to exposure information, many such organizations take environmental health issues seriously and have substantial amounts of information, often of good quality, that is relevant to their members or residents of a community. On the other hand, the role of adversarial relations among parties may slant or bias information from these sources (see Chapter 9).

Prior medical records

Direct site visit

These records tend to be accessible and obtainable without risk of any disclosure for the patient. Although these records may not further etiologic assessment, they can confirm the patient’s complaints on previous occasions and provide objective measures of his or her previous physiologic status. This information can help corroborate or modify the history and may be useful in applying one of the cardinal principles of occupational and environmental diseases: the biologic plausibility inherent in the time between exposure and effects (see below).

When the issue is a current or recent exposure, there is perhaps no more satisfactory way to evaluate the environment than onsite inspection. This practice offers the advantages of contacting employees at the site, relating the history to observable facts, and directly assessing exposure and dose. The opportunity to correlate illness with the work environment is one of the special advantages of clinicians who are based at the workplace. Lack of this capability puts the diagnostician at a considerable disadvantage. Whether he or she is based inside or outside the plant, however, the clinician must recognize that the direct assessment of exposures is a highly complex and specialized process, requiring assistance of an industrial hygienist or a comparably trained professional (see Chapter 4); not every relevant exposure possibility may be evident to the untrained eye.

Exposure records from an employer Under current regulations in the United States and most other developed countries, employers are obligated to maintain material safety data sheets (MSDS) for each potentially hazardous material with which employees may come in contact. The employer must make this information available to the employee or his or her physician in a timely fashion, together with any available information about exposure doses (e.g., air sampling information, blood tests). Despite this, there remain numerous problems. The MSDS themselves are often of limited quality. Much potentially useful information is lacking, such as information on minor ingredients that may be responsible for important health effects, especially allergic ones. In addition, the health information is often presented uncritically and without adequate discussion, much as the way adverse effects are uncritically listed in the Physician’s

DIAGNOSTIC DECISION MAKING Specialized use of the laboratory Clinicians without experience in OEM diagnosis and practice may imagine that the laboratory could be used to compensate for the difficulties in obtaining reliable information about exposure or putative effects. This is not surprising, given the remarkable progress of clinical toxicologists in quantifying xenobiotics and the burgeoning array of technical capabilities that now allow measurement

38 Approach to the Patient of many contaminants down to the level of parts per trillion in numerous body tissues. Unfortunately, at this time, the overall role of the laboratory in OEM remains limited. In this section, the role of the laboratory in tertiary OEM practice is described. Each of the organ system (Section 3) and exposure chapters (Section 4) in the text further details the role of the laboratory in the assessment of particular hazards and diseases. Despite some overlap, laboratory tests can be conceptualized in one of three ways.

1. Tests that elucidate pathophysiology These tests include almost all of the routine ‘medical’ tests, such as imaging studies, chemistry panels, and hemograms. Also included are tests that play more specific roles in OEM diagnosis, such as non-specific inhalational challenge tests (e.g., methacholine challenge test) or measurement of enzyme levels (e.g., cholinesterase or aminolevulinic acid dehydratase [ALA] test). The net effect of this kind of study is to clarify what is or is not wrong pathophysiologically with the patient. Some such studies, of course, may be invaluable for assessing causality, but the primary role is to evaluate effects, not exposures or causes.

2. Tests that elucidate or quantify exposures Other tests may be performed to establish the presence of a suspected causal agent in an organ or body tissue. Such tests are often referred to as biologic monitoring because, in effect, they use the body as a sampling device to assess exposure. Examples include measurement of the whole blood lead level, which documents that lead at a given concentration is present in red cells, or polarizing light microscopy on a lung biopsy to quantify crystalline silica particles. These kinds of studies may lead to inferences about health effects, but they do not measure them, only exposures. Although there may be some rationale in using a measure of exposure as a surrogate for identifying ‘cases’ of disease (e.g., identifying an adult with a lead level over 40 g/dL as ‘lead poisoned’) for surveillance purposes, the identification of the toxin is not tantamount to demonstration of a causal health effect for clinical use.

3. Tests that directly assess the relationship between an exposure and an effect A few tests are ‘dynamic’ in the sense that they inherently capture causal information. An example is determining the presence of a specific antibody to a sensitizing agent. The presence of the antibody in such a case confirms both that exposure has occurred and that it has generated an immunologic reaction (which may or may not prove to be related to a presenting symptom or sign!). Similarly, patch testing and specific inhalational challenge tests are types of tests that document that an exposure and sensitization have occurred and may even document the relation between the level of exposure and a specific health effect. At times, a test of one of these three types may be used appropriately to establish information about another type. For example, zinc protoporphyrin (ZPP) in red cells, a

measure of lead effect in the blockade of the enzyme heme synthetase, is a test commonly used as a surrogate measure of lead exposure because of its dose–response characteristics. Similarly, while measurement of urine cadmium level is a good index of recent absorption in workers exposed to the metal, in another setting – long after exposure has ceased – it may serve as a good measure of the renal effect of cadmium. In individuals who have not had recent cadmium exposure but who have suffered impaired renal function, the affected kidney leaks stored cadmium, whereas the intact kidney will not. In this case, an apparent measure of an exposure is used to measure an effect. Even when the logic for ordering a laboratory test is clear to the clinician, there remain several problems with the application of results to diagnosis.

1. Limitations inherent in the laboratory itself The clinician must always be alert to the factors that limit the quality of any data that are returned from the laboratory. These include (1) the ability of the laboratory to detect a substance or an effect, (2) the reliability of the laboratory at measuring such substances, (3) the validity of results, (4) the precision of the results, and (5) the standardization of the results. Notably, for example, the indiscriminant use of hair analysis to detect heavy metals has undergone scrutiny; in all but a small number of research labs, hair analysis has been found wanting in almost every one of these dimensions.4

2. Strategies for obtaining tests Although many tests in clinical medicine do not vary significantly according to the time they are obtained (e.g., chest x-ray studies), others can be interpreted only on the basis of careful planning of how and when samples are obtained (e.g., plasma triglycerides). In many clinical situations in OEM, the strategy for sampling is crucial to interpretation of results. For tests of effects, the timing must be planned to avoid either missing the effect or confounding one effect with another. An example of the first situation is the use of spirometry to detect bronchospasm; even an individual with severe occupational asthma may have normal spirometry if the test is not timed to coincide with an expected effect. On the other hand, an audiogram performed within several hours of noise exposure is likely to document the temporary effects of noise but is incapable of establishing baseline hearing function. Failure to consider testing strategy may also limit the interpretation of tests that directly assess the relationship between exposure and effect. For example, it is now well recognized that early in the course of occupational asthma, sensitization to agents such as isocyanates may rapidly reverse after removal of the individual from exposure. Therefore, even in a previously sensitized individual, a specific challenge with the agent may fail to produce bronchospasm if that person has been away from the exposure for a period of time before testing.

Diagnostic Decision Making 39

3. Interpretation of normal and abnormal results Laboratories commonly supplement reports with statements as to whether or not a result is ‘normal’. By convention, the term ‘normal’ usually means that a test result falls within the range of results for 95% of the healthy population. For a few tests, other guidelines are used by convention, such as within 20% of the mean result of a reference group. In OEM, unlike general medical practice, laboratory test results need to be placed in their appropriate context. For example, a young worker exposed to a known respiratory tract toxin may have ‘normal’ lung function on spirometry, yet comparison with a previous value from the same individual documents that loss in function has occurred. The laboratory’s interpretation in this case is falsely reassuring. Conversely, a lead battery worker may be identified by the laboratory as having an ‘abnormal’ whole blood lead level of 20 μg/dL. Although this level is indeed higher than that found in any general population, it is unlikely to reflect significant lead toxicity, nor is it an unexpected value in a well-controlled battery plant. Recognition that this level of exposure is higher than that seen in other adults may have some utility, but cannot be used as the basis for the diagnosis of lead poisoning. Further, as with all tests, the likelihood that an individual with an abnormal test in fact has the disease under evaluation (positive predictive value) is influenced not only by the test’s sensitivity (that those with disease will test positive) and specificity (that those without disease will test negative) but also by the prevalence of the disease in the specific population from which the individual comes. For example, a minimal finding of interstitial fibrosis on a chest radiograph in an asbestos-exposed worker is, in itself, more predictive of disease than is the same finding in someone who has not been exposed to a fibrogenic material (see further, below).

The available databases for diagnostic inference Having completed the basic evaluation, including the history, environmental exposure assessment, physical examination, and basic laboratory evaluation, the information needs to be integrated to answer three central questions. 1. Given everything already known about the patient, is it plausible that he or she has a disease or health effect related to environmental exposure? 2. If it is plausible, how likely is it, based on the exposure assessment and clinical pattern? 3. Given the exposure assessment and clinical setting, how should various laboratory tests be interpreted? What further studies or tests offer the possibility of substantially altering the likelihood that any clinical abnormality(ies) is (are) related to the environmental exposure? In this section, we address the task of identifying existing sources of information and establishing how these sources can be used to assist in arriving at a working diagnosis.

Many different types of resources are available to address questions about disease plausibility and likelihood. Most, but not all, rely on a review and assessment of the scientific literature related to the exposures and the health effects in question. The most important are as follows.

1. Exposure assessment databases For a variety of reasons, it is often not possible to confirm directly the occurrence of an exposure of interest or to obtain a reliable estimate of dose. Fortunately, there are many available resources for translating historic information into at least semi-quantitative estimates about exposure and likely dose. Texts, including this one, often summarize the hazards associated with particular kinds of activities and the range of doses to which such workers are potentially exposed. Large surveys, such as the ones performed periodically by the National Institute for Occupational Safety and Health (NIOSH), allow determination of the jobs and industries where certain hazards are likely to be found.5 Scientific papers, often from the industrial hygiene literature, can frequently provide valuable summaries that may be relevant to a particular clinical problem. Such papers are easily searched using National Library of Medicine and related databases such as Medline or Toxline.

2. Epidemiologic databases When there is some basis for estimating exposure and dose, epidemiologic studies often provide the most compelling data relating exposure to risk for disease Epidemiologic studies can provide direct evidence that an exposure causes an effect in humans. Moreover, such studies often establish certain limits for biologic plausibility of an association between exposure and disease, such as the timing between exposure and emergence of increased disease risk. These kinds of data are often helpful in the quantitative assessment of individual risk. If individuals who do precisely the same activity have been studied, quantitative determination of risk is relatively straightforward. Even if the patient’s exposure setting differs from groups that have been reported, one can often learn enough about the patient’s exposure to fit him or her into the range of exposures that have been studied. Access to epidemiologic studies is readily achieved by using available texts and computerized literature searches. Unfortunately, occupational epidemiology studies vary dramatically in quality and applicability to the patient under evaluation. Issues related to the quality of epidemiologic studies are discussed in Chapter 6. The relevance of epidemiologic studies to the patient at hand is no less important. First, the exposure dose in the patient should be reasonably similar to the exposure dose of at least a portion of the study population. Study findings can be extrapolated to results at higher and lower doses, but this approach is not always valid biologically and requires complex scientific judgements (see Chapter 5). Second, the population under study ideally should resemble the patient’s demographic characteristics, such as age and gender. Unfortunately, most studies historically

40 Approach to the Patient have been limited to white males, so there is often little choice other than to use the data on white males and modify the interpretation as needed. Whether this is a meaningful limitation depends on the particular health effect and the likelihood that host susceptibility plays an important modifying role.

3. Toxicologic databases Epidemiologic data are often inadequate alone for determining the likelihood that a patient’s health problem can be related to an environmental exposure. Alone or in combination with epidemiologic data, the results of animal studies performed under experimental conditions may be helpful. One advantage of these studies is that they often provide very strong evidence of dose-related effects of hazards. Further, animal studies may provide information about certain laboratory findings, such as the presence of toxins in diseased organs, biochemical changes, and histopathology. Modern toxicologic studies of this kind are plentiful and are easily identified by computer searches of the scientific literature. Good summaries are often available as well in texts and in monographs such as the toxicologic profiles compiled by the United States Agency for Toxic Substances and Disease Registries located in Atlanta.6 Serious limitations of toxicologic studies include differences between species and routes of exposure. For example, gavage (tube feeding) differs greatly from the routes by which patients typically have been exposed. In addition, animal studies almost invariably depend on use of relatively high doses of single toxins, and invariably require extrapolation to be applicable to clinical situation. As with epidemiologic studies, such extrapolation sometimes requires complex scientific judgment. Additional limitations to the use of toxicologic studies for clinical purposes are identified in Chapter 5.

4. Clinical studies and case reports Clinical reports and case studies, although of limited value for establishing cause and effect per se, often are exceptionally useful for clinicians in OEM. Compared with epidemiologic studies, very rich information is often provided regarding the actual characteristics of the patients being reported and the specific nature of their exposure and dose. Specific clinical information of importance may include the results of a wide array of tests and descriptions of the patterns of illness, clinical course, and responses to treatment. At their best, clinical case reports and studies may be sufficiently applicable to provide a rational basis for all subsequent steps in diagnosis and management. At their worst, clinical reports may create the illusion of a causal relationship between an exposure and an outcome that cannot be substantiated. Case reports relevant to an exposure, therefore, should be carefully reviewed, but strong weight should be given to them only if corroborating evidence (e.g., toxicologic or epidemiologic) is also available.

5. Clinical experience Although direct clinical experience may create certain biases of perception because exposed individuals with prob-

lems are far more likely to have been seen than those without problems, relevant experience should be drawn upon whenever feasible. Specifically, when possible, physicians should consult those who may have previous experiences with patients like their own, such as physicians practicing in workplaces where relevant exposures are commonplace, or those caring for such workers. Such databases, while not ‘scientific’ in the usual sense of the term, have obvious relevance.

Quantitative clinical reasoning Once the patient’s health status has been evaluated, and the exposure history obtained and supplemented to the extent feasible, the task remains to establish a final diagnosis. In particular, it is necessary for the OEM practitioner to address the specific question regarding the suspect occupational and environmental cause. In most cases, despite every effort, this determination requires dealing with at least some uncertainty. Not only will the diagnosis often fail to be unequivocally established by information at hand, in many instances there is no single further test or source of information that could provide certainty even were patient safety, cost or practicality not at issue. Going back to the principles outlined in Chapter 1, for many situations there are no discriminatory tests to resolve, for example, whether a bladder cancer was or was not caused by exposure to benzidine dye, or asthma by exposures in an aluminum potroom. For other situations, extensive quantitative data regarding former exposure might theoretically resolve the matter, or a test might in theory exist to assist, but the information or test cannot be obtained. Unfortunately, though in many domains of medicine such specific answers are not required – physicians treat such disease all the time without establishing their causes – this luxury does not extent to OEM where preventive and therapeutic efforts, as well as appropriate assignment of medicolegal benefits, require such a diagnosis. In this section we discuss the use of Bayesian reasoning to reach a diagnosis in the face of uncertainty. A fuller discussion of this subject can be found in various texts.7,8 Whether they do it explicitly or not, all clinicians rely on Bayesian reasoning in everyday practice, when deciding that a fever in a young adult is most likely mono, or a rash in an elderly patient most likely a reaction to a medication – and which medication is the most likely culprit. The theory, based on Bayes theorem, is straightforward: the probability of any diagnosis being correct is the mathematical product of two probabilities. The first probability is the ‘risk’ for the patient (given his/her age, race, gender, background, the season, etc) getting such a disorder, sight unseen. This is referred to as the ‘prior’ probability, and refers to the frequency with which such a diagnosis occurs in the population from which the patient comes. For example, the prior risk of an adolescent getting mono in a given year is very high – perhaps as high as 1 per 100 – based on incidence in that age group. Likewise, the risk of an elderly patient on antibiotics getting a skin

Diagnostic Decision Making 41 rash is also high, depending on the antibiotic. Contrarily, the risk of a young patient in otherwise good health getting tuberculosis in the US is very low – perhaps 1 per 100,000 – compared to the same person in South Africa, where the background rate is several orders of magnitude higher. Obviously, some knowledge of disease rates in the appropriate population is necessary for estimating this prior probability, but even in the absence of such detail, almost all practitioners should have enough experience to know what is common and what is rare. More detailed information comes from medical texts, periodicals like MMWR and descriptive epidemiologic studies. In occupational and environmental medicine practice, the sources may be the same, or knowledge is available from some of the sources of exposure information described earlier in this chapter. The second factor in the calculation of the probability of each diagnostic consideration is the probability that some major finding – typically a physical sign or laboratory anomaly – would be expected in patients with that diagnosis. For example, 80–90% of all patients with mono have persistent fever. On the other hand, less than 1 in 10 with mono will be expected to be negative on the mono-spot test. Determining these ‘posterior’ probabilities depends on performing the most appropriate exams and/or tests for the major diagnostic possibilities dictated by knowledge of the patient’s (prior) risks. The sources for this information, at least for diseases of OEM concern, are discussed in Section 3 of this text in relation to each of the major disorders. Once these two probabilities are estimated for each major diagnostic possibility, comparison should yield either a clear ‘winner’ – the most likely – or leave a small number, usually two, about equally possible. In the former instance, the work-up is essentially complete, unless one of the more remote possibilities has health consequences that mandate exclusion. For example, even if metastatic cancer is a very unlikely possibility for a young coal miner with diffuse nodules on chest x-ray (unless he/she had a melanoma removed the year before!), consideration of cancer might dictate additional tests to further exclude this possibility if only because of the harm to the patient of overlooking it. More important is the use of additional tests when there remain competing probabilities, as the following case examples illustrate. The first example is a 60-year-old African-American male, a foundry worker who has been a ‘chipper-grinder’ – a job entailing the grinding of sand off metal castings – for 35 years. He complains of cough and shortness of breath and has a symmetric reticulo-nodular infiltrate on his x-ray (Fig. 3.3). The ILO grade of the infiltrate is 2/2 (see Chapter 19.1 for a discussion of the ILO rating system for x-rays). The complaint and the x-ray are compatible with silicosis, but also with sarcoidosis, chronic beryllium disease, disseminated tuberculosis and other rarer diseases. Using the approach suggested above, it is first important to learn roughly how likely a chipper-grinder of this era (1968–2002) would be to get silicosis, i.e., the prior probability of that diagnosis.

Figure 3.3: Abnormal chest x-ray for both hypothetical patients. The x-ray shows symmetric, bilateral reticulo-nodular opacities in the mid and upper lung zones, rated 2/2 by a B-reader using the ILO scale (see Chapter 19.1).

Based on review of the literature or, better yet, a call to the company or physicians caring for workers there, you learn that as many as 1 in 5 long-term workers have developed this common outcome of daily, historically uncontrolled exposure. Next, reviewing Chapter 19.9 of this text, you learn that his x-ray is typical of the disease, present in almost all cases, let’s say 90% for the calculation purposes. Applying Bayes theorem, his likelihood of having silicosis is proportional to: 0.2 (prior) × 0.9 (posterior) = 0.18. The absolute number is unimportant, only its value compared to other possibilities. Compare it, for example, with his risk for sarcoid, the next most likely consideration from a clinical perspective. Sarcoid has a prevalence in this patient’s ethnic group of about 1 per 1000. Even assuming the x-ray is seen in all such cases, the likelihood of sarcoid would be proportional to: 0.001 (prior) × 1.0 (posterior) = 0.001. Crudely estimating, silicosis is 180 times more likely than sarcoid in this case. For all of the other possibilities the prior would most likely be lower still, even though tuberculosis occurs more frequently in silica-exposed men than in the rest of the population. Although it makes sense to exclude tuberculosis because of the gravity if this were missed, the overwhelming likelihood is that this man has silicosis. Unless his illness takes an unexpected turn, no biopsy or further tests are easily justified, especially not for medicolegal reasons. Now consider a second patient with the same complaint and the same x-ray from the same foundry. This second man, however, is only 35, and has worked in this environment only since 1995. For him, the prior probability of silicosis, by comparison to the previous case, is very low –

42 Approach to the Patient he has worked fewer years, all in the more modern period since environmental controls have been instituted (you learn using the approach above) and in a less exposed job category (other facts you could confirm). With silicosis far less likely a priori, his likelihood of having sarcoid, or even one of the rarer causes of this x-ray picture (Fig. 3.3) become greater by comparison. So likely, in fact, that additional tests, up to and including lung biopsy, would be justified to reach the correct answer, whereas in the first case the risk of such a test may exceed the risk of initially missing an alternative diagnosis (as might occur very rarely, of course). Not every case is as clear as these, nor in every case is there an obvious test available to break ‘ties’. Consider, for example, the elderly painter with heavy solvent exposure who develops depression, or mild dementia. There are numerous possibilities and no trivial test to distinguish chronic solvent intoxication. Another concern in many cases is the possibility of no disease at all. This must often be factored in as a possibility, when signs or laboratory findings are minimally deranged (not really a possibility in the previous examples). In such a case one calculates the probability of no disease in much the same way as a particular diagnosis, e.g., what is the (prior) probability that a 60-year-old foundryman is free of any respiratory disease? Having said that, and with full appreciation of the difficulty of providing estimates of many prior probabilities with great confidence, the overall approach is itself very robust, and OEM practitioners, whether in training or highly experienced, are well advised to make such estimates as explicitly as information allows, the more so the more complex and uncertain the case may be.

PATIENT MANAGEMENT There are four conceptual stages in the management of patients suspected of occupational or environmental disease. Although actual treatment choices depend on the particular hazardous exposure, clinical disorder, and relevant social issues, the conceptual framework remains uniform.

The diagnostic period In every case, there is a period of time from the first contact with the patient until the best possible working diagnosis can be achieved. Although this period may be as short as a single visit in cases of clear-cut exposure and a well-delineated physiologic response (or its absence), often this period spans days to weeks and occasionally even longer, during which relevant records of medical and exposure history are obtained, environments evaluated, and additional tests performed. Several important aspects of management should be kept in mind during this period. 1. Formulate the diagnostic plan at the outset. For reasons already given, it must be anticipated that many potentially knowable facts may remain unlearned. Therefore, at the outset, it is important to decide how the diagnostic process will proceed,

whether or not one succeeds in obtaining desired information. This is another way of saying that the patient cannot be held hostage to impediments in getting access to information. 2. Use the diagnostic period to the fullest. An advantage of small delays in achieving best clinical diagnosis is that it facilitates ascertainment of various social issues, such as the agendas of all parties and likelihood of reactions if certain choices are proposed (see Chapter 9). 3. Do not initiate management decisions until a sound working diagnosis is achieved. Although a few occupational and environmental illnesses are true medical emergencies, the vast majority are not. In fact, many diagnostic issues, such as the temporal relationship between physiologic responses and exposures or the measurement of biologic exposure dose, are best studied while a patient remains in an exposed situation. Further, given the ramifications of certain courses of action, such as removal of a worker from a job or the designation of a particular health problem as occupational, it is important not to take such steps prematurely when there is a reasonable likelihood that subsequent facts may alter diagnostic thinking. Although further exposure may seem on the surface to place a patient at unnecessary additional risk, it is important to place that risk in the context of exposures that have already occurred and the very real risk of taking mis-steps that may be harmful in themselves, especially if they are premature and wrong. 4. In certain circumstances, diagnosis itself depends on the individual’s response to a therapeutic trial, such as whether or not the symptoms abate on removal of the individual from an environment. Although such trials are sometimes necessary or desirable, they must be conducted as trials, with the explicit understanding that the goal is to establish a diagnosis and not misconstrued as treatment based on a putative or tentative diagnosis. Only in this way can future confusion be avoided and, with it, the attendant social costs to the patient.

Formulation of treatment plans Once the diagnosis has become sufficiently clear, treatment plans can be developed. Sometimes the choice is strongly dictated by clinical circumstances, as in cases of disabling chronic diseases. More often, however, there are alternatives. Although one choice or another may seem preferable from a strictly medical perspective, it cannot be presumed that this choice satisfies the needs of the patient, or that it is included among the options offered by the employer or other relevant circumstances. Whenever feasible and safe, it is ideal to consider alternative possibilities for the management of every case. On other occasions, though, medical realities dictate clearly one course of action, such as discontinuing exposure of a patient with heavy metal or pesticide poisoning when the documentation is clear-cut.

Patient Management 43

Establishing and communicating the therapeutic plan However limited or diverse the options considered, the actual choice of a treatment plan cannot be made without discussion with all parties who must participate, first and most importantly the patient. The benefits, costs, and health risks associated with each option must be identified so the patient can make an intelligent decision based on all the facts, including social factors (see Chapter 9). If necessary, he or she should be encouraged to include a spouse or other family member who may be affected by the choice. Advice from other advocates, such as a social worker, union representative, or advisor, may be invaluable – especially when one choice or another may disrupt the individual’s normal life activities, work, or income. Once the patient has agreed to a plan, it is reasonable to communicate that plan to the other parties who must cooperate, especially in the occupational setting. In every case, the patient should be aware of the planned communication and it should be limited to issues that are within the purview of the doctor’s relationship with non-medical parties. Once an approach has been selected, it is important that results of the evaluation be put in writing for the patient, his or her personal physician, and others who need to have this information. Dissemination of this information can often be best accomplished by a single letter, which is sent to everyone. Although this method may limit communication of certain private issues and require use of language not technically ideal for every reader, the use of a single letter reduces the likelihood of anyone misunderstanding what has been found and recommended.

Therapeutic follow-up Whatever plan is chosen, clinical and non-clinical consequences that arise during the initiation of the plan may lead to reconsideration by patient, physician, or both. For example, the impact of altered life circumstances, such as joblessness, or homelessness in environmental cases, may provoke re-evaluation of options, however strongly they are indicated from a clinical perspective. By planning visits for no grander purpose than a progress update and plan re-evaluation, the physician offers the most flexible and supportive milieu in which the patient can accommodate to changes demanded in the management of occupational and environmental health problems.

Where does the physician’s responsibility end? One of the most inescapable realities of modern OEM practice in most parts of the world, including the United States, is that many hazardous situations will be recognized about

which little can be done. Reasons include: limitations of resources available in the professional community, corporate sector, and governmental agencies; technical uncertainty about the solution; economic incentives to ignore or resist recognition of the problem or solution; inadequate regulatory, technical and administrative structures to resolve the matter; and last, but not least, widespread ignorance. None of these factors should form the basis for physician apathy or reluctance to consider and act on the public health implications of a serious occupational or environmental disease or injury. At a minimum, the record should clearly indicate the physician’s opinion that others may be at risk. The choice of strategy should be well documented in the patient’s record. Finally, although various approaches may best be carried out verbally or informally, in the end it is important that the individuals who control the environment in question be formally notified that concern has been raised. This notification may come directly from the physician or, more often, indirectly through a government agency or their consultant. The physician’s effort should be documented in writing, which is far more effective in motivating action than informal approaches. Once these steps have been taken, the practitioner can and should return to the care and needs of the patient, whose OEM problems and their clinical and social sequelae may take months, often years, to resolve. This is the same timetable that public intervention may follow. Clinicians may choose early in the process of public health actions to turn responsibility over to others. Regardless, the often delayed but sometimes highly effective link between clinical evaluation and resultant greater public health good is one of the great satisfactions of OEM practice.

References 1.

2.

3.

4.

5. 6. 7. 8.

Frank AL. Occupational and environmental medicine: approach to the patient with an occupational or environmental illness. Primary Care Clin Office Pract 2000; 27:877–94. Lax MB, Grant WD, Manetti FA, Klein R. Recognizing occupational disease – taking an effective occupational history. Am Fam Phys 1998; 58:935–44. Rosenstock L, Logerfo J, Heyer N, Carter W. Development and validation of a self-administered Occupational Health History Questionnaire. J Occup Med 1984; 26(1):50–54. ATSDR. Summary Report on Hair Analysis Panel Discussion, June 12–13, 2001. The website is: www.atsdr.cdc.gov/HAC/hair_analysis/index.html. NIOSH. National Occupational Exposure Survey. The website is: www.cdc.gov/niosh/89-103.html. ATSDR toxicologic profiles may be accessed at: www.atsdr.cdc.gov/toxfaq.html. Albert DA. Reasoning in medicine: an introduction to clinical inference. Baltimore: Johns Hopkins Press, 1988. Cutler P. Problem solving in clinical medicine: from data to diagnosis. Baltimore: Williams and Wilkins, 1998.

Chapter 4 Principles of Industrial Hygiene 4.1 Occupational Hygiene Robert F Herrick, John M Dement Occupational (industrial) hygiene is the health profession dedicated to the anticipation, recognition, evaluation and control of hazards in the workplace environment. The scope of interest includes chemical, physical and biologic hazards as well as ergonomic and human factors, that cause or contribute to impaired function, disease, disability, injury and discomfort resulting from work. As the profession grew up with the Industrial Revolution, it has been known as industrial hygiene, but the term occupational hygiene more accurately describes the field and the practitioner’s range of activities. In fact, the term occupational hygiene is prevalent everywhere except in the United States. The term industrial/occupational hygiene is derived from the Greek Hygieia, the goddess of health and prevention, daughter of Asklepiose, god of medicine. Its roots trace back to Bernardino Ramazzini (1633–1714), considered the father of occupational medicine. The modern history of occupational hygiene starts with the industrialization of the United States and western Europe. This process was chronicled by Theodore Hatch, who summarized the ‘Major Accomplishments in Occupational Health in the Past Fifty Years’ on the 50th anniversary of the Division of Occupational Health of the Public Health Service in 1964.1 Hatch recounted that prior to the First World War (about 1914), societies were predominantly rural and based upon agriculture. Industrial processes were few, and conducted by manual labor. The only plastic available was celluloid, petroleum refining discarded most product as waste, and Henry Ford had just introduced the radical concept of a $5 daily wage. This was the industrial world Alice Hamilton found when she began to trace health problems among immigrant families back to the workplace. Pioneers like Hamilton and Hatch identified important problems, but they also had the vision to develop interdisciplinary approaches to solve them. Occupational hygienists share responsibility with physicians and other occupational health practitioners and researchers in the identification of adverse health effects associated with the workplace environment. The occupational hygienist focuses upon the factors that are potential causes of work-related conditions, providing information on workplace processes, and exposures to physical, chemical and biologic agents that result from those workplace processes and conditions. The evaluation of those hazards frequently includes measurements to identify and quantitate contaminants in the workplace, as well as measurements of physical factors such as noise, radiation, heat and ergonomic conditions. In the practice of

occupational hygiene, the evaluation of hazards leads to the selection and application of an appropriate exposure control strategy. These strategies include engineering controls, improvements in work practices and materials, and personal protective equipment to reduce workplace risks. In addition to the occupational hygiene practice of hazard recognition, evaluation and control, research on exposures as potential causal factors for occupational disease advances the goal of promoting worker health and safety. Exposure assessment in epidemiologic research and hazard evaluation is a vital role for occupational hygienists. These assessments frequently involve current and past exposures, so occupational hygienists apply familiarity with workplace processes, controls, and exposure information in conjunction with the work histories of employees, to reconstruct exposures in retrospective epidemiologic studies and risk assessments. Occupational hygienists integrate information and knowledge from disciplines including engineering, chemistry, physical science, toxicology and medicine. While occupational hygiene practitioners are usually trained in one of these disciplines, the majority have graduate degrees in occupational (industrial) hygiene, environmental health, or an allied field. There are programs for professional certification of occupational (industrial) hygienists in several countries, requiring demonstration of proficiency in the following technical areas: basic science; biohazards; biostatistics and epidemiology; engineering and other controls; ergonomics; ethics and management; analytical chemistry; sampling, monitoring and instrumentation; noise and vibration; ionizing radiation; nonionizing radiation; regulations, standards and guidelines; thermal and pressure stressors; toxicology; and general topics including community exposures, hazardous wastes, risk communication, indoor environmental quality, and others (unit operations, process safety, and confined spaces). Certification is achieved through a combination of work experience and a comprehensive written examination. In the United States, approximately 6500 industrial hygienists are certified by the American Board of Industrial Hygiene.

OCCUPATIONAL HYGIENE AND DISEASE PREVENTION Prevention of environmental diseases may be thought of as a two-stage process involving primary prevention

46 Occupational Hygiene and secondary prevention.2 The ultimate objectives are: (1) to avoid the establishment of disease, (2) to reduce the likelihood of disease recurrence or progression and (3) ameliorate the morbidity associated with the disease. Prevention of environmental diseases involves hazard recognition, hazard evaluation, and hazard control/ intervention. Hazard recognition: The hazard associated with a given exposure is a function of both the toxicity of a material and the extent of human exposure. Surveillance of both exposure and disease provides clues and hypotheses for further evaluation. Health data may be generated through environmental/occupational medicine and surveillance programs or through epidemiological studies. Toxicology often provides valuable information with regard to hazard recognition. Hazard evaluation: Prevention strategies require knowledge of the effects caused by exposures as well as the dose levels where effects occur. These data allow development of risk assessment and strategies to reduce or eliminate significant human exposures. Toxicology, occupational medicine, and epidemiology provide the means for identifying chemical, physical, or biologic hazards. Toxicology testing in animals is an important component of early hazard recognition as well as hazard evaluation. Hazard control/intervention: Primary prevention involves identification of environmental hazards which are factors or co-factors in disease development, followed by application of methods to reduce or eliminate human exposures. This represents the classical public health approach. Figure 4.1.1 depicts the components of the pathway from the source of a contaminant in the environment, through exposure, dose, and adverse health outcomes. The opportunities for intervention at various steps along the pathway are also shown. Principles and methods for controlling occupational hazards are discussed more fully later in this chapter.

• Substitute less hazardous materials • Enclose/isolate source • Modify the process • Local ventilation to contain emissions

Ambient concentration

Source

RECOGNITION OF OCCUPATIONAL AND ENVIRONMENTAL HAZARDS Classification of hazards Workers may be exposed to contaminants by inhalation, absorption through the skin, ingestion, or injections (e.g., through accidental puncture wounds). Inhalation and skin absorption represent the predominant routes of exposure for most materials in the occupational environment. Ingestion may be an important source of exposure where poor hygiene practices, such as consumption of food and beverages in a contaminated work area, is allowed. Workers in healthcare facilities often are exposed to infectious agents through punctures with contaminated needles. Mucous membrane exposures to infectious agents from blood and body fluids may also be an important route of exposure in healthcare facilities. Environmental agents are broadly classified by the Occupational Safety and Health Administration (OSHA) in the Hazard Communication Standard.3 The following is a summary and elaboration of hazards listed in the OSHA standard.

Physical hazards Materials such as explosives, flammable or combustible liquids, oxidizers, compressed gases, organic peroxides, pyrophoric materials, unstable (reactive) chemicals, or water reactive chemicals are regarded as physical hazards by OSHA. Other exposures in the workplace, such as excessive noise, ionizing and non-ionizing radiation, and temperature extremes are further examples of physical hazards. Ergonomic hazards include repetitive and forceful movements, vibration, temperature extremes, and awkward postures that arise from improper work methods and improperly designed workstations, tools and equipment.

• General ventilation to remove contaminants • Administrative/work practice controls

Exposure

Personal protective devices (gloves, respirators)

Molecular/ clinical outcomes

Dose

Figure 4.1.1: Hazard recognition, evaluation, control.

Recognition of Occupational and Environmental Hazards 47

Biologic hazards include bacteria, viruses, insects, plants, birds, animals and humans. Most biologic health hazards can be classified as infectious or immunologically active. As an example, accidental injection of blood-borne viruses is the major hazard of needlestick injuries, especially the viruses that cause AIDS (the HIV virus), hepatitis B, and hepatitis C. Anthrax is an example of a bacterium (Bacillus anthracis) that can affect the skin, the lungs, as well as the mouth, throat, and gastrointestinal tract. The infection sometimes can spread to other parts of the body, especially if treatment is not started early. The anthrax bacteria can form spores under certain conditions when, for example, body fluids infected with the bacteria are exposed to the air. The bacteria cannot live for long outside an animal; however, the spores can survive in soil and some other materials for decades.

trial lighting conditions, may represent a serious overexposure condition for substances such as asbestos and silica. Air-borne dusts show wide variability in particle shape. Figure 4.1.3 shows typical dusts from a rubber-processing operation using industrial talc. The particles vary from flat to rounded and compact. In comparison, Figure 4.1.4 shows a microscopic photograph of dusts from an industrial operation generating talc contaminated with asbestiform minerals. The photograph demonstrates greatly elongated asbestiform fibers, similar in appearance to fibers seen with commercial asbestos minerals. Fumes: Fumes are formed when the material from a volatilized (evaporated or vaporized) solid condenses in cool air. The solid particles that are formed make up a fume that is extremely small, usually less than 1 μm in diameter. In most cases, the freshly generated fume reacts with the oxygen in the air to form an oxide. Welding, metalizing, and other operations involving heating of metals to high temperatures produce vapors from the molten metal which produce fumes. Arc welding volatilizes metal to form a vapor that condenses, usually as the metal or its oxide. Fumes, because they are extremely fine, are readily inhaled. Smoke: Smoke is usually produced by the incomplete burning of carbonaceous materials such as coal and oil. The resulting aerosol consists of carbon or soot particles less than 0.1 μm in size. Smoke, such as tobacco smoke, generally contains droplets as well as dry particles. Mists: Mists are finely divided liquid droplets which are air-borne. Mists may be generated by condensation of liquids from the vapor back to the liquid state, or by breaking up a liquid into a dispersed state, such as by splashing, foaming, or atomizing. In industrial operations, mists are produced during paint spraying, spray application of pesticides and herbicides, and during cutting and grinding operations. Acid mists are produced during metallurgical pickling operations and during electroplating.

Types of air-borne contaminants

Particle respiratory deposition

Aerosols: Aerosols are composed of liquid droplets or solid particles fine enough to remain dispersed in air for a prolonged period of time. Aerosols may also be referred to as air-borne particulate matter with a wide range of particle size. Typical size ranges for aerosols are shown in Figure 4.1.2. Dusts: Dusts are solid particles suspended in a gaseous medium. Dusts result from the mechanical disintegration of materials, such as grinding, with enough mechanical energy to propel particles into the air. Air-borne dust particles vary widely in size from approximately 50 μm to less than 1 μm. Only the larger particles may be seen without the use of a microscope. Most dusts produced from industrial operations as well as non-industrial operations such as construction or demolition consist of particles that vary widely in size, with the small particles greatly outnumbering the large ones. In general, when visible dust is noticeable in the air near a dust-producing operation, exposures to large numbers of smaller particles can be anticipated. The presence of a visible dust cloud, under typical indus-

The hazard associated with air-borne particulate matter is a function of: (1) the biologic activity of the material, (2) concentration of the air-borne material, and (3) air-borne particle size. Particle size determines the site of deposition within the respiratory system. Many occupational diseases such as silicosis and asbestosis are associated with material deposited in particular regions of the respiratory tract. Criteria have been developed to define critical size-fractions most closely associated with various health effects.4 The various critical fractions established by the American Conference of Governmental Industrial Hygienists (ACGIH) are shown in Figure 4.1.5 and are defined as follows. Inhalable fraction: This is the fraction of air-borne particulate matter which can be hazardous when deposited anywhere in the upper or lower respiratory tract. Thoracic fraction: Those particles which are hazardous when deposited anywhere within the pulmonary airways and the gas exchange region. Respirable fraction: Those particles which are a hazard when deposited in the gas exchange region of the lungs.

Ergonomic factors are increasingly important causes of injury in the workplace. Collectively, the term musculoskeletal disorders (MSDs) refers to conditions that involve the nerves, tendons, muscles, and supporting structures of the body. Carpal tunnel syndrome is an example of a wellrecognized work-related MSD of the wrist.

Chemical health hazards Many chemicals are capable of producing adverse acute or chronic health effects. Chemical hazards include exposures to chemical mists, vapors, gases or particulates (dust and fumes) through inhalation or by absorption through the skin. OSHA defines hazardous chemicals to include carcinogens, reproductive toxins, irritants, corrosives, sensitizers, hepatoxins, nephrotoxins, agents which act on the hematopoietic system and agents which damage the lungs, skin, eyes or mucous membranes. Certain biologic materials are health hazards.

Biologic health hazards

48 Occupational Hygiene

Aerosols Normal impurities • Quiet outdoor air

Fog

Mist

Rain drops

Metallurgical dust and fumes Smelter dust and fumes NHCl3 fumes Range of sizes

Alkali fumes

Small range- average

Foundry dust Flour mill dust, sprayed zinc dust Ground limestone Sulphic ore, pulps for flotation

Doubtful values Sulfuric acid mist Cement dust Condensed zinc dust Zinc oxide fumes Tobacco Tobacco mosaic necrosis virus virus

Pulverized coal Insecticide dusts

Plant spores

Bacteria Pollens

Carbon black Tobacco smoke Diameter of gas molecules

Sneezes Fly ash

Oil smoke Magnesium oxide smoke

Sand tailings

Rosin smoke

Washed foundry sand

(Enamels) Pigments (Flats) Silver iodide

Spray dried milk Human hair diameter

Combustion nuclei Sea salt nuclei

Reference sizes

Visible to eye

Screen mesh 0.0001 0.0005 0.001

0.005 0.01

0.05 0.1

0.5 1 5 Particle size (microns)

10

400 325 200 100 65 48 35 28 10

50 100

500 1000

5000 10000

Figure 4.1.2: Size ranges for airborne contaminants, and comparisons with reference sizes.

Gases: Gases are formless fluids that expand to occupy the space or enclosure in which they are confined. The gaseous state is restricted to temperatures and pressures which would normally be present in the ambient or occupational environments.

Vapors: Vapors are the gaseous form of substances that are normally in the solid or liquid state at room temperature and pressure. Evaporation is the process by which a liquid is changed into the vapor state and mixed with the surrounding atmosphere. Many solvents will volatilize

Deposition, collection efficiency

Figure 4.1.3: Electron microscope image of airborne particles from an industrial rubber operation using non-fibrous talc.

Deposition, collection efficiency

Deposition, collection efficiency

Recognition of Occupational and Environmental Hazards 49 1.0

a

IPM

0.8 0.6 0.4 0.2 0 0.01 1.0

DETotal

0.1

1

10

100 b

TPM

0.8 0.6 0.4 0.2 0 0.01 1.0

DEba

0.1

1

10

100

RPM

c

1 10 Diameter (μM)

100

0.8 0.6 0.4 0.2 0 0.01

DEa

0.1

Figure 4.1.5: ACGIH reference curves for inspirable (a), thoracic (b), and respirable (c) fractions of airborne particulates. In (a), total inhalable fraction, or IPM, demonstrates that most particulates under 100 μm are inhalable. The DEtotal curve reflects the fact that all but particles between 0.1 and 1.0 μm are well retained in the respiratory tract. In (b), it is apparent that particles under 10 μm are able to enter the bronchial tree and are called thoracic, or TPM. DEba shows the proportion of particles of each size deposited in the bronchoalveolar compartments. Particles of smaller size ( rabbit > rat > guinea pig > mouse > forearm skin of humans.20 The pig is often considered to be the best animal model for studies of transdermal penetration of drugs and chemicals.26 Although skin is generally not considered a major site of biotransformation of chemicals, biotransformation of some substances in the skin may be of critical toxicologic significance. The activity of most drug-metabolizing enzymes in the skin is low relative to the liver, but even low rates of metabolism for chemicals that are slowly absorbed or accumulate in the epidermis can be important in the ultimate disposition. The skin has the ability to perform a variety of oxidative biotransformations, including the oxidation of hydrocortisone, testosterone, aliphatic hydrocarbons, alicyclic hydrocarbons, and polyaromatic hydrocarbons. Indeed, the cytochrome P450-mediated oxidation of benzo[a]pyrene to the vicinal diol epoxide in the skin is thought to be essential to the ability of this compound to induce papillomas and carcinomas following repeated dermal application. Although the activity of aryl hydrocarbon hydroxylase in skin is only about 2% that of the liver, it is highly inducible following repeated dermal applications of polyaromatic hydrocarbons. The ability of the skin to perform a variety of other biotransformation reactions, including reduction reactions, hydrolysis of ethers, and glucuronide and sulfate conjugation, has been reported. Evidence for cutaneous first-pass metabolism of chemicals following dermal application has been demonstrated. Among the various physicochemical factors that can influence percutaneous absorption, the extent of hydration is one of the more important. The permeability of skin has been shown to increase as much as 4-5-fold following hydration. The increase in penetration after hydration may result from an increase in the size of the pores in the stratum corneum. However, at high water concentrations, there are changes in both the diffusion and activity coefficients of the penetrating agent, in addition to physical changes in the stratum corneum. Dehydration also may enhance absorption by causing damage to the integrity of the stratum corneum. At water concentrations below 10%, the stratum corneum becomes brittle and loses its functional integrity.20 Occlusion of the skin is an especially effective means of enhancing percutaneous absorption; in one study, occlusion with an impermeable barrier resulted in a 50-fold increase in penetration compared with that of the same chemical in an identical formulation without occlusion. The significance of occlusion on skin permeability is especially noteworthy in the occupational setting. The use of

gloves that serve as an incomplete barrier to chemicals may actually enhance percutaneous absorption by (1) increasing permeability by increased skin hydration and elevated temperature and (2) increasing the contact time and epidermal concentration, especially for volatile chemicals that would otherwise evaporate from the surface of the skin. Thus, it is imperative that gloves worn to protect against skin contact from chemicals be truly impermeable to the chemical of concern. Occupational and environmental exposure to chemicals often occur in mixed media. For example, many pesticides are formulated in organic solvents. The nature of the vehicle that contains the solute may be of prime importance in determining dermal bioavailability. The rate and extent of penetration of a chemical is dependent on the relative partitioning of the substance between the vehicle and the epidermis. The use of barrier creams may reduce dermal absorption by preventing contact of the material with the dermis or by providing a ‘skin-vehicle’ partition coefficient that highly favors partitioning of the chemical into the vehicle. Conversely, chemicals that are only sparingly soluble in a solvent but are readily soluble in the lipid matrix of the epidermis may penetrate the skin much more rapidly than if they were applied directly or in a vehicle in which they are highly soluble. For example, when pure alcohols are applied to the skin, their rates of penetration are much lower than their corresponding rates in aqueous solutions. Some substances, such as dimethyl sulfoxide, dimethylformamide, and propylene glycol, can act as ‘sorption enhancers’, greatly increasing the permeation of other substances through the skin.27 Percutaneous absorption of some substances has been increased 1000-fold by sorption promoters. Detergents and other surfactants also can alter the percutaneous absorption of other substances. Among the various types of surface-active agents, anionic surfactants are the most effective at enhancing percutaneous absorption, followed by non-ionic and cationic agents.

Classification of chemicals in terms of risk from dermal exposure. The American Conference of Governmental Industrial Hygienists (ACGIH) and a variety of occupational health regulatory agencies worldwide have identified compounds that are regarded as especially hazardous following dermal contact by use of the ‘skin’ denotation. The ACGIH lists 179 different compounds regarded as skin hazards. Although there is a large degree of inconsistency among the different lists that have been compiled, with nearly 400 chemicals appearing on at least one list, there is a smaller group of chemicals that are commonly identified as skin hazards (Table 5.5). Although the limitations of such lists have been discussed, they do provide a useful and quick reference for chemicals that may be especially problematic if they do come into contact with human skin. However, as noted by Granjean,28 ‘. . . the lack of skin denotation for a particular substance does not necessarily exclude that hazardous quantities can be absorbed, provided that, e.g., the vehicle is right or the skin is occluded or damaged’. Finally, the contributions of dermal

96 Toxicology Acrylamide Acrylonitrile Aldrin Alyl alcohol Aniline Benzene Carbon disulfide Carbon tetrachloride Cellosolve (glycol ether) Chlordane Chlorinated naphthalenes Chloroprene Cresol Demeton-methyl Diazinon 1,2-Dibromoethane 2,2′-Dichlorodiethyl ether Dichlorvos Dieldrin Diethylaminoethanol Dimethylacetamide Dimethylaniline 1.1-Dimethylhydrazine* Dinitrobenzene Dinitro-o-cresol Endrin

Epichlorohydrin EPN Ethylene chlorohydrin Ethylene glycol dinitrate Ethylene glycol monobutylether Ethylene glycol monoethyl ether acetate Ethylene glycol monomethyl ether Ethylene glycol monomethyl ether acetate Ethyleneimine Furfural Heptachlor Hydrazine Hydrogen cyanide Lindane Malathion Mercury Methyl acrylate Methyl alcohol Methyl bromide Methyl sulfate Mevinphos Morpholine n-Butylamine n-Methylaniline N,N-Dimethylformamide

Nicotine Nitrobenzene Nitroglycerin Nitrotoluene o-Methylcyclohexanone p-Nitrochlorobenzene p-Dioxane p-Nitroaniline p-Phenylenediamine Parathion Pentachlorophenol Phenol Phenylhydrazine Picric acid Polychlorinated biphenyls Propargyl alcohol Sodium fluoroacetate 1,1,2,2-Tetrachloroethane Tetraethyl lead Tetryl Thallium Toluidine Trinitrotoluene Xylene Xylidine

* Only 1,1-dimethylhydrazine does not have a ‘skin’ denotation on the ACGIH list. Adapted from Granjean P, Berlin A, Gilbert M, et al. Preventing percutaneous absorption of industrial chemicals: the ‘skin’ denotation. Am J Ind Med 1988; 14:97-107. © 1988 John Wiley & Sons, Inc. Reprinted by permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.

Table 5.5 Chemicals commonly identified as hazardous by dermal exposure

absorption to overall exposure to toxic substances is often underestimated, especially for volatile chemicals. For example, using an experimentally determined dermal permeability value for benzene, the amount of benzene absorbed through the hands during long-term exposure to solvents contaminated with 0.1% benzene could increase the theoretical risk estimate for leukemia by 42%.29

Absorption of chemicals by inhalation exposure The lung serves as an important site of contact with chemicals in the external environment. Such contact can result in direct damage to the respiratory epithelium or may lead to systemic toxicity following absorption into the blood stream. Like the gastrointestinal tract, the lung is designed for optimal absorption. It has a large total surface area (50-100 m2) and a high blood flow (4 to 5 L/min) that is in intimate contact with the respiratory epithelium, which itself is of minimal thickness (10 μm diffusion distance). Absorption of chemicals from inspired air can occur regardless of the physical form. For gases and fine vapors, absorption occurs by direct diffusion from alveolar air spaces across the epithelial cells, whereas for aerosols and other types of particles, deposition occurs along various aspects of the tracheobronchial tree, with the specific location depending on the size and density (mass median diameter). (See Roth30 for a comprehensive review of toxicology of the respiratory system.)

Absorption of gases and vapors. The rate of absorption of gases and vapors by the lung is largely a function of the

‘blood-gas partition coefficient’. Applying the principles of Henry’s law to blood and alveolar gas, the concentration of a gas in the blood as it leaves the lungs is dependent on the solubility of the gas in blood, which is defined as the ratio of the concentration of the dissolved gas in the blood to the concentration in the gas phase at equilibrium.31 Thus, chemicals with high blood-gas partition coefficients will have a high rate of uptake into the blood stream, relative to those chemicals with low blood-gas partition coefficients. For chemicals with a low solubility in blood, only a small fraction of the gas present in the alveolar space will be removed, and an increase in respiration will not enhance uptake. An increase in blood flow through the lung will greatly enhance uptake (perfusion limited). Conversely, an increase in respiration rate can significantly enhance the extent of absorption of gases that are readily soluble in blood because the delivery of gas to the blood, and not the dissolution into the blood, may be rate limiting in regard to uptake (ventilation limited). The solubility of a gas in the blood should not be equated to its solubility in water because components of blood other than water can greatly affect the solubility of some gases. For example, carbon monoxide, because of its high affinity for hemoglobin, has an extremely high apparent blood-gas partition coefficient even though it is only sparingly soluble in water. The site of absorption of gases and vapors from the lungs is largely dependent on water solubility. Highly water-soluble gases present in low concentrations in inspired air, such as hydrochloric acid or sulfur dioxide, are usually removed in the upper respiratory tract, although higher concentrations and longer durations of exposure

Factors That Influence Toxic Responses 97 Particle size (μm) ≥ 11.0

Outside respiratory tract

7.0 – 11.0

Nose

4.7 – 7.0

Pharynx, larynx

3.3 – 4.7

Trachea and primary bronchi

2.1 – 3.3

Secondary bronchi

1.1 – 2.5

Terminal bronchi

0.43 – 1.1

Alveoli

allow bioavailability to the alveolar regions of the lung where they may cause serious pulmonary damage. Gases that are less soluble in water, such as nitrogen oxides and ozone, are not efficiently removed by the upper respiratory tract and tend to reach alveolar regions at relatively low concentrations. For those gases and vapors that impart their toxic properties systemically, the site of absorption is of far less concern than the rate and extent of absorption.

Deposition and clearance of particles from the lung.

In contrast to gases, for which contact with the pulmonary epithelium occurs exclusively by diffusion, particles (aerosols) deposit on the pulmonary epithelium primarily through physical forces dictated by the particle size, shape, and mass. Inhaled aerosols are heterogeneous in size, and the size distribution usually follows a log-normal pattern. The geometric mean and geometric standard deviation of the particle size distribution is commonly obtained by plotting the cumulative percentage of particles less than a stated size increment against the log of the stated size on log probability paper. The ‘mass median diameter’ is a common means of characterizing aerosol size distribution because it considers not only the physical diameter of the particles, but also their density. For particles that are not spheric (e.g., asbestos fibers), a consideration of aerodynamic drag becomes important, and the determination of ‘aerodynamic diameter’ is most useful. This term describes the settling behavior of a non-spheric particle, whatever its size, shape, and density, in terms of a unit density sphere having the same settling velocity as the particle under study. The aerodynamic diameter of particles is a primary determinant of the site of deposition within the respiratory tract. Figure 5.9 shows the influence of particle size distribution on the regional deposition of an inhaled

Figure 5.9: Particle deposition by size distribution in various regions of the lung. Particles deposit in different regions of the lung, based largely on the size of the particle. (From Kennedy GL. Inhalation toxicology. In Hayes AW, ed. Principles and methods in toxicology, 2nd edn. New York: Raven Press, 1989.)

aerosol. It is evident that aerosols with aerodynamic diameters greater than 10 μm are not capable of passing beyond the nasopharyngeal area.32 Particles may be deposited on the respiratory tract epithelium by three fundamental processes. 1. Inertia. Particles with sufficient mass will collide with the surface of the respiratory epithelium at points of branching and curvature. As the direction of air velocity changes, the inertial force of the particles will prevent them from changing directions at the same rate as that of the air flow. The greater the mass, the less the ability of the particles to change direction with air flow. Thus, deposition of particles occurs by impaction. For particles that are non-spheric (e.g., asbestos fibers), the term ‘interception’ is sometimes used to describe the deposition that results when an edge of the particle contacts the surface of the epithelium. 2. Sedimentation. Particles that are of sufficiently small size to escape deposition by inertia may deposit on the respiratory epithelium through sedimentation when the velocity of air flow becomes slow. Gravity is the predominant force behind sedimentation, and the rate of sedimentation is proportional to the density of the particle and the square of its diameter. 3. Diffusion. For extremely small particles, Brownian motion, in which suspended particles are bombarded by surrounding gaseous molecules, is the principal means of deposition. In contrast to both inertia and sedimentation, the rate of diffusion is not appreciably influenced by the particle density, but it is inversely proportional to the diameter. Although particle size (i.e., aerodynamic diameter) is the principal determinant of deposition, other factors, such as

98 Toxicology breathing pattern, airway diameter, and the anatomy of the nasal, oral, and pharyngeal areas, are also important. Thus, pathologic or irritant conditions that result in narrowing or obstruction of airways, excessive bronchial secretions, or other conditions that alter the size and shape of the respiratory tract can alter the extent and pattern of particle deposition.

Clearance of particles from the lungs. Particles may be removed from the lungs by several mechanisms. Particles deposited in the upper nasopharyngeal area are cleared primarily by entrapment in the mucous lining and removal by reflex responses, such as sneezing or coughing, or through the continuous movement of mucus up the mucociliary escalator, where the entrapped particles are either swallowed or expectorated. The former is an important toxicologic consideration because it can result in significant ingestion of toxic materials even though exposure occurred by inhalation of particles. The mucociliary escalator is the primary means of clearance of particles in the tracheobronchial tree and functions down to the level of the terminal bronchioles. The clearance half-time by this process is between 12 and 24 hours. Phagocytosis of particles by pulmonary macrophages is the principal route of elimination of particles that deposit in the alveoli and terminal bronchioles. Macrophagecontaining particles may migrate to the ciliated epithelium of the terminal bronchioles and move up the mucociliary escalator, or they may enter the lymphatic vessels. This process takes from 2 to 6 weeks for effective elimination of particles. Some particles may actually be solubilized by body fluids and the constituents absorbed into the blood stream or lymph. Obviously, any toxicologic response that reduces or eliminates particle clearance can result in more extensive accumulation of particles and potentially enhance the toxicity of the particles themselves or of toxic constituents that are slowly dissolved from the surface of the particles. Cigarette smoking has been demonstrated to reduce significantly the rate and efficiency of the mucociliary escalator. The inefficient removal of particles in heavy smokers may contribute, at least in part, to the well-described synergistic interaction in the lung cancer risk between asbestos and smoking.

DISTRIBUTION OF CHEMICALS IN THE BODY First-pass effect Because the critical dose parameter for any toxic effect relates to the concentration of a chemical at the biologic receptor site, the extent and nature of distribution of the absorbed chemical in the body can have a significant impact on the toxicity of that chemical. After a substance has entered the systemic circulation, it is potentially available to all tissues in the body. However, biologic activity at or near the site of absorption may drastically reduce the availability of the original chemical to distant sites, even

though absorption into the circulation is complete. This phenomenon, referred to as the ‘first-pass effect,’ is most often described in association with absorption of substances following ingestion.33 Generally, the first-pass effect after oral absorption is the result of efficient extraction and metabolism of a chemical by the liver. The blood perfusing the gastrointestinal tract is collected in the portal vein, and thus, all substances absorbed from the gastrointestinal tract must pass through the liver prior to distribution to other organs. The liver maintains a high capacity for both extraction and biotransformation of exogenous substances, and can effectively prevent some chemicals from ever reaching the rest of the body. Although hepatic extraction is generally recognized as the most important site of removal for chemicals demonstrating a first-pass effect following oral administration, extraction and biotransformation by the gastrointestinal epithelium may also contribute.34 The skin is also capable of first-pass metabolism of chemicals following dermal exposure, although the extent of this effect is generally far less important than that for oral exposure.

Binding and storage Binding of chemicals to proteins can also have a dramatic effect on the toxicologic response to a given absorbed dose. Many chemicals bind to plasma proteins, especially albumin and globulins. Because only the concentration of ‘free’ (unbound) chemical is available for interaction with biologic receptors, extensive protein binding can greatly reduce the toxicity of a compound. For chemicals that are highly bound to plasma proteins, even a small shift in protein binding can have a large effect on their pharmacologic functions. For example, if the equilibrium binding of an absorbed chemical to plasma proteins is 98%, a change in binding status from 98% to 96% would result in a doubling of the concentration of free chemical and an attendant increase in its toxicity. This is especially a problem when multiple exposures occur to substances that share the same binding site, and it is a common mechanism of drug–drug or drug–chemical interactions. If protein binding of toxic substances to intracellular proteins occurs, the tissue may serve as a ‘sink’ for accumulation of a chemical, prolonging the biologic half-life of the chemical. For example, Cd is a highly toxic element that binds avidly to intracellular binding proteins called metallothioneins. These low molecular weight proteins, which are inducible (their quantity increases) on repeated exposure to Cd and certain other metals, are present in relatively high concentrations in the liver and kidney. Following oral exposure to a dose of Cd, the majority of the absorbed dose (usually only a small percent of the exposed dose) is concentrated in the liver and bound largely to metallothioneins.35 In this instance, the preferential binding serves as a protective mechanism against Cd-induced liver damage by preventing Cd from binding to essential cellular proteins. However, the Cd-metallothionein complex is slowly transported out of the liver to the kidney, where the complex is

Distribution of Chemicals in the Body 99 extracted by the proximal tubules. As the metallothioneins protein is degraded by normal catabolic processes, Cd is released and binds to either renal metallothioneins or other intracellular thiols, including critical proteins. The net effect of this is an extraordinarily long biologic half-life for Cd (up to 30 years), with the majority of the body burden eventually concentrated in the renal cortex. After the concentrations exceed the binding capacity of renal metallothionein, irreversible and potentially fatal kidney damage can result. Lipid-soluble chemicals that are poorly metabolized will redistribute to poorly perfused fat depots following equilibration in the rapidly perfused tissues. Many lipid-soluble chemicals rapidly affect central nervous system function after exposure. A relatively high fraction of the absorbed dose enters the brain because of the high rate of perfusion of the brain and partitions into brain tissue because of the relatively high lipid content. The duration of action of this effect following an acute exposure is often short because the chemical quickly redistributes from the brain to the more slowly perfused peripheral fat stores, or to sites such as the liver where the chemical may be more rapidly metabolized to water-soluble forms and eliminated by urinary or biliary excretion.

Special barriers to distribution The so-called ‘blood–brain barrier’ is an effective anatomic barrier to penetration of highly water-soluble chemicals into the brain.36 This barrier is the result of several unusual anatomic characteristics, including the presence of tight junctions (desmosomes) between capillary endothelial cells, the presence of glial foot processes that effectively surround the capillary endothelium, a contiguous basement membrane, and a relatively low protein content in the interstitial fluid of the brain. The relative contribution of the last two elements to the barrier function is probably small compared with that of tight junctions and glial cells. Thus, in contrast to most other organ systems, where intercellular gaps in the capillary endothelium allow free access of chemical into parenchymal cells, in the brain a chemical must pass through several biologic membranes (capillary endothelial cells and glial cells) and interstitial fluid before coming into contact with neurons. This barrier is effective against most water-soluble chemicals, except very small substances that can diffuse with water through membrane pores, (e.g., lithium). However, it presents little impediment to the diffusion of lipid-soluble substances into the brain, and consequently, nearly all organic solvents can readily enter the brain and produce aberrant effects on central nervous system function. A similar concept has been described for the testis. The ‘blood–testis’ barrier, like the blood–brain barrier, does limit the bioavailability of some water-soluble substances to germinal cells, but again, it provides little protection against lipid-soluble compounds.37 Sertoli cells appear to be primarily responsible for the blood–testis barrier because adjacent Sertoli cells form occluding junctions between the germ cells in various stages of development.

The distribution of chemicals to different organs systems may also be affected by the presence of specific carriermediated transport processes. Such processes are most active in the liver, kidney, and intestine, although transport processes have been described in the choroid plexus and lung. The transport processes in the liver and kidney function primarily to aid in the uptake and elimination of foreign substances and metabolic byproducts from the blood, whereas those in the intestine function primarily to aid in the absorption of water-soluble nutrients that would otherwise not be absorbed. The physiologic functions of transport processes for organic substances identified in the lung are not known, but the presence of these transport systems can have important toxicologic consequences. For example, the herbicide paraquat is highly and specifically toxic to the lung, regardless of the route of administration, due in part to an active transport system that concentrates paraquat in type II pneumocytes.38

Biotransformation of chemicals For many chemicals, the toxic effects are highly dependent on the metabolic fate of the chemical in the body. Because few xenobiotics are actually fully metabolized to carbon dioxide and water, the metabolic processes that change the structure and characteristics of a chemical are appropriately referred to as ‘biotransformation’ reactions. There are a multitude of enzymatic pathways capable of such reactions, and the quantitative and qualitative differences in the ability of different organs to conduct such processes is often responsible for the ‘organotropic’ (i.e., organ-specific) effects of many chemicals. For example, the liver is the most common site of toxicity for chloroform and carbon tetrachloride, largely because of the ability of this organ to biotransform these compounds rapidly into reactive free radical intermediates.39 Regardless of the specific pathways involved, conceptually the ultimate ‘goal’ of biotransformation reactions is to render potentially toxic chemicals less toxic. This is accomplished in two ways by: (1) the addition of polar groups to lipid-soluble chemicals, which decreases the ability of chemicals to penetrate cell membranes on the one hand and enhances the rate of elimination in urine or bile on the other, and (2) the alteration of chemical structure so that the chemical no longer ‘fits’ the specific biologic receptor (e.g., catalytic sites of enzymes or neurotransmitter receptors). However, it is now widely recognized that intermediates formed in the process may actually be of far greater toxicity than the ‘parent’ molecule, and thus, some biotransformation reactions may actually be deleterious ‘activation reactions,’ whereas others are considered ‘detoxification’ reactions. Biotransformation reactions are commonly divided into two broad categories: phase I and phase II reactions.40 Phase I reactions are so named because they are generally the first biotransformation step in what is often a multistep process leading to the eventual excretion of the biotransformed products. Phase II reactions are those enzymatic processes that use the products of phase I reactions to impart further

100 Toxicology endoplasmic reticulum is referred to as the ‘microsomal fraction,’ and thus cytochrome P450-mediated reactions are often called microsomal oxidation reactions.41 Microsomes are readily prepared by differential centrifugation of homogenized tissues. The cell membranes, nuclei, mitochondria, ribosomes, and most other intracellular organelles are pelleted after centrifugation at 9000-15,000 g. Because the fragments of the smooth endoplasmic reticulum are less dense than these other organelles, they remain suspended in the supernatant (sometimes referred to as the S9 fraction). The microsomes can be separated from the soluble fraction of the cell homogenate by centrifugation for 1 hour at 100,000 g. The microsomal pellet obtained by this process can then be resuspended and used for in-vitro studies. The 100,000 g supernatant contains only soluble enzymes, and it can also be used for studies of soluble biotransformation enzymes. It is important to recognize that, in addition to the cytochrome P450 system, microsomes contain many other important biotransformation enzymes, and thus, microsomal metabolism is not completely synonymous with cytochrome P450-mediated reactions. There are actually numerous forms of cytochrome P450 enzymes (more than 30 distinct enzymes have been identified in the human liver), and the genetics of these enzymes is becoming well understood. Although there is generally a broad overlap in substrate specificity between individual enzymes, specific P450 enzymes play a major role in the biotransformation of specific xenobiotics, and genetic differences in the expression and activity of specific P450 enzymes can be of substantial toxicologic and pharmacologic significance (Fig. 5.10). Genetic polymorphisms in xenobiotic metabolism are discussed in detail subsequently. The number of chemicals that can undergo oxidation by the cytochrome P450 complex is large, as is the variety

structural changes, usually greatly increasing the water solubility. However, this classification can be confusing because some biotransformation enzymes may act as either phase I or phase II enzymes, depending on the substrate. For example, the hydrolysis of epoxides by the enzyme epoxide hydrolase could be considered a phase I reaction if it were the first enzyme to metabolize an exogenous epoxide, such as trichloropropene oxide or heptachlor epoxide, whereas it would be considered a phase II reaction if it were acting on an epoxide generated endogenously by an oxidative pathway. However, with numerous notable exceptions, oxidation, reduction, and hydrolytic pathways are generally considered phase I reactions, whereas conjugation reactions are usually classified as phase II reactions. Virtually all biotransformation reactions of toxicologic significance can be identified as proceeding by one of the following four basic categories of pathways: oxidation, reduction, hydrolysis, and conjugation.

Oxidation reactions The majority of oxidative biotransformation reactions are mediated by the cytochrome P450-containing mixedfunction mono-oxygenase system. However, there are other important oxidative pathways, and thus, oxidation reactions are most conveniently divided into ‘cytochrome P450-mediated’ and ‘non-cytochrome P450-mediated’ pathways.

Cytochrome P450-mediated oxidation reactions Cytochrome P450 is a heme-containing, membrane-bound complex located in the smooth endoplasmic reticulum. For experimental purposes, this enzyme system is most easily studied in crude subcellular fractions of homogenized tissues. The subcellular fraction containing the smooth

O2 O2

NADPH

Flavoprotein [oxid]

O2

S

Flavoprotein [red]

S

-1 Cytochrome b3 [red]

P-450[Fe+3]

P-450[Fe+2]

Cytochrome P-450 reductase NADP+

S

P-450[Fe+2]

[e-]

Flavoprotein [oxid]

NADPH

Cytochrome b3 reductase O2

S

-2

Cytochrome b3 [oxid]

Flavoprotein [red]

NAD+

P-450[Fe+3] P-450[Fe+3] S (Substrate)

H2O S-OH (Product)

Figure 5.10: Catalytic cycle of the cytochrome P450-dependent, mixed-function mono-oxygenase system. The cytochrome P450-dependent, mixed-function mono-oxygenase system catalyzes the addition of atomic oxygen to lipophilic substrates such as hydrocarbons. The iron in the heme portion of the molecule undergoes oxidation and reduction, transferring electrons from an electron donor (NAD[P]H) to molecular oxygen (O2) and the substrate (S). The net result is the insertion of one atom of molecular oxygen into the substrate, and the second atom combines with hydrogen ions to form water.

Distribution of Chemicals in the Body 101 Name

Substrate

Product(s) CH

Aliphatic hydroxylation

R-CH2-CH2-CH3

R-CH2-CH-CH3

Alkene epoxidation

R-CH=CH-CH3

R-CH–CH-CH3

O

X Aromatic epoxidation

X

H O H

O Aromatic hydroxylation

N-dealkylation

CH

R

R-N

CH2R1 CH2R1

R-N

CH2R1 CH-R1

R-N

CH2R1 + H

RCH2O

CH S-dealkylation

O-dealkylation

N-hydroxylation

R-S-CH2R1

R-O-CH2R1

RCH2-NH2

S Oxidative desulfuration

(RCH2)2-P-O-R1

CH

R-SH

+

RCH2O

R-OH

+

RCH2O

R-S-CHR1

OH R-O-CHR1

RCH2-NHOH

RCH2N=O

RCH2NO2

O (RCH2)2-P-O-R1

O Oxidative dehalogenation

CH2X2

HO-CH-X2

HC-X + H+ + XC=O + H+ + X-

Figure 5.11: Common oxidation reactions mediated via cytochromes P450. The cytochromes P450 system is capable of mediating the oxidation of a wide variety of drugs and other chemicals. See text for further discussion.

of biotransformed products that can result from a single substrate. However, it is possible to predict with some degree of confidence the possible array of metabolites that could result from cytochrome P450-mediated oxidation reactions by understanding the basic reactions that have been described. Cytochrome P450-oxidation reactions are commonly classified into ten different categories (Fig. 5.11). The oxidation of aromatic and aliphatic hydrocarbons is one of the most common of the P450-mediated oxidations. The products of such reactions are generally hydroxylated metabolites. The hydroxyl group increases polarity

and also provides a molecular site for conjugation with highly polar groups, such as glucuronic acid or sulfate, by phase II biotransformation pathways. However, oxidation of hydrocarbons can also lead to the formation of highly electrophilic epoxides or other reactive intermediates, which may bind to nucleophilic sites within the cell, disrupting cellular function. Two common nucleophilic sites within the cell are reduced thiols in proteins and certain bases in DNA. The binding of electrophilic intermediates to thiols or other nucleophilic sites in essential proteins may result in cell injury or death, with a resultant loss in organ function. The formation of reactive epoxide

102 Toxicology intermediates by cytochrome P450 oxidation is thought to be the initiating step in chemical carcinogenesis by a wide variety of chemicals, including polyaromatic hydrocarbons and the fungal toxin, aflatoxin B1. Oxidative dechlorination of carbon tetrachloride by cytochrome P450 is thought to be essential to the hepatotoxic effects of this chemical. Oxidative desulfuration by cytochrome P450 is essential to both the insecticidal activity and toxicity of many organophosphate insecticides because it is only the ‘oxon’ analogue that is capable of binding to and inhibiting acetylcholinesterase. Thus, although cytochrome P450-mediated oxidation reactions serve as a principal means of elimination of a wide variety of chemicals, this same pathway is also responsible for the activation of a variety of chemicals to more highly toxic and carcinogenic forms. Whether biotransformation through the P450 pathway results in net activation or detoxification is dependent on many other factors and is not always readily apparent. The kinetics of different cytochromes P450 enzymes toward the same substrate, relative to the kinetics of competing reactions, are critical to the ultimate outcome. For example, the widely used analgesic acetaminophen at normal therapeutic doses is largely metabolized by conjugation reactions (sulfation and glucuronidation), with only about 4% biotransformed by the cytochrome P450 system. Although one product of P450 biotransformation of acetaminophen is quite reactive, it is rapidly detoxified by intracellular reduced glutathione. However, at high doses, the ratio of biotransformation shifts more to the P450 system; with increasing doses, the tissue stores of glutathione become depleted, and the reactive intermediate then reacts with cellular thiols, binding to intracellular proteins and creating cellular damage, primarily in the liver where the majority of biotransformation of acetaminophen occurs.42

Induction of cytochrome P450. An interesting and clinically relevant phenomenon of most cytochrome P450 enzymes (and many other biotransformation pathways) is their ability to increase activity after repeated exposure to certain exogenous agents. Enzyme induction occurs by gene activation in which the rate of messenger RNA synthesis increases, with a concomitant increase in enzyme production and activity. A wide variety of chemicals have been identified as ‘microsomal enzyme inducers’, but the specific pattern of enzyme induction differs. The two most widely studied enzyme inducers are phenobarbital and 3-methylcholanthrene (3-MC). Phenobarbital induces a broad spectrum of P450 enzymes (P4502B, 2C, 3A, and several other important biotransformation enzymes, including some glutathione S-transferases and glucuronyl transferases). 3-MC, other polyaromatic hydrocarbons, and the controversial and potent toxicant 2,3,7,8-tetrachlorodibenzo-p-dioxin (dioxin) induce a much narrower spectrum of biotransformation enzymes, most notably the cytochrome P4501A family, via the aryl hydrocarbon hydroxylase receptor (AHH) complex. Although the spectrum of enzymes induced via the AHH receptor complex is more limited, the extent of

induction is often great, because constitutive (background) expression of the cytochrome P501A family of enzymes is very low. Thus, in the absence of exposure to AHH receptor agonists, enzyme activity is very low but increases dramatically on exposure. Because this enzyme system is principally responsible for the biotransformation of polyaromatic hydrocarbons, induction can be viewed as an attempt by the organism to respond to external stimuli by altering its capacity to deal with those stimuli. It is unclear whether induction of cytochrome P4501A enzymes is indeed beneficial to the organism because this enzyme system is responsible for the activation of polyaromatic hydrocarbons to carcinogenic intermediates, as well as formation of less toxic forms of the parent molecule. Microsomal enzyme induction by phenobarbitallike inducers, which include many chemicals found in the workplace and general environment (e.g., some forms of polychlorinated biphenyls [PCBs] and organochlorine insecticides), can greatly accelerate the rate of biotransformation of numerous chemicals.43 The significance of this effect is dependent on the relative toxicity of the products of biotransformation. Thus, induction with phenobarbital may decrease the toxicity of some compounds while enhancing the toxicity of others. Such chemical–chemical interactions are well described for therapeutic drugs, in which patients receiving chronic barbiturate therapy for epilepsy may require substantial dosage adjustment for other medications to be effective.

Non-cytochrome P450 oxidation reactions Alcohol and aldehyde dehydrogenases. The pathways of oxidation of short-chain alcohols, such as methanol, ethanol, and isopropanol, are very important biotransformation routes because exposure to these substances is very common. Although cytochrome P450 2E1 enzymes are capable of oxidizing ethanol, the relative importance of this reaction to the overall metabolism of ethanol in humans is small. Primary oxidation of short-chain alcohols occurs by a two-step sequence involving alcohol dehydrogenase and aldehyde dehydrogenase.44 The relative rates of metabolism of these two enzymes are important to the toxicologic effects of various alcohols. In the case of ethanol, the oxidation to acetaldehyde proceeds relatively slowly and in a zero-order kinetic fashion at doses capable of inducing even mild inebriation. That is, the quantity of ethanol metabolized is a fixed amount, rather than a fixed percentage, of the available dose. Thus, with repeated consumption of alcohol, the fraction of the dose metabolized becomes less as the dose increases. Aldehyde dehydrogenase functions relatively more efficiently than does alcohol dehydrogenase, preventing any significant accumulation of acetaldehyde under normal circumstances. Acetaldehyde is toxic and causes numerous adverse effects (headache, nausea, vomiting, hypotension, and flushing), and thus, chemicals that inhibit aldehyde dehydrogenase may produce these effects when combined with relatively low doses of alcohol. Several drugs and a few non-drug chemicals are capable of this effect. Disulfiram has been

Distribution of Chemicals in the Body 103 used clinically for this purpose as a form of aversion therapy in the treatment of alcoholism.45 Methanol metabolism proceeds by the same pathway as that for ethanol, but at a rate about seven times slower than that of ethanol. However, the consequences of formation of the two sequential metabolites, formaldehyde and formic acid, are much more severe. Virtually all of the severe and potentially irreversible effects of methanol can be attributed to these two metabolites. Retinal damage and blindness is thought to result from the localized production of formaldehyde in the retina. Life-threatening metabolic acidosis results from the formation of formic acid by aldehyde dehydrogenase. Because ethanol is preferentially metabolized by alcohol dehydrogenase and methanol will be eliminated largely unchanged in the urine if given adequate time, the inhibition of methanol biotransformation by intravenous infusion of ethanol or administration of fomepizole, an inhibitor of alcohol dehydrogenase, is standard therapy for methanol poisoning.46 If the amount of methanol ingested is large, pharmacologic treatment may be combined with extracorporeal hemodialysis to aid in the elimination of unchanged methanol. Alcohol and aldehyde dehydrogenases are also important in the biotransformation of various other alcohols, aldehydes, glycols, and glycol ethers and in the reduction of ketones to alcohols.

Flavin-dependent mono-oxygenases (FMOs). Oxidation of certain secondary amines, tertiary amines, imines, arylamines, and hydrazines, and many sulfur-containing chemicals proceeds by microsomal mono-oxygenases that are distinct from the cytochrome P450 enzyme family. This family of flavin-dependent mono-oxygenases appears particularly active in the human liver and competes with the cytochrome P450 system for the oxidation of nucleophilic nitrogen and sulfur atoms.47 The primary function of FMOs is in the detoxification of xenobiotics, and is likely the major route of oxidation of nucleophilic nitrogen, sulfur, phosphorus and other heteroatom-containing chemicals, including several important pesticides. Like the cytochromes P450, the FMOs are a multi-gene family of enzymes, with five different genes identified in humans to date. Interestingly, a genetic variant in one of these genes, FMO3, is associated with a malodorous syndrome known as ‘fish odor disease’. Individuals who inherit the variant form of FMO3 are unable to metabolize trimethylamine, a highly malodorous component of fish oil.48 Lack of metabolic elimination results in secretion of the compound in skin. Although not physiologically detrimental, this condition may be associated with pyschosocial disorders.

Reduction reactions Reduction reactions are relatively uncommon because cells function principally in an oxidizing environment. However, enzymatic reduction is important in the disposition of at least three general classes of chemicals: azo compounds, aromatic nitrates, and certain halogenated

hydrocarbons. Surprisingly, these reactions are also mediated by the cytochrome P450 system, but the substrate, rather than molecular oxygen, accepts the electrons and is reduced. Reduction reactions occur commonly in the gut flora and can be quite important in the disposition and toxicity of substances. As noted, the formation of nitrite from nitrate is a reduction reaction that can occur in the gut flora of infants, resulting in potentially severe methemoglobinemia. The reduction of azo dyes by the gut flora may alter the absorption and toxicity of these compounds. Complex azo dyes such as Direct Black 38 or Direct Blue 6 may release carcinogenic products, such as benzidine, after azo reduction, whereas the reduction of single azo compounds usually is a detoxification step. NAD(P)H:quinone oxidoreductase (NQO1) is widely distributed in mammalian species and tissues, and is involved in the reductive activation and detoxification of a variety of chemicals that contain, or can form, quinones or their derivatives. NQO1 is thought to play an important role in protection against oxygen radical formation that can occur from redox cycling of quinones.49 It also is involved in the metabolic activation of several widely used antitumor agents, including mitomycin C and streptonigrin.

Hydrolytic reactions The hydrolysis of esters, amides, and epoxides is important in the disposition of a wide variety of drugs and chemicals. The products of the hydrolysis of esters are an organic acid and alcohol, whereas the products of amide hydrolysis are an organic acid and a primary or secondary amine. The hydrolysis of epoxides results in a dihydrodiol. There are a variety of different esterase and amidase enzymes with broad and somewhat overlapping substrate specificities.50 Arylesterases are principally active on aromatic esters, whereas carboxylesterases are active on aliphatic esters. One form of arylesterase, called paraoxonase, is active in the hydrolysis of parathion and a few other organophosphate insecticides and has a demonstrated polymorphic distribution in the human population.51 Whether a genetic deficiency in paraoxonase activity places such individuals at increased risk for organophosphate insecticide poisoning is uncertain and the subject of current study. Genetic polymorphism in butyrylcholinesterase (pseudocholinesterase) is responsible for the prolonged muscular paralysis that occurs in 1-3% of the population following clinical use of succinylcholine. Cholinesterase is a more substrate-specific enzyme with important physiologic functions. A number of drugs and chemicals inhibit acetylcholinesterase activity, including all organophosphate insecticides. Esters may be hydrolyzed by plasma and hepatic esterases, whereas the metabolism of amides is more complicated, sometimes involving cytochrome P450-dependent N-dealkylation prior to amide hydrolysis. This is why amide-type local anesthetics (e.g., lidocaine) are generally longer acting that ester-type anesthetics (e.g., procaine).

104 Toxicology The hydrolysis of epoxides by microsomal epoxide hydrolase is generally regarded as a detoxification pathway because the dihydrodiol product is far less reactive than the epoxides.52 However, in some circumstances, such as in the biotransformation of polycyclic aromatic hydrocarbons, the hydrolysis of the first arene epoxide generated by cytochrome P450 oxidation is required for formation of the ultimate carcinogenic form, the ‘diol-epoxide,’ and thus, epoxide hydrolysis is considered part of the activation process.

Conjugation (Phase II) biotransformation reactions Glucuronide conjugation UDP-glucuronosyl transferases are a multigene family of microsomal enzymes that function to add a residue of glucuronic acid to a variety of xenobiotics.53 Because these enzymes are not capable of adding glucuronic acid directly to hydrocarbons, functional groups such as alcohols (COH), carboxylic acids (CCOOH), amines (CNH2), thiols (CSH), or sulfonamides are necessary. If such groups are present on the parent compound, conjugation can occur directly. However, for hydrocarbons or other chemicals that lack these sites, prior biotransformation to generate or ‘expose’ such functional groups is required. The addition of the glucuronic acid moiety to the xenobiotic imparts a substantial degree of water solubility and generates a bulky addition that would likely interfere with any structure-specific receptor interaction. Glucuronic acid per se is not used in the reaction; rather, an activated form (UDPGA), in which glucuronic acid is linked to the terminal phosphate residue of uridine diphosphate (UDP), serves as the co-factor for the reaction. Typically, glucuronidation reactions have been viewed primarily as ‘detoxification’ events, although some glucuronidation reactions are considered ‘activation’ reactions.54 Glucuronide conjugates are generally endproducts of biotransformation and are commonly found in both the urine and bile. Because bacterial flora contain the enzyme β-glucuronidase, which is capable of hydrolyzing the β-glycosidic linkage to re-establish the less polar substrate, glucuronide conjugates excreted in the bile may undergo extensive enterohepatic circulation if the degluconated molecule is sufficiently lipid soluble to be reabsorbed through the intestinal epithelium. β-glucuronidase activity has also been identified in the bladder epithelium and is thought to contribute to the etiology of bladder cancer for certain aromatic amines, which are conjugated with glucuronic acid, concentrated in the bladder, and then subsequently hydrolyzed by this enzyme to release a reactive, mutagenic form of the chemical.

Sulfate conjugation The arylsulfotransferases are a multi-gene family of soluble enzymes present in the liver, kidney, and intestine that function to add inorganic sulfate to phenols, aliphatic alcohols and hydroxylamines.55

Sulfate is donated to the phenol or alcohol by 3′-phosphoadenosine-5′-phosphosulfate (PAPS), a molecule analogous to adenosine diphosphate, except that the terminal phosphate is a sulfate residue and the 3′-hydroxyl group is also phosphorylated. Because the products of this reaction are ionized regardless of pH, sulfate conjugates are rapidly excreted in urine. Although sulfate conjugation usually results in products that are less toxic than the substrate, sulfate conjugation of some hydroxylamines (R-NH-OH), such as the sulfate conjugate of N-hydroxy-2-acetylaminofluorene, can rearrange to form reactive electrophilic species capable of interacting with nucleophilic sites in DNA, and thus become potent mutagens and carcinogens.56 The sulfate conjugation of phenol is a major metabolite of benzene, and the ratio of organic to inorganic sulfate in the urine was used at one time as a crude biologic indicator of benzene exposure in the occupational environment. However, much better biomarkers of benzene exposure are now available.57

Glutathione conjugation Glutathione (γ-glutamylcysteinylglycine) is an intracellular tripeptide that provides a number of critical physiologic functions and is the most important intracellular antioxidant. In addition to its important physiologic functions, it serves as the co-factor for a multigene family of cytosolic enzymes, the glutathione S-transferases, which catalyze the conjugation of a variety of exogenous substances to glutathione.58 The best studied, and perhaps toxicologically most important, of these reactions is the conjugation of alipathic and aromatic epoxides. Glutathione S-transferases can also catalyze the addition of glutathione directly across certain unsaturated aliphatic sites, such as with diethylmaleate, and can also facilitate substitution reactions on certain halogenated organic compounds, with replacement of one halogen atom with glutathione. Like the cytochrome P450 family of enzymes, the glutathione S-transferases contain a relatively large number of different enzyme forms that are the products of separate genes. There are at least 13 different isoenzymes that have been characterized in humans, with some forms exhibiting tissue-specific expression. The different enzymes have broad and overlapping substrate affinities but are readily classified into one of six families: alpha (hGSTA), mu (hGSTM), pi (hGSTP), theta (hGSTT), zeta (hGSTZ) and omega (hGSTO).58 Glutathione S-transferase P1 is highly expressed in many neoplastic tissues, and its expression is frequently used as an early marker of neoplastic change in experimental carcinogenesis. There is also a microsomal form of glutathione S-transferase, but its function in the biotransformation of xenobiotics appears limited, relative to the various cytosolic forms.59 Several GSTs are polymorphic in the human population, and have been the subject of considerable investigation as possible ‘environmental susceptibility genes’ due to their involvement in a wide variety of detoxification reactions.60 Glutathione conjugates are seldom excreted directly in the urine, although they may appear in the bile as intact glutathione conjugates. Most glutathione conjugates are

Distribution of Chemicals in the Body 105 sequentially metabolized by (1) γ-glutamyltranspeptidase (GGT), which removes the N-terminal glutamic acid residue from cysteine; (2) non-specific peptidases, which cleave the peptide bond between cysteine and the C-terminal glycine to generate a cysteine conjugate; and (3) Nacetylation of the cysteine conjugate to form the N-acetyl cysteine (mercapturic acid) conjugate. Mercapturic acid conjugates are then readily excreted in the urine. Although glutathione conjugation is an important detoxification pathway, recent studies have demonstrated that certain haloalkanes, such as ethylene dibromide, when conjugated to glutathione, can rearrange to form highly reactive episulfonium ions.61 Such glutathione metabolites may be at least partially responsible for the nephrotoxicity and carcinogenicity inherent in some of these compounds.

otics thus occurs primarily in circumstances in which the xenobiotic mimics some endogenous substrate. Examples where this occurs are the methylation of pyridine and catechols. The co-factor for methyltransferase reactions is S-adenosylmethionine. A few examples where methylation of xenobiotics is important include the methylation of dopamine and related catechol drugs by catechol-Omethyl transferase (COMT) and the methylation of the chemotherapeutic agent, 6-mercaptopurine by thiopurine methyltransferase (TPMT).66 The presence of a single nucleotide polymorphism in TPMT is an important consideration in treatment of leukemia patients with TPMT because those who are homozygous for the variant allele may suffer serious toxicity at standard therapeutic doses. Conversely, for those who are homozygous for the high activity allele, a standard dose may not be efficacious because of rapid systemic clearance of the drug.67

N-acetyl transferases (NATs) Many aromatic amines are metabolized primarily by conjugation of the primary or secondary amine with an acetyl group.62 Two different genes for NAT are commonly expressed in human tissues, referred to as NAT1 and NAT2. Because most primary aromatic amines are in the positively charged state at physiologic pH, this conjugation can actually result in products that are not appreciably more water soluble and, in some instances, may be less water soluble than the parent compound. Nevertheless, this pathway is a major route of biotransformation of many aromatic amines, such as hydralazine, isonicotinic acid hydrazide, aniline, 2,6-dinitrotoluene, and some sulfonamide drugs. Both NAT1 and NAT2 are polymorphic in the human population, and have been the subject of considerable study as potential determinants of adverse drug reactions and as environmental susceptibility genes.63 N-acetylation results in loss of pharmacologic activity of some widely used drugs and, as such, is a detoxification pathway. Occasionally, N-acetylated aromatic amines can undergo further metabolism by N-hydroxylation to yield the N-hydroxy-N-acetyl derivative. Both NAT1 and NAT2 are capable of catalyzing this ‘activation’ reaction. A different enzyme, called arylhydroxamic acid N,O-acyltransferase, catalyzes the transfer of the acetyl moiety from aromatic nitrogen to the hydroxyl group on the nitrogen, resulting in the formation of a highly unstable acyloxyarylamine that can react with DNA and proteins.64 This complicated biotransformation pathway is implicated in the carcinogenesis of numerous aromatic amine compounds, including benzidine.65

Methyl transferases The methylation of xenobiotics is not a frequent route of elimination. In contrast to most other biotransformation pathways, products of methylation are almost always less polar than the substrate. Thus, methylation reactions do little to enhance the elimination of non-polar compounds from the body. Methylation of endogenous proteins, nucleic acids, and catecholamines occurs frequently, and is an important biochemical process for the regulation of many intracellular functions. The methylation of xenobi-

Amino acid conjugation Organic cyclic carboxylic acids can undergo conjugation with several amino acids, including glycine, glutamine, and taurine.68 For example, benzoic acid is rapidly conjugated with glycine to form hippuric acid. Because benzoic acid is the primary oxidative metabolite of toluene, hippuric acid is a primary urinary metabolite of toluene, and methylhippuric acids are primary urinary metabolites of xylenes. Both taurine and glycine are used biologically for the conjugation and biliary excretion of primary and secondary bile acids. Taurocholate, glycocholate, taurochenodeoxycholate, and glycochenodeoxycholate are the primary bile acids in human bile, and conjugation of newly synthesized bile acids in the liver is a prerequisite for biliary excretion.

Genetic polymorphisms in xenobiotic biotransformation As noted throughout the previous discussion, several genetic polymorphisms in biotransformation pathways have been described. One of the best studied genetic polymorphisms is that associated with N-acetylation of various primary aromatic amines.65 A bimodal distribution in N-acetylation activity has been described, with approximately 50-70% of caucasians having a slow acetylation phenotype, with only 10-15% of Japanese demonstrating this phenotype. The slow-acetylator phenotype has been associated with a high incidence of adverse drug reactions at normal therapeutic doses of isoniazid, hydralazine, procainamide, dapsone, and some sulfa drugs. In addition, a number of occupationally important arylamines, including naphthylamine, benzidine, 4-aminobiphenyl, and 4-nitrobiphenyl, are detoxified by N-acetylation. In occupations in which exposure to certain carcinogenic arylamines occurs (e.g., dyestuff workers), epidemiologic evidence has suggested that slow acetylators are at a slightly increased risk for bladder and perhaps a few other types of cancer.69 Genetic polymorphisms in several P450 cytochromes have been described in humans, and may have implica-

106 Toxicology tions for individualized drug therapy.70 For example, the alicyclic hydroxylation of the antihypertensive drug debrisoquin is mediated via human CYP2D6, which is polymorphically distributed, with 7-10% of the population characterized as extremely poor metabolizers. The differences in activity between poor metabolizers and the rest of the population are remarkable, with an approximately 20,000-fold difference in debrisoquin hydroxylation. This is a single-gene polymorphism, but the enzyme has broad substrate specificity and is important in the oxidation of a variety of other drugs and chemicals. Although several studies have evaluated whether the CYP2D6 ‘slow metabolizer’ phenotype/genotype is associated with increased risk for certain cancers, no clear association has been found.71 Thus, the importance of this genetic polymorphism in environmental and occupational toxicology remains to be established. The cytochrome P450-mediated hydroxylation of polyaromatic hydrocarbons to mutagenic and carcinogenic intermediates is well established, although no clear polymorphic distribution in this activity has been identified in humans. However, some studies have suggested that high inducibility of aryl hydrocarbon hydroxylase activity may be associated with an increased risk of lung cancer in smokers.72 A genetic polymorphism in the M1 form of glutathione S-transferase (hGSTM1) has been extensively studied in human populations as a possible risk factor for cancer and other environmentally related diseases. Approximately 5060% of the population are genetically deficient in hGSTM1, and this deficiency has been associated with a modest increase in the risk of lung73 and bladder74 cancer in numerous studies. Much remains to be learned about the biochemical and molecular mechanisms that underlie important interindividual differences in susceptibility to environmental and occupational pollutants. The study of the relationship between genetic factors and individual susceptibility to drugs and chemicals is the basis for the rapidly emerging field of ecogenetics.75 Recent advances in molecular biology now provide the experimental tools necessary to identify important genetic differences in individuals, and the identification of variant alleles in the human population has grown at a remarkable rate.76 Application of these tools to toxicology, epidemiology and other areas of public health will be critical to the rational evaluation of individual risk from chemicals in the workplace and general environment, but will also raise important ethical, legal and social issues that must be addressed in lockstep with the science.77

Excretory pathways The ability of the body to rid itself of exogenous chemicals is largely dependent on the physicochemical characteristics of the chemical. Chemicals that have very low blood-gas partition coefficients (e.g., are poorly soluble in blood and have a high vapor pressure) may be effectively eliminated by exhalation, whereas chemicals that are highly water

soluble will generally be eliminated by excretion into the urine or bile. Because many chemicals of occupational and environmental concern lack either of these characteristics, accumulation in the body is likely to occur unless biotransformation processes alter the chemical to a more readily excretable form. Thus, elimination of a chemical from the body occurs by two processes, direct excretion of the unchanged substance or biotransformation to a different chemical form that may then be excreted as a metabolite.

Urinary excretion The kidney is highly efficient at removing many foreign substances from the blood. However, the extent and rate of urinary excretion are highly dependent on the water solubility of the substance. Most chemicals in the blood are readily filtered by glomeruli; the only exceptions are those chemicals that bind avidly to high molecular weight plasma proteins. The rate at which plasma is filtered through the glomeruli (glomerular filtration rate or GFR) is approximately 125 mL/min in normal adults, and the rate of urine production is less than 1% of this volume. As the volume of glomerular filtrate is reduced by reabsorption of electrolytes, nutrients, and water in the nephron, the filtered xenobiotics are concentrated in the remaining tubular fluid, generating a large driving force for diffusion of the xenobiotics from the tubular lumen to the interstitial space and blood stream. Thus, relatively lipid-soluble chemicals are not effectively eliminated in the urine following glomerular filtration because they readily diffuse back out of the tubular fluid into the blood. However, water-soluble chemicals are incapable of diffusing across the tubular membrane, can thus be concentrated to a very high degree in tubular fluid, and eventually are eliminated in the urine. For weak organic acids and bases, the extent of urinary excretion can be greatly influenced by the pH of the urine. In contrast to blood, where the pH must be maintained in a narrow range, urinary pH can readily vary between 5 and 8, resulting in large potential differences in the fraction of chemical in the ionized form. This knowledge has been put to practical therapeutic use in the treatment of poisoning with several weak organic acids or bases. The renal elimination of a weak organic acid with a pKa in the range of normal urinary pH can potentially be increased by 4-6fold by simply alkalinizing the urine with an intravenous infusion of sodium bicarbonate. The renal elimination of salicylates and barbiturates has also been enhanced by alkalinization of the urine. For some organic acids and bases, urinary excretion occurs in part by the presence of specific carrier-mediated transport processes in the proximal tubules.78,79 Such highly efficient active transport systems have been described for a number of weak organic acids and bases and are important in the elimination of certain drugs, such as penicillin, and a variety of glucuronide and sulfate conjugates. Because these carrier-mediated transport processes may be saturated at high doses or may be competitively inhibited by the presence of other substances, for the purposes of risk characterization, it is important to identify

Toxicokinetics 107 whether such processes are functional in the elimination of chemicals. For example, phenoxy acids and several other weak organic acid herbicides appear to be relatively more toxic to dogs than most other species, apparently because the renal organic anion transport system is saturated at relatively low doses, resulting in decreased elimination and somewhat greater rate of accumulation in dogs compared with other species given the same dose.80 In humans, renal elimination of these weak organic acids can be enhanced by alkaline diuresis and/or hemoperfusion.81 For chemicals that enter the tubular lumen exclusively by glomerular filtration and are not reabsorbed across it, the rate of renal elimination from the blood stream equals that of the GFR. For chemicals that are partially reabsorbed or are incompletely filtered because of protein binding, the renal clearance rate may be much less than the GFR. Because the rate of renal plasma flow is about five times higher than the GFR (e.g., only 20% of total renal plasma flow passes through the glomeruli), chemicals that are cleared from the blood by active transport processes can have renal clearance rates much higher than the GFR.

Biliary excretion In contrast to urine, bile is not an ultrafiltrate of plasma, and the biliary tree has very little direct contact with the vascular compartment. Therefore, all substances that enter the bile from the plasma must do so by passing first across the hepatic sinusoidal membrane and then from the hepatocyte across the canalicular membrane to the bile.82 Bile flow is produced by the flux of water from the hepatocyte to the canalicular space in response to an osmotic gradient produced by active transport of bile acids and bicarbonate into the canalicular space. After chemicals enter the hepatocyte, they are generally biotransformed to polar metabolites, which may either re-enter the circulation for elimination in the urine or may be actively transported into the bile. As a general rule of thumb, polar chemicals with molecular weights in excess of about 325 to 350 Daltons will be secreted preferentially in the bile, whereas chemicals of lower molecular weights will be excreted primarily in the urine. The liver possesses several distinct active transport systems for endogenous and exogenous hydrophilic compounds. For example, biliary elimination of anionic compounds, including glutathione S-conjugates, is mediated by a transport protein known as MRP2, whereas bile salts are excreted by a bile salt export pump (BSEP); class I-P-glycoprotein (P-gp) is involved in the secretion of amphiphilic cationic drugs, whereas class II-Pgp is a phospholipid transporter.83 A widely used liver function test, based on the rate of plasma clearance of the blue dye sulfobromophthalein (BSP) or indocyanine green, is in fact a measure of the functional integrity of a hepatic organic anion transport system distinct from that which transports bile acids. Because the removal of these dyes from plasma occurs almost exclusively by hepatobiliary elimination, a decline in the rate of plasma clearance is a useful indicator of hepatic dysfunction. Bilirubin is also excreted in the bile (primarily as the diglucuronide conjugate), by active transport processes,

and a dysfunction in this system will result in elevated plasma bilirubin levels (jaundice; see Chapter 26.1). Indeed, genetic defects in biliary transport processes have led to new insights into how hepatobiliary transport processes function in humans.84 Mutations in specific bile acid or lipid transporters have been identified within specific cholestatic disorders, and genetic polymorphisms have been established for specific diseases.85 Some metals are eliminated in the bile, albeit slowly, and this can be an important route of elimination. For example, methyl mercury is secreted in the bile, but it is largely reabsorbed in the intestinal tract. Interruption of this ‘enterohepatic recirculation’ by the administration of thiol-binding resins greatly reduces the half-life of methyl mercury and is a useful therapeutic approach to the treatment of methyl mercury poisoning.86 Some relatively non-polar compounds can also be excreted in the bile, probably by dissolution in biliary micelles. These are formed by the aggregation of bile acids and phospholipids, and are an important means of solubilizing the large quantity of cholesterol that is normally secreted in the bile. Too high a ratio of cholesterol to bile acids and phospholipids results in the precipitation of cholesterol as gallstones.87 Because micelles are extremely effective at dissolving non-polar compounds, lipid-soluble xenobiotics can partition into the micelles from the hepatocyte. However, after biliary micelles reach the intestinal tract, the non-polar xenobiotics will be rapidly absorbed. Interruption of the enterohepatic circulation with the non-absorbable anion-exchange resin cholestyramine (used clinically to lower blood lipids) has been used successfully in humans to reduce the half-life and toxicity of the highly lipid-soluble pesticide kepone following very high occupational exposures.88 Hepatic dysfunction from disease or chemical toxicity can alter the kinetics of elimination and, thus, the toxicity of a wide variety of chemicals.89 For example, the adverse effects of some digitalis glycosides can be enhanced in the presence of hepatic disease because hepatobiliary clearance is decreased, although this is a less common cause of digitalis sensitivity than other factors such as poor renal function, hypokalemia or hypothyroidism.90 Some drugs that induce microsomal enzymes also enhance bile flow and may increase biliary excretion of other xenobiotics. Phenobarbital treatment increases bile flow and enhances the biliary elimination of methyl mercury. The potassiumsparing diuretic, spironolactone, has been shown experimentally to decrease the toxicity of several chemicals, including mercury and cardiac glycosides, by enhancing biliary excretion.

TOXICOKINETICS The absorption, distribution, biotransformation, and excretion of xenobiotics in the body is a dynamic process. The concentration of toxicant at its receptor site is thus dependent on the various rates of reactions that affect absorption, distribution, biotransformation, and excretion. The field of pharmacokinetics developed largely out of a need to under-

108 Toxicology a

One-compartment open model Chemical in environment Slope = κel/2.303

Onecompartment open model

Log conc

κab

κel Time

b

Two-compartment open model Chemical in environment

Central compartment κel Biotransformation, urine, bile, breath

κ1 κ-1

Peripheral compartment

Log conc

κab

β α Time

stand the factors that dictate the determination of efficacious and non-toxic doses of pharmaceuticals. Essentially the same biologic factors also determine the biologic fate of non-drug chemicals in the body; therefore, toxicokinetics and pharmacokinetics are nearly identical disciplines, and toxicologists frequently use the terms interchangeably. The mathematic modeling of the fate of chemicals in biologic systems commonly uses compartmental models and rate constants to reflect various physiologic functions.91 A ‘one-compartment open model’ assumes that a chemical is instantaneously distributed equally throughout the body, and uses the concentration of a chemical in plasma as representative of the concentration throughout the compartment (Fig. 5.12a). For chemicals that more slowly redistribute from the vascular compartment to tissues, a two-compartment model is used (Fig. 5.12b). The central compartment conceptually represents the vascular space and rapidly perfused tissues, whereas the rest of the body represents the peripheral compartment. There is a rate constant for exchange between the two compartments, a rate constant for influx to the central compartment (where absorptive processes are involved), and a rate constant of elimination (kel) from the central compartment. For chemicals that slowly redistribute to ‘deep’ compartments, such as fat and bone, a three-compartment model, with different rate constants between the central compartment and the two peripheral compartments, is sometimes used to explain the very long terminal half-life of some chemicals.

Figure 5.12: One- and two-compartment pharmacokinetic models. A one-compartment model (a) represents the simplest approach to understanding the distribution of chemicals in the body and assumes that the chemical is instantaneously distributed throughout the compartment (body) and that elimination from the compartment occurs by a first-order process (a constant per cent of remaining compound is eliminated per unit time). A two-compartment model (b) takes into account a slower, distribution phase between the central compartment (blood) and peripheral compartments (tissues). Elimination proceeds only from the central compartment, which is in dynamic equilibrium with the peripheral compartments. Absorption rate constants (kab), equilibrium rate constants (k1, k–1), and elimination rate constants (kel) can be determined, which describe these processes under conditions of first-order kinetics.

The volume in which a chemical is dispersed, of course, varies greatly, depending on its solubility in water and fat, whether it binds to proteins in the plasma, or whether it binds to intracellular sites. This so-called apparent volume of distribution is an important pharmacokinetic parameter and is essentially a constant that relates the concentration of a chemical in the plasma to the total amount of chemical in the body. Thus, for chemicals that are tightly bound to plasma proteins, the majority of the chemical is confined to the plasma (central compartment), and the apparent volume of distribution (Vd) is relatively small, perhaps on the order of 5-10 liters. In contrast, for chemicals that are highly lipid soluble or are sequestered in intracellular sites, the Vd may be very large, exceeding the ‘true’ volume of the body by many fold. The apparent Vd is a necessary factor to estimate the ‘body burden’ or total amount of chemical in the body, at any point in time. Thus, the body burden is equal to the concentration of the chemical in the plasma times the Vd. ‘Clearance’ is a term that refers to the ability of the body to ‘clear’ a chemical from the blood and has units of flow rate (e.g., milliliters per minute). Thus, a chemical with a clearance of 50 mL/min is eliminated from 50 mL of blood in 1 minute. Clearance is thus a measure of the overall efficiency of the removal of a chemical from the body. Total body clearance (ClT) can be proportioned to specific pathways of elimination, such as the liver and kidney, such that:

Toxicokinetics 109 a

ClT = ClH + ClR + . . . ,

Elimination of a chemical from the central compartment usually occurs exponentially by a first-order rate process, in which the fraction of chemical removed per unit time remains constant (a constant per cent of chemical is eliminated per unit time, e.g., 10% per minute). A plot of the log of the plasma concentration versus time will result in a straight line (Fig. 5.13). For first-order elimination rates, the plasma half-life (time required for the concentration of a chemical in plasma to decrease by 50%) is proportional to the elimination rate constant, kel, by the following equation: 0.693 T1/2 = –––––– kel If elimination occurs by a process that is saturable (e.g., enzymatic metabolism or carrier-mediated transport), the rate of elimination will follow zero-order kinetics (a constant amount of chemical is eliminated per unit time, e.g., 10 mg/min), regardless of the concentration. The importance of enzyme saturation at high doses to toxicity is illustrated by the hypothetic situation shown in Figure 5.13b. This figure illustrates the hypothetical one-compartment elimination kinetics of two individuals given the same dose of a chemical, in which the elimination pathway in subject A is saturated at plasma concentrations above about 50 μg/mL, but saturation of the elimination pathway in subject B does not occur within the dose range shown. If toxic effects were evident at plasma concentrations above 10 μg/mL in both subjects, the period of time necessary for subject A to reduce the plasma concentrations to below the toxic level is over twice as long as that for subject B. The upper panel shows the plasma concentration versus time curve for the two subjects; the lower panel plots these same data as the log plasma concentration versus time. The linear response for subject B is typical of a first-order rate of elimination from a one-compartment model, with a half-life of 1 hour. The convex shape of the log-plasma concentration versus time curve in subject A is indicative of saturable metabolism or zero-order kinetics. Toxicokinetic modeling of repeated exposure to chemicals in the workplace and/or environment requires the important additional consideration of the period between exposures, or the ‘dosing interval’. If the dosing interval is longer than the elimination half-life, then accumulation of the chemical in the body will occur. Assuming a firstorder rate of elimination at all concentrations, the amount eliminated between doses will eventually become equal to the amount taken in, and no further accumulation will occur. The point in time at which the amount of chemical eliminated equals the amount taken in during a dosing

80 Plasma concentration

ClT = Vd. kel

90

70 60 50 40 A

30 20 B

10 0 b 100

Toxic plasma concentration Plasma concentration

where ClH and ClR represent the rate of hepatic and renal removal, respectively. Clearance is thus related to the apparent Vd and a first-order elimination rate constant, kel, according to the simple equation:

100

10 A

B 1

0 0

1

2

3

4

5 6 7 Time (hours)

8

9 10 11 12

Figure 5.13 First-order and zero-order kinetics of elimination. When the plasma concentration of a chemical is plotted directly with time (a), a straight line is achieved only when elimination processes are saturated (A), and elimination from the plasma occurs via zero-order kinetics (a constant amount of chemical is removed per minute). However, elimination of most toxic substances occurs via first-order kinetics, in which a constant per cent (or fraction) of the dose is eliminated per unit time. When the log of the plasma concentrations is plotted against time (b), a straight line results when elimination is first order (B). Note how much longer chemical A remains in the region of toxic plasma concentration compared with chemical B, even though both have the same first-order elimination rates.

interval is called the ‘steady state’. The average amount of chemical in the body (body burden) at the steady state is described by the following equation: Xss =

(1.44)t1/2D τ

where Xss = the body burden at steady state, t1/2 equals the elimination half-life, D = the amount of the dose at each dosing interval, and τ = the dosing interval (in the same time units as the t1/2). If the apparent Vd for the chemical is known, then the average plasma concentration at steady state can be

110 Toxicology obtained by simply dividing the body burden at steady state (Xss) by Vd. From this relationship, it is evident that repeated daily exposure to a chemical with a half-life of 1 day would result in an accumulation of chemical in the body to an amount about 1.5 times greater than the amount of the daily exposure. For chemicals with very long half-lives, such as many chlorinated polyaromatic compounds (e.g., PCBs, DDT, and dioxins, which may have half-lives greater than several years) and some metals (e.g., Cd), accumulation may occur throughout a lifetime, potentially reaching toxic levels many years after the initial exposure. The relationship between toxicity and body burden is not always straightforward because accumulation may occur in body ‘compartments’, such as adipose tissue or bone, which may not be target organs for the toxicity of the chemical. As long as the chemical remains highly partitioned in non-target organ compartments, its presence may be of little toxicologic consequence. However, if for some reason this site of storage becomes mobilized (e.g., fasting in the case of fat storage of chlorinated hydrocarbons or osteolysis in the case of bone storage of lead), the subsequent release of stored chemical to the circulation and, thus, its redistribution to target organs could result in toxicity many years following the exposure.

Chemical–chemical interactions The evaluation of toxicity for drugs and chemicals often assumes that the exposure to the suspect drug or chemical is isolated. However, this is seldom the case because simultaneous or sequential exposures to multiple drugs and chemicals are commonplace in our society. Interactions of prescription, over-the-counter, and/or recreational drugs with other such drugs, drugs with chemicals in the home or workplace, or with dietary factors can and do occur. There are many well-known adverse interactions, such as the interaction between monoamine oxidase inhibitors prescribed as antidepressants with other pressor drugs (sympathomimetics) and/or with tyramine, which occurs naturally in some foods (e.g., some types of wine and cheese). However, the number of chemicals for which specific interactions have been documented is likely to be only a small fraction of what really occurs. Of course, beneficial chemical–chemical interactions are the basis of antidotal therapy. The number of potential drug–chemical combinations is, by definition, nearly infinite, and it would be impossible to study all such combinations systematically. Nevertheless, clinicians familiar with the basic mechanisms of action of drugs and chemicals can reasonably identify many potential interactions. Particularly because analytic techniques have emerged over the past few decades, the clinical problems seen as a consequence of such interactions have been recognized. Moreover, certain principles have been developed to estimate the likelihood of such problems, often in advance of their actual occurrence. These can be summarized as follows.

1. Interactions between two compounds prior to actual exposure by the host. For example, an acidic drug, when combined with an alkaline compound, may become insoluble and precipitate out of solution, thus becoming unavailable for absorption. Other examples include the high binding affinity of some drugs, such as cholestyramine, for other drugs, thus reducing their bioavailability. 2. Interactions that occur after ingestion but have an impact on the rate or the degree of absorption. Classic examples include antibiotics, such as penicillin, tetracyclines, many barbiturates, dicumarol, and a variety of hydroxide compounds. In these instances, multiple mechanisms may be involved so that the physician managing a given case will do well to consult the listing of such interactions. As a generalization, these types of interactions may serve to minimize therapeutic efficacy but seldom contribute to the enhancement of poisoning. 3. Drug interactions resulting from alterations in metabolism and/or disposition may lead to clinical problems.92 For drugs that are largely bound to proteins, the displacement of as little as 1% or 2% from protein can greatly increase the biologically active compound. An example of this would be the interactions between various cardiac glycosides and dicumarol. Moreover, such displacement also alters the biologic half-life of the displaced drug. In other instances, concomitant administration of drugs, such as phenobarbital, with a wide variety of other drugs produces induction of enzyme systems (e.g., the cytochrome P450 series), with the result that the second drug will have enhanced elimination. In other instances, such as adding erythromycin to the regimen of a patient already taking theophylline, metabolism will be impaired such that the theophylline concentration may reach harmful levels. 4. ‘Competition at the receptor site’, the locus of pharmacologic activity, can occur. Classically, naloxone’s ability to displace a variety of opioid entities is well known; other drugs, such as phenothiazines, tricyclic antidepressants, and antihistamines, exert comparable alterations at other receptor sites within the nervous system. With increasingly sophisticated analytic techniques, still more unanticipated interactions are being uncovered, although many have little or no clinical significance. 5. Alteration of pH of some body fluid and modifying the pharmacologic activity of a second compound. One example is the administration of sodium bicarbonate to treat an overdose of salicylate; the resultant alkalinization of the renal tubular fluid enhances the ionization of filtered salicylate, precludes its resorption, and significantly increases its excretion.

Toxicity Testing and Predictive Toxicology 111

TOXICITY TESTING AND PREDICTIVE TOXICOLOGY Although epidemiology can be an effective tool to identify and characterize chemical risks to humans, a major limitation of human epidemiologic studies is that the damage has been done by the time it is identified. Furthermore, the utility of epidemiology for purposes of ‘predictive’ toxicology is limited by a number of factors, including: (1) the presence of confounding, (2) difficulties in dose and exposure assessment, (3) difficulty in identifying suitable exposed populations, and (4) limited statistical power associated with many environmental and occupational epidemiologic studies. Increased public awareness of chemical hazards, coupled with a few notable and highly publicized and potentially preventable chemical disasters, has stimulated the search for biologically meaningful and relevant tests to predict hazards from chemicals before human exposures occur. Most predictive toxicity tests rely on the use of common laboratory animals, such as rats, mice, and guinea pigs, although bacterial tests systems and cell and tissue cultures have gained widespread use in some types of toxicity testing, most notably mutagenicity. It is beyond the scope of this chapter to describe in detail the design and conduct of the multitude of toxicity tests that are currently available. However, a summary review of the nature and approaches used in toxicity assessment is given. (For a detailed description of these approaches and techniques, the reader is referred to the excellent reference text on this topic by Hayes.93) The fundamental principle that dictates the utility of animal models for predicting human response is that the adverse effects of a chemical on a test organism, when properly qualified, are applicable to humans; that is, laboratory animals are useful biologic surrogates for the human response to toxic substances. This premise underlies all experimental biology and medicine and is not unique to toxicology. The basic premise that laboratory animals are useful surrogates for human responses to toxic substances is well supported by a wealth of scientific data.94 However, there are many circumstances in which the data obtained from animal models may differ from those in humans substantially, both quantitatively and qualitatively, because of mechanistic, pharmacokinetic, and/or pharmacodynamic properties. Thus, basic research into the biochemical and molecular modes of action and the biologic fate of chemicals in experimental animal and human tissues is a critical component of predictive toxicology. Toxicity testing is generally divided into several major categories, based on the duration of exposure and/or the specific endpoint to be measured.

Acute toxicity testing Acute toxicity studies involve the administration of a single dose of chemical to test animals. Acute toxicity tests can be conducted by a variety of routes of administration, including oral, dermal (percutaneous), inhalation, or

parenteral (intravenous, intraperitoneal, or subcutaneous). A common measure of acute toxicity is the LD50, although much additional useful information is generally obtained from acute toxicity studies. In addition to identifying the lethal and sublethal doses, acute toxicity tests provide information on target organs, mode of action, duration, and the reversibility of the non-lethal effects. In product safety evaluations, the acute toxicity test is used primarily to establish appropriate doses for subacute or subchronic studies, where repeated dosing occurs.

Subacute toxicity testing For chemicals that are expected to accumulate in target organs or produce irreversible effects from individual doses, repeated administration of an agent over a period of 2-4 weeks is sometimes used to provide additional dose range information for the design of longer term studies. Additional information on target organs, pharmacokinetics, pharmacodynamics, and the mechanism of action is generated from such studies but does not generally fulfill regulatory requirements for the product safety evaluation of pesticides, food additives, and other chemicals for which routine exposures are likely to occur.

Subchronic toxicity testing Routine toxicologic evaluation of a chemical for potential human health risks almost always requires the inclusion of studies in which the test chemical is administered daily for a period of 90 days.95 Usually four to five doses are selected, the highest dose of which produces overt toxicity and limited mortality at the end of 90 days of dosing and the lowest dose selected approximates the maximum dose that would produce no observable adverse effects (NOAEL) even with 90 days of repeated dosing. For regulatory purposes, the objective of a subchronic bioassay is to identify the NOAEL. Because estimates of acceptable human doses are frequently derived from the animal NOAEL, the care and thoroughness of such studies can have a major impact on the regulatory standards for the workplace and environment. In general, most regulatory agencies establish acceptable levels of exposure by identifying the animal NOAEL (in units of milligrams of chemical per kilogram animal body weight per day), then dividing that value by an arbitrary ‘safety’ or uncertainty factor. Frequently a value of 100 or 1000 is used, depending on the degree of confidence the regulator has in the quality and relevance of the animal data to humans. This approach assumes that a threshold exists in the dose–response relationship; that is, there is some dose below which no response will occur, regardless of the size of the population exposed. The animal NOAEL is an experimental estimate of the threshold dose in that strain of animal, and the safety factor is used to account for possible differences in species sensitivity, differences in response among various human individuals, and for ‘scaling factor’ differences (on a body weight basis, small animals generally require relatively larger doses than large

112 Toxicology 40 35 % Tumors possibly undetected

animals to elicit the same response; the use of body surface area, rather than body weight, is often a better scaling factor). Because dose–response relationships for many toxic effects are steep, a 100-fold uncertainty factor in most circumstances may afford a large degree of protection. However, where there are significant differences in the mechanism of action, pharmacokinetics, or pharmacodynamics between the animal species used for estimation of the NOAEL and humans, the use of a 100-fold uncertainty factor may be significantly underprotective or overprotective. Thus, the establishment of relevant exposure standards for humans depends on a good understanding of the many factors that influence toxic responses, and how these factors may be similar or different between test animals and humans.

30 25 20 15 10 5 0 0

Chronic toxicity testing and carcinogenicity evaluation In circumstances in which a chemical has a very long halflife, it causes irreversible effects at doses well below the lethal dose, and/or it is suspected of being potentially carcinogenic, a chronic bioassay may be warranted. These studies generally involve the administration of the test substance to animals for an entire lifetime, which for rodents is about 2 years. Generally, the route of administration is the one that is most relevant to human exposures. Administration of the test substance in the diet or drinking water is most common, although chronic bioassays using inhalation exposure are occasionally conducted. The latter are very expensive and, technically, much more difficult to conduct. Although chronic bioassays may occasionally be conducted to assess endpoints other than cancer, by far the most frequent purpose for such studies is the assessment of oncogenic (carcinogenic) potential. Most chronic bioassays utilize both sexes of two species (almost always rats and mice), at least two exposure doses, and one unexposed control group, with approximately 40-50 animals per group. Dose selection is a major consideration in the design and conduct of a chronic bioassay, especially if the study is intended to evaluate the oncogenic potential of the chemical.96 For the purposes of quantitative cancer risk assessment, most regulatory agencies assume that all chemicals that increase tumor incidence do so in a non-threshold manner. Although studies in experimental animals have demonstrated that the extent of DNA damage is often proportional to dose at relatively low concentrations, experiments have not been, nor could they be, conducted at the very low lifetime doses that would be associated with cancer risks of less than about 0.1% (1 in 1000). The statistical power of studies that use even maximum experimentally manageable sample sizes (e.g., 50 animals per dose) is extremely limited. For example, consider a study that exposed 50 animals for a lifetime at dose x, with no tumors found in either the control or dose x group (Fig. 5.14). The most that could be said about such a negative experiment is that we would be 95% confident that, if a

100 200 300 400 500 600 700 800 900 1000 Number of animals in test group

Figure 5.14: Statistical power of animal tests - power versus number. The statistical power of an animal test to determine a small positive response is highly dependent on the sample size. This curve tells us that we can be 95% confident that a chemical that tested negative in 100 animals will not produce more than about 4% incidence of tumors in a larger population. Thus, a negative test, even using 100 animals in a single dose, cannot be used to prove that a chemical is not capable of causing a significant incidence of cancer in a larger population exposed to the same dose.

human population equal in sensitivity to the test animals were exposed at dose x for a lifetime, the true incidence of cancer would not be greater than about 8%. Obviously, it is experimentally impossible to demonstrate that a chemical poses no significant cancer risk using doses in the range of human exposures. For example, to detect a statistically significant (P = 0.05) tumor incidence of 1 in 1000 (0.1%), it would require the use of 460,000 animals in each dose group, plus control, assuming that the background incidence of tumors was near zero.97 Thus, the only reasonable alternative is to assume that there is some describable dose–response relationship, test animals at doses high enough to give statistically measurable response levels (tumor incidence above about 5% to 10% in most instances), and then extrapolate the dose–response data down to the doses encountered by humans. Based on this logic, current carcinogenicity testing guidelines require that animal studies used for the quantitative risk assessment of potentially carcinogenic chemicals utilize the maximum tolerated dose (MTD) and some fraction (usually one half or one fourth) of the MTD to characterize tumor dose–response relationships and then extrapolate the dose–response data to the very low doses associated with ‘acceptable’ lifetime cancer incidence (e.g., one additional lifetime cancer per million exposed individuals). Thus, it is not uncommon to find that cancer risk estimates are based on extrapolation from animal studies with dose–response curves containing only two or three highdose data points to doses four to six orders of magnitude below those in the measured response range. Figure 5.15 shows the process of extrapolation of observed animal data to the very low doses generally deemed necessary to

Toxicity Testing and Predictive Toxicology 113 a 100

Aflatoxin

% Animals with tumors

90 80 70 Dimethylnitrosamine

60 50 40 30

Nitriloacetic acid

20

Vinyl chloride

10 0 10-1

100

101

102

103

b 100

Cancer incidence

10-1 10-2 10-3

?

10-4 10-5 10-6 10-4

10-3

10-2

10-1 100 Relative dose

101

102

103

Figure 5.15: dose–response extrapolation to low doses. The upper panel represents the log-dose–response curve for four different chemical carcinogens. The lower panel represents the same data plotted on a log-log scale to demonstrate the extent of extrapolation typical of most quantitative risk assessments. The dose data for each carcinogen have been normalized by setting the lowest dose of each carcinogen that gave a positive result to 1, and subsequent doses as a multiplier of that dose. Actual dose ranges and routes of exposure for the four carcinogens were vinyl chloride, 50-6000 ppm (inhalation); aflatoxin, 1-100 ppb (diet); nitrilotriacetic acid, 7500-20,000 ppm (diet); dimethylnitrosamine, 5-20 ppm (diet). (From Amdur MO, Doull J, Klaassen CD. Casarrett and Doull’s Toxicology: the basic science of poisons, 4th edn. New York: Pergamon Press, 1991.)

protect public health. Regulatory agencies frequently use additional lifetime risk of 10–4–10–6 (1 in 10,000 to 1 in million) as the level of socially acceptable risk. Because of the uncertainty in the actual shape of the dose–response curve at very low doses, a great deal of potential error is introduced in modeling and extrapolation from the high-dose animal data to the low doses of environmental concern. This uncertainty is aggravated by the realization that high doses may produce tissue toxicity that does not occur at low doses, and that this toxicity in itself may produce a strong promotional effect by stimulating tissue repair and cell division.

The selection of a mathematic model to extrapolate the observed animal dose–response data to the very low dose levels encountered by humans can have a large effect on the projected acceptable risk level. Although numerous mathematic models have been proposed, there are few scientific data that would allow one to support the use of one model over another. In addition to model selection, there are many other areas of uncertainty in the quantitative risk assessment process, such as the selection of animal data studies when more than one is available, the type of tumor responses to be used (e.g., total tumors or malignant tumors only), exposure assumptions, and human dose estimations. The current practice of making conservative assumptions (e.g., assumptions that tend to overestimate risk) when uncertainties exist is often justified as necessary to ensure that risks are not underestimated. Although the accuracy of risk projections based on these methods has limitations, and may overestimate risk, the process provides a useful means of comparing the relative magnitude of potential cancer risks posed by different chemicals, and is a useful addition to other scientific information necessary for the establishment of regulatory guidelines that are adequately protective of public health. There is currently much controversy over the use of high-dose animal testing in cancer risk assessment. There is little question that the quantitative interpretation of the apparent magnitude of risk can be significantly altered by relatively small changes in the slope of the dose–response curve in the measured region, which can result from tissue damage and/or saturation of biotransformation pathways likely to occur at doses near the MTD. Consideration of the differences in pharmacokinetics and pharmacodynamics at high doses versus low doses will substantially enhance the reliability of predictive toxicology tests and represents a very active area of research in toxicology.98

Mutagenicity testing There are numerous different tests commonly used to assess the mutagenic ability of chemicals.99 There are two fundamentally important types of mutations, i.e., mutations in somatic cells and mutations in germinal cells. Mutations in somatic cells are not passed from generation to generation but may be associated with the development of cancer in the mutated somatic tissue. In contrast, germinal mutations may not express themselves as overt toxicity in the host but can be passed on to offspring in the form of a heritable genetic alteration. Most chemicals encountered in the workplace and general environment are not sufficiently chemically reactive to interact directly with DNA, but they may be oxidized by enzymes in the liver and other tissues to highly electrophilic intermediates that covalently bind with nucleophilic sites in DNA. Because the interaction of electrophilic chemicals with nucleophilic sites in DNA is largely independent of the source and type of DNA, simple organisms such as bacteria, coupled with mammalian tissue fractions containing the enzymes necessary for activation,

114 Toxicology are commonly used. Thus, in contrast to most other predictive toxicity tests, mutagenicity assays most frequently utilize responses to bacteria or cells in culture. The most widely used method for determining the mutagenic potential of a chemical is the Ames Salmonella mutagenicity assay. This test employs cultures of S. typhimurium bacteria that have been genetically altered from the wild type. These bacteria have a mutation in the gene that normally functions to synthesize the essential amino acid histidine. Because this gene is defective, the bacteria are unable to synthesize their own histidine (termed His), and thus, they cannot grow in a histidinedeficient medium. However, the genetic alteration responsible for the histidine deficiency is readily altered back to the wild type (a ‘reverse’ mutation) in the presence of genotoxic chemicals. In addition to the His- mutation, these strains of bacteria have additional mutations that enhance their responsiveness to mutagenic chemicals, such as genetic alterations that reduce the efficiency of DNA repair and increase the cell coat permeability to exogenous chemicals. Thus, one can plate millions of bacteria on a medium deficient in histidine, add a potentially mutagenic test chemical and the mammalian enzymes necessary for metabolic activation to the culture plates, and culture the bacteria for 48 hours. Readily visible bacterial colonies will form on the plate only where individual bacteria have been mutated back to the wild type. The number of ‘revertant’ colonies formed per unit of mutagenic chemical is thus an index of the mutagenicity of the chemical. Generally, the 9000 g supernatant fraction (S9 fraction) of a rat liver homogenate is added to the culture plates as the source of the biotransformation enzymes. Because millions of bacteria can be used in a single plate, only a very small fraction of bacteria actually have to be mutated at the His- locus to produce a readily measurable response. This simple and relatively inexpensive test has been used to screen thousands of different chemicals for mutagenic potential. There are many other assays that employ other types of bacteria, yeast, and mammalian cells in culture to identify the mutagenic potential of chemicals.99 The Ames test and other similar bacterial assays generally identify point mutations or ‘microlesions’, e.g., single base-pair substitutions or frame-shift mutations (deletions or additions). There are also a number of in-vitro and in-vivo tests that have been developed to assess major alterations in DNA, sometimes referred to as ‘macrolesions’. For example, incorporation of radiolabeled thymidine into high molecular weight DNA (unscheduled DNA synthesis) can be readily determined as a measure of DNA damage following exposure of normally quiescent mammalian cells in culture to a mutagenic chemical. Because the extent of DNA damage is reflected by the rate of DNA repair and thymidine incorporation, this test can be used to assess widespread but non-specific damage to DNA. Mammalian cells in culture are frequently used to identify chromosomal aberrations such as ‘sister chromatid exchanges’, chromosomal rearrangements or deletions, and micronuclei formation.99 Increases in chromosomal aberrations in

circulating mononuclear leukocytes from occupationally exposed populations have been used as an in-vivo indicator of human exposure to mutagenic chemicals.100 However, one should always insist on the use of appropriate controls because the association between exposure and effect can only be made when an appropriate non-exposed population is used for comparison. Although short-term mutagenicity assays are extremely valuable in assessing the potential mutagenic activity of a chemical, the quantitative interpretation of such assays to human health risks has limitations. Generally, a chemical is tested in a variety of different mutagenicity test systems before qualitative judgments about the potential genotoxicity to humans are made. Chemicals that test positively in multiple different test systems are likely to be mutagenic and, thus, potentially carcinogenic in humans, whereas the mutagenicity of chemicals that test positively in only one or two assays but test negatively in several other tests may be of questionable relevance to humans. Short-term mutagenicity assays are frequently used to assess carcinogenic risk. Although the qualitative relationship between mutagenicity and carcinogenicity is significant, it is not absolute. There are chemicals that may test positively in short-term mutagenicity assays that do not present a significant carcinogenic risk to humans (falsepositive result relative to cancer risk) because of pharmacokinetic and/or pharmacodynamic factors. Similarly, there are chemicals that may pose a carcinogenic threat to humans that test negatively in short-term mutagenicity assays (false-negative result). For example, several inorganic substances, such as arsenic, chromium, and asbestos, test negative in the majority of mutagenicity assays, yet they are known human carcinogens. In general, all chemicals that modify the carcinogenic response by epigenetic mechanisms (e.g., promoters, co-carcinogens, and immune suppressants) may be important human carcinogens, even though they are not mutagenic. Conversely, the identification of chemicals that are mutagenic provides supportive, but not confirmatory, evidence of potential human carcinogenicity.

Reproductive and developmental toxicity tests Well-defined experimental protocols have been developed for the assessment of the reproductive and developmental effects of chemicals in both male101 and female102 test animals. These tests often involve the exposure of laboratory animals throughout gestation, weaning, and early reproductive life. For example, a typical three-generation study (Fig. 5.16) to determine the effects of chemicals on reproductive capacity involves continuous exposure of both male and female animals (Fo, or Parental), generally by the diet or drinking water route, after weaning. Males are generally exposed for 8-11 weeks, and females for 2 weeks, prior to mating. The animals are then mated, reproductive success is evaluated, and the offspring (F1A generation) are examined by autopsy at weaning for malformations. The parental generation (F0) continues to

Toxicity Testing and Predictive Toxicology 115 Fo (P1) Male and female animals continuous treatment -Males: 8 to 11 weeks premating -Females: 2 weeks premating Mated

F1a Follow to weaning

F2c 'Teratology'

F 1b

F1x

Mated

F2a

F2b

F2c

Mated

F3a

F3b

F3c Treatment

be exposed and mated a second time after 1 to 2 weeks to produce the F1B generation. These offspring are then continued on the study, receiving potential exposure by lactation from the exposed dams, and then are switched to the exposure diet at weaning. Selected parents from this generation (F1) are continued on the study and mated twice in the same manner, and the offspring (F2A and F2B) are treated as described previously, such that the cycle is repeated to a third generation (F3A, F3B). At one or more of the mating points in the three-generation study, one half of the pregnant female animals are killed and evaluated for the number and distribution of embryos, early implantation sites, and resorptions. These comprehensive studies provide useful information about the potential effects of the test chemical on male and female reproductive capacity and teratogenic effects. If abnormal responses are obtained in either category, more specialized tests can be performed to evaluate male and female reproductive effects individually and/or evaluate the teratogenic effects by limiting the exposure to specified portions of gestation at doses that are not associated with maternal toxicity. Interpretation of reproductive and developmental tests must include an assessment of 0 relationships, the presence or absence of frank histopathologic damage to the reproductive organs, and evidence of maternal toxicity in which teratogenic effects have been identified. Because of the well-known occurrence of significant species differences in response to teratogenic agents, information on the mechanism of action, pharmacokinetics, and pharmacodynamics are again critical to the rational extrapolation of laboratory animal data to human health risks.

Figure 5.16: Three-generation reproduction study. Generations and time intervals involved in a threegeneration study of effects on the reproduction process in rats. Adapted from Christian MS. Test methods for assessing female reproductive and developmental toxicology. In: Hayes AW, ed. Principles and methods of toxicology, 4th edn. Philadelphia: Taylor & Francis, 2001;1301–80.

Specialized tests to assess other forms of toxicity In addition to the basic toxicity testing discussed, numerous other special tests have been developed to assess ocular toxicity,103 dermal toxicity,104 neurobehavioral effects,105 and immunologic alterations.106 Ocular and dermal toxicity have classically been determined by direct application of the materials to the cornea or skin, usually using rabbits or guinea pigs as test animals. The ‘Draize’ test was originally developed in 1944 to identify human eye irritants using rabbits. Although this test is sometimes a regulatory requirement prior to manufacture of products that might result in eye contact, concerns over animal welfare have stimulated the search for better ways to assess the eye irritation potential of chemicals. Modifications of the procedure in recent years have reduced both the number of animals necessary and the level of discomfort, and new invitro procedures using corneal cells in culture and other similar approaches are now being used to screen for ocular irritancy potential. The shaved surface of the guinea pig is frequently used to assess dermal irritation. A variety of protocols and methods of interpretation are available that attempt to quantify dermal irritation, delayed hypersensitivity, and other types of dermatoxicity. These tests generally involve the direct application of material to the shaved surface of the animal, followed by occlusion of the site with an impervious material for a defined period. A scoring system for erythema, eschar formation, and edema is then used to provide an overall score of dermal irritation. The assessment of allergic reactions, such as delayed hypersensitivity, usually involves the application of the test

116 Toxicology Carcinogen Vinyl chloride Aflatoxin Dimethylnitrosamine Nitrilotriacetic acid

One-hit 1 1 1 1

Linearized multistage

Weibull

1 20 600 10

1 × 10 1000 600 30,000

(0.03) (0.07) (0.04) (50 or 3 μg/g creatinine, urine β2M > 300 μg/g creatinine, or blood cadmium > 5 μg/L whole blood. Medical removal of workers may be required when urine cadmium levels are > 7 μg/g creatinine, β2M > 750 μg/g creatinine, or blood cadmium > 10 μg/L whole blood. The complete OSHA standard and monitoring protocol should be consulted and utilized to guide individual patient care.30,31 Abnormal proteinuria is rarely observed when the urine cadmium concentration is below 5–10 μg/g creatinine, but urine cadmium levels at or above this threshold are associated with a higher risk for renal tubular dysfunction. The risk of renal involvement increases progressively with higher urine cadmium levels. It has been well established that the nephrotoxic effects of cadmium can be irreversible and can progress even after cessation of exposure. Periods of follow-up extending for 5 or more years after workers have been removed from exposure because of increased urine β2M levels have demonstrated significant average increases in urine proteins and serum creatinine, as well as progressive nephrocalcinosis. Nevertheless, exposure cessation may delay the development of clinically significant abnormalities in these parameters. The natural history of cadmium toxicity is characterized by gradual progression of proteinuria with renal failure or osteomalacia described only in isolated cases among occupationally exposed workers. Environmental studies have generally demonstrated progressive but persistent subclinical tubular damage, manifested by increasing excretion of tubular proteins and enzymes. In contrast, urinary calculi are not uncommon among long-term cadmium-exposed workers; the prevalence is undetermined but was reported to be as high as 44% in one series. The dramatic Itai-Itai disease, which affected individuals who consumed food contaminated by cadmium in one region of Japan, is characterized by painful bone disease, osteomalacia, and pseudofractures attributed to disordered calcium, phosphate, and vitamin D metabolism caused or influenced by the cadmium-induced renal damage. Renal pathologic findings in advanced cases have included frank kidney contraction, tubular atrophy and dilation, interstitial fibrosis, with relative sparing of glomeruli at the microscopic level. Varying degrees of proteinuria occur with some cases progressing to renal failure. The significance of minor signs of renal dysfunction attributable to cadmium may have greater significance in the context of other diseases. For example, one study of individuals residing in areas of Belgium with relatively high levels of environmental cadmium contamination found that those with diabetes were more susceptible to

Disorders of the Kidney and Urinary Tract 579 chronic cadmium nephrotoxicity. Other groups at potentially higher risk include women, the elderly, smokers, and those with iron deficiency. Historic cohort mortality studies of cadmium-exposed working populations have yielded mixed results in studying end-stage renal disease as a cause of death. At least two studies of worker cohorts have reported greater than expected numbers of deaths from renal disease, with insufficient power to demonstrate statistical significance. Two larger mortality studies found no cadmium-associated risk after adjustment for potential confounding factors, but a nested case-control study conducted within the combined cohorts from those two studies found a two-fold increased risk of death from nephritis and nephrosis among workers with relatively high career exposure to cadmium. Several studies have demonstrated increased all-cause mortality in environmentally exposed subjects, with evidence of cadmium-induced kidney damage at baseline; however, the proportion of these deaths that is attributable to renal disease is not known.29,32 There is no established treatment for cadmium-induced tubular nephropathy beyond removal from further exposure. Reductions in urine protein excretion have been described when individuals with mild and presumably early dysfunction reduce or cease their exposure to cadmium. One retrospective study found that workers whose urine β2M levels were less than 1000 μg/g creatinine and whose urine cadmium levels never exceeded 20 μg/g creatinine when exposure was reduced or stopped, showed evidence of reversible toxicity.3 The long-term significance of this initial improvement is questionable in light of multiple studies describing slow progression of dysfunction. It has also been reported that calcium and vitamin D replacement can slow the progression of associated osteomalacia, but such treatment is probably contraindicated other than in severe cases because of the risk for nephrocalcinosis. Chelating agents have no demonstrated effectiveness for the condition.11

Lead.

Lead remains a widely used metal, and toxic exposures still occur commonly in both occupational and general environments. Chronic renal failure and the pathologic manifestations of end-stage renal disease were identified as late manifestations of chronic occupational lead exposure during the 19th and early 20th centuries, but with improved control of lead exposures in contemporary workplaces in the developed world, chronic lead nephropathy has become a relatively rare clinical diagnosis. The progression from acute and reversible lead nephropathy to the chronic and irreversible form has been demonstrated reproducibly in rodent models, and there are several lines of human evidence that indicate that lead exposure causes or contributes to chronic renal disease.33 The strongest evidence for the existence of a chronic lead nephropathy comes from clinical and pathologic investigations in Queensland, Australia. At the turn of the last century, chronic nephritis occurred there with excessive frequency among young adults, particularly those with histories of childhood lead poisoning. For at least 10

years before the recognized increase in renal disease, acute lead poisoning had been a problem among children in this region who played on and under wooden verandas that were routinely coated with lead-containing paints. The use of lead paint was prohibited there in 1922, and the occurrence of chronic renal disease progressively declined, approaching the rates seen in other provinces by the 1940s and 1950s. A retrospective follow-up study of children hospitalized for plumbism between 1915 and 1935 established the status of 352 (of 401) former patients in 1954 and found that 165 had died by age 40 years or younger, including 107 who had died with causes of death listed as nephritis variants. A 1956 autopsy study, which included 67 Queensland natives who died between the ages of 20 and 50 years with chronic nephritis, found the lead content of skull and rib bones to be twice as high among cases of idiopathic renal disease than among cases with non-renal disease or renal disease of other established causes. A follow-up study of childhood lead poisoning victims from the 1920s and 1930s in Boston, however, did not reveal either any significant predilection for renal disease or premature death. It has been hypothesized that chronic renal disease was averted because these children received chelation therapy.34 Other supportive clinical evidence comes from the historically observed cases of chronic renal failure among moonshine drinkers. In addition to having impaired renal function, drinkers of lead-contaminated alcohol had evidence of high total body lead by EDTA lead mobilization testing; renal biopsy specimens revealed intranuclear inclusion bodies in renal epithelial cells, a characteristic of acute lead nephropathy in children. The other major body of evidence for a chronic leadinduced nephropathy comes from studies of occupationally exposed individuals. One study of 140 deaths among a subcohort of 241 Australian lead smelter workers, who had formerly been acutely lead poisoned, found a five-fold increased risk of death from chronic renal disease in comparison to other lead-exposed, but never acutely poisoned, workers from the same smelter. In addition, three large historic cohort studies of battery or smelter workers exposed to lead demonstrated twice the expected risk for mortality from chronic renal disease (and some hypertension-related diseases), with 20 years of exposure being associated with a four-fold increased risk for death from renal disease. With extended follow-up, however, this excess mortality from non-malignant renal disease became attenuated.35,36 Many epidemiologic studies have documented an association between higher blood lead levels (greater than 60 μg/dL) and elevated serum creatinine in leadexposed workers.37 Low level environmental lead exposures have been associated with elevated blood lead levels and increases in serum creatinine as well, although the magnitude of the rise in creatinine is small and the clinical significance uncertain.38 In general, clinical tests have not been valuable in assessing asymptomatic lead-exposed workers. Crosssectional studies have not found clinical parameters, such as BUN, to be higher than laboratory reference values, or

580 Renal and Bladder Disorders to correlate with historic or laboratory measures of lead exposure. A number of studies have demonstrated increased excretion of tubular proteins such as NAG by patients with elevated blood lead levels, but the clinical significance of these findings remains unclear. A comprehensive clinical evaluation of 140 lead-exposed workers (only five of whom were symptomatic) found 57 who had abnormally high lead excretion (>650 μg) with EDTA chelation challenge and no other disease. Of these 57 workers, 21 had reduced GFRs (less than 87 mL/min/1.73m3 body surface area, by iodothalamate clearance) and comparable reductions in effective renal plasma flow (by PAH clearance), with no obvious alternative explanation for renal dysfunction.39 Only three had elevations of serum creatinine or BUN. Six of 12 asymptomatic subjects with abnormal kidney function underwent renal biopsy, with light microscopic changes demonstrating focal tubular atrophy, interstitial disease, and generally normal glomeruli. Two subjects had some evidence of glomerular sclerosis. Fluorescence microscopy revealed various patterns of immunoglobulin deposition in tubular and glomerular basement membranes in the seven cases so studied. Among eight subjects who had reduced GFRs and who underwent thrice weekly EDTA chelation treatment for 6 to 50 months, four showed 20% or greater improvements in GFR (two worsened and two showed no change), with normalization of EDTA-mobilized lead excretion. The authors characterized the observed treatment response as evidence of probable lead causation and as effective reversal of preclinical lead-induced renal dysfunction. They did not, however, recommend EDTA therapy for established lead nephropathy. In another study, the same investigators also found that the degree of EDTA-mobilized lead excretion correlated highly with increases in serum creatinine among 44 men with gout (including 26 with histories of industrial lead or moonshine exposure), a condition that historically has been associated with lead exposure. The investigators speculated that, not only some proportion of non-specific or idiopathic chronic renal disease cases, but also some cases with an identifiable cause, such as gout or hypertension, may be primarily attributable to chronic lead toxicity.40 Once diagnosed, the treatment for chronic interstitial nephritis attributed to lead is non-specific, other than elimination of further lead exposure. There is no clear evidence that long-term EDTA chelation therapy improves the course of clinically established chronic lead nephropathy. The possibility exists that long-term EDTA administration in renally compromised patients may itself have an adverse renal effect. Wedeen and coworkers recommend that such treatment only be undertaken with careful follow-up and with a clearly defined endpoint, such as normalization of EDTA-challenge response or improvement in renal excretory function within an a priori designated time frame.39

Mercury.

Acute poisoning with mercuric salts, described previously, can result in persistent renal impairment characterized primarily by tubular dysfunction. Workers chronically exposed to mercuric salts have been reported to have increased urinary excretion of certain lysosomal enzymes,

particularly NAG, suggesting the presence of proximal tubular dysfunction or injury. Tubular proteinuria also occurs with Minamata disease, a severe neurotoxic disorder that occurred in an outbreak caused by dietary ingestion of seafood contaminated by methyl mercury. However, despite the generally higher affinity of methyl mercury for renal than neurologic tissues, the proteinuria of Minamata disease was not associated with marked azotemia and renal impairment.

Beryllium. Chronic beryllium exposure is well established as a cause of granulomatous pneumoconiosis (berylliosis) or chronic beryllium disease (CBD) and has also been associated with extrapulmonary involvement; skin and, to lesser degrees, liver and lymph nodes are the most commonly described extrapulmonary sites of beryllium deposition and pathologic changes. There have been reports of pathologic changes in other organs, including the kidneys. Renal pathologic changes consistent with old or scarred granulomas (Schaumann bodies associated with local beryllium deposition) have been reported. Most of the reported cases with renal pathologic changes have described hyperemia and intrarenal calculi. Hypercalcemia and hypercalciuria have also been reported frequently with CBD, and nephrolithiasis has been present in as many as 10% to 30% of cases, with passage of calcium oxalate- (and beryllium-)containing urinary stones.41

Other elements. Cross-sectional epidemiologic studies of chrome platers and of workers with long-term exposures to uranium dust have found slight but statistically significant increases in urine tubular protein excretion. However, no links have been established between chronic exposures to either of these agents and development of clinically significant renal dysfunction. Germanium is sometimes used (in inorganic forms, such as germanium oxide) as a component of folk remedies or health elixirs. There have been at least 13 reported cases of nephropathy associated with regular ingestion of such mixtures over periods spanning 6 to 20 months. The pathologic findings on renal biopsy have uniformly appeared as chronic tubulointerstitial nephritis, typically with proximal or distal tubular degeneration and interstitial fibrosis, and without significant glomerular injury or evidence of immunologic mechanism. The systemic toxicity of germanium in experimental exposures is low, but the kidneys and liver are usually affected.42

Chronic glomerulonephropathy Mercury. Chronic elemental mercury poisoning primarily affects the central nervous system but also may produce proteinuria. There is human and animal evidence that elemental and inorganic mercury are capable of inducing an immunologically mediated glomerular abnormality and proteinuria that occasionally may reach nephrotic proportions. Young children with exposure to mercury ointment may have acrodynia, an idiosyncratic and presumably allergic reaction whose central features are dermatologic, but

Disorders of the Kidney and Urinary Tract 581 which may be associated with proteinuria. There have also been a number of reports of transient proteinuria and overt nephrotic syndrome occurring idiosyncratically among workers with chronic or subacute exposure to metallic or inorganic mercury. The syndrome has occurred with various degrees of mercury exposure but, typically, without other manifestations of mercury toxicity. Most of the reported cases have resolved completely after cessation of exposure. The most frequent pathologic pattern on renal biopsy has been a membranous-like glomerulonephropathy, with deposits of immune complex in the glomerular basement membrane, although normal and other immunofluorescent patterns have also been described. An analogous phenomenon can be induced experimentally in rodents by repeated subcutaneous or intramuscular injection of mercuric chloride or by injection or inhalation of a variety of organic mercury compounds, with glomerular deposition of antiglomerular antibodies. A progressive membranous nephropathy with glomerular deposition of immune complexes ensues, followed by resolution of both immunologic and histopathologic changes. Despite consistent reproducibility in selected genetically susceptible experimental animals, it appears that the mercury-induced proteinuric syndrome occurs rarely in humans. Recent cross-sectional studies of exposed workers have found no increase in urinary albumin, anti-GBM antibodies or other autoantibodies.43,44 Other metal exposures have been reported in association with proteinuria of probable glomerulotoxic origin. As discussed earlier, organic gold salts, used for the treatment of rheumatoid arthritis, can cause hematuria and proteinuria up to nephrotic proportions. A pharmacologic preparation of bismuth (bismuth tartrate), which previously also was used in the treatment of rheumatoid arthritis, reportedly caused nephrotic syndrome. Nickel carbonyl inhalation in experimental rats produced proteinuria.

Silica. There is mounting evidence that silica exposure increases the risk for renal disease. Animals exposed to silica under experimental conditions have developed a variety of nephrotoxic effects, including glomerulosclerosis. Three studies have documented subclinical renal dysfunction in silica-exposed workers as evidenced by increased urinary excretion of both glomerular (albumin and transferrin) and tubular (RBP and NAG) proteins. These abnormalities were found independent of silicosis, and with either short- or long-term exposure. Numerous case reports have described biopsy proven proliferative glomerulonephritis in persons exposed to silica (either with silicosis or without any pulmonary disease), and two recent studies document an increased risk for end-stage renal disease (particularly due to glomerulonephritis) in silica-exposed workers.45,46 In addition, two case-control studies of patients with either rapidly progressive glomerulonephritis and antineutrophil cytoplasmic antibodies (ANCA) consistent with Wegener’s granulomatosis found an increased incidence of silica exposure in these patients compared with controls.47,48 The mechanism

for this putative nephrotoxicity is not well understood and may involve both a direct nephrotoxic effect as well as an immune-mediated mechanism of injury. Supporting the argument for a direct toxic effect is a case report of acute renal failure following acute overexposure to silica without associated immune deposits on renal biopsy. However, the lack of a clear dose–response relationship and the frequent finding of immune deposits on biopsy specimens argue in favor of an immune-mediated mechanism. In all likelihood, both mechanisms of injury may play a role. Silica-related nephrotoxicity may progress to chronic renal failure and even death, despite steroid treatment and hemodialysis. In one reported case, aggressive immunosuppressive therapy was judged to be helpful.

Organic solvents. Organic solvents have been implicated as either causative or contributing agents in the development of a variety of primary glomerular disorders. While this has not been replicated in studies, a number of investigations have described increased levels of selected urine proteins among groups of workers with chronic exposure to solvents. Cohort mortality studies have not revealed clearly increased risk for death from primary glomerular diseases. The main evidence in support of there being a causal association between organic solvents and glomerulonephritis comes from an extensive series of case reports and casecontrol epidemiologic investigations.49 The case reports have generally described instances of anti-GBM glomerulonephritis occurring in the context of either chronic, subchronic, or acutely excessive exposures to a variety of organic solvents in both occupational and domestic settings. The predominance of anti-GBM glomerulonephritis among reported cases has indicated an anti-GBM immunopathogenic mechanism, possibly initiated by interactions between inhaled solvent vapors and pulmonary alveolar basement membrane, which share common antigenic properties. However, the findings of case-control studies indicate that the increased risk for glomerulonephritis associated with solvent exposure is greater than can be explained by anti-GBM glomerulonephritis alone, indicating a broader association with many types of primary glomerulonephritis. At least 15 case-control studies have examined the relationship between solvents and glomerulonephritis, and many have described significantly increased risks, ranging up to nine-fold increases, for solvent exposures that exceed the usual population experience.50 The studies have included all types of primary glomerulonephritis, suggesting either that the estimated risk extends to all types or that an even greater risk may be attached to some subgroup of glomerulonephritis. Two studies that examined single categories of glomerular disease, proliferative in one study and membranous in the other, both found substantially increased risks associated with solvent exposure. The casecontrol studies have been criticized for methodologic limitations, such as the recall bias inherent to retrospective interview studies. However, the credibility of these studies is supported by the consistency and reproducibility of

582 Renal and Bladder Disorders findings across numerous investigations (each having different types and degrees of methodologic limitation), the high magnitude and statistical significance of the risk estimates, and the evidence for a dose–response relationship in each study that examined subgroups defined by time or intensity of exposure.28 The chemical structural diversity of the associated solvent compounds, the widespread prevalence of solvent use, and the relative rarity of primary glomerulonephritis suggest that any pathogenic mechanism involving organic solvents is probably multifactorial or relatively idiosyncratic. Although the risk for glomerulonephritis may be greater with solvent exposure, solvent exposures appear to be involved in only a small proportion of glomerulonephritis cases. The weight of evidence favors any effect of solvents being non-specific and not limited to specific solvent compounds. Immune mechanisms are probably involved, and it is plausible that there is an immunogenetic predisposition. It may be that potentially glomerulotoxic immune factors arise independently of solvent exposure and that solvents interact with those factors or with renal tissue to precipitate or facilitate the development of glomerulonephritis. For example, when anti-GBM antibodies were injected into rabbits, the antibodies adhered to alveolar basement membrane after intratracheal instillation of gasoline, but not after saline instillation. Other than cessation of further exposures, there is no specific treatment for cases of glomerulonephritis associated with organic solvents. The clinical severity of primary glomerulonephritis has been observed to fluctuate adversely with variations in intensity of continued solvent exposure, and it is prudent to recommend avoidance of non-incidental solvent exposure for any individual with glomerulonephritis.

urologic, and systemic conditions. In clinical series, it has been associated most often with diabetes mellitus and, in others, with excessive analgesic use. It is particularly strongly associated with excessive use of phenacetincontaining analgesics, although most, if not all, analgesics and non-steroidal anti-inflammatory agents have the potential to cause this condition in humans and animals. Other medical agents also have been incriminated as possible causes, including phenothiazines, cyclophosphamide (Cytoxan), and radiocontrast agents, and it has been speculated that other drugs and chemicals may also contribute to or cause renal papillary necrosis. Although it has not been demonstrated to occur in humans with occupational exposures to many chemicals, it has been produced in experimental animal exposures using a variety of chemicals that either have been used in industry, are structurally related to industrial chemicals, or are recognized as industrial pollutants. These include methylaniline, tetrahydroquinoline, ethylenimine, bromoethanamine, diphenylamine, diphenylmethyl alcohol, phenylanthranilic acid, aminopyrene, dioxin, and PCBs (specifically, Aroclor 1242). In managing cases of renal papillary necrosis, therefore, it is reasonable at least to consider the possible role of antecedent occupational chemical exposures.

Renal cystic disease

Chronic exposure to carbon disulfide may accelerate atherosclerosis, and epidemiologic studies of chronically exposed workers have demonstrated significantly increased risks for cardiovascular disease. In addition, an increased prevalence of hypertension has been reported among exposed workers. One study of artificial silk workers reported that half of those with 20 or more years exposure were hypertensive. Carbon disulfide-induced renovascular disease has been invoked to explain this observation. This hypothesis is supported by one clinical study that evaluated 26 carbon disulfide-exposed workers, 16 of whom were hypertensive or had a history of hypertension. The effective renal plasma flow (ERPF, as measured by PAH clearance) was significantly lower on average in the entire group in comparison to eight normotensive and non-exposed control subjects. When creatinine clearance was measured in the five subjects with the lowest ERPFs, four had normal creatinine clearances and relatively high filtration ratios (creatinine clearance/ PAH clearance). These findings support the presence of intrarenal vascular changes.

Renal cystic disease (RCD) in humans is most often attributable to heritable conditions, particularly autosomal dominant polycystic kidney disease. It is clear, however, from clinical experience and animal experiments that heritable factors are not solely involved in the development of RCD and that cyst development can occur in the absence of heritable factors. For example, RCD occurs in up to 40% of patients undergoing chronic dialysis therapy, including patients with previously non-cystic kidneys. Chemical factors have been considered as possible causes of this acquired condition. Although certain experimental animal species have been identified as having inherited predispositions to renal cyst development, renal cyst formation can be induced reproducibly and in a dose-related manner by experimental exposures to a variety of drugs and chemicals, independent of identifiable predisposition. Diphenylamine and diphenylthiazole are well characterized experimentally as cystogenic agents. Diphenylamine also produces renal cysts in the offspring of female rats exposed during pregnancy. Cyst formation has also been produced experimentally by lead acetate, diethylhexyl phthalate, dibutyl phthalate, 2,4,5-trichlorphenoxyacetic acid, alloxan, biphenyl, lithium (lithium chloride), and cisplatin, the latter two of which are pharmaceutic agents that have been related adversely to cyst formation in humans. RCD has not been recognized to occur in association with any occupational chemical exposure, but the possibility should be considered in patients with RCD and no family history or obvious alternative explanation.

Renal papillary necrosis

Radiation nephritis

Renal papillary necrosis is an uncommon condition that may occur in association with a variety of intrarenal,

Renal involvement is generally not a direct feature of ionizing radiation exposure produced either by internal

Vascular nephropathy

Disorders of the Kidney and Urinary Tract 583 emitters, which gain entry to the body, or by acute wholebody external radiation (radiation sickness). Renal injury commonly occurs following excessive partial-body external radiation exposure to sites including the retroperitoneum; however, this is described only with therapeutic radiation and usually at cumulative doses exceeding 2300 rads (23 Gy) over a several-week period. Such exposure produces no clinically detectable acute changes other than local skin changes, but it may result in acute radiation nephritis up to 6–12 months later. The acute nephritis usually manifests with hypertension, which in turn, may produce proteinuria, microhematuria and azotemia. It may occur in the setting of cardiac failure and a relatively refractory anemia. The initial renal pathologic injury includes degenerative changes of the tubular epithelial and glomerular endothelial cells and may progress (as the chronic condition) to include extensive atrophic changes and interstitial fibrosis. The mortality rate is generally high, and survivors often have persistent proteinuria or renal insufficiency. Chronic radiation nephritis may develop years after exposure, with or without acute radiation nephritis in the interim. It is characterized by clinical features similar to the acute condition, although the condition is generally slowly progressive and does not show spontaneous improvement. Malignant hypertension and its complications are common, particularly in the chronic syndrome.

Hypertension Lead. Chronic lead exposure is associated with subclinical increases in blood pressure and may also increase the risk for developing hypertension. By the early 20th century, it was generally accepted that hypertension was a clinical manifestation of chronic lead poisoning. However, subsequent controlled epidemiologic studies of lead-exposed working populations have generally not demonstrated substantial increases in either hypertension or mortality from hypertensive diseases. Many authorities now question the existence, or at least the clinical significance, of any inherent hypertensive effect of lead beyond that secondarily attributable to lead-associated chronic renal insufficiency. There is, however, evidence indicating that lead exposure has some degree of hypertensive effect. Each of two large population-based studies in the United States and Great Britain found that higher blood lead levels were significant predictors of hypertension, and that the blood lead level showed a statistically significant, although small, positive association with both systolic and diastolic blood pressure.51,52 Experimental animal studies have demonstrated that dietary lead can increase blood pressure and that lead, even at low levels, may produce changes in the balance of renin and angiotensin. Blood lead appears to make a small proportional contribution to blood pressure in the general population, relative to other risk factors. However, in situations that differ from the general population experience, lead may be a more substantial contributor to blood pressure elevation. For example, a study in the west of Scotland, where plumb-

ing systems commonly contain lead parts, found significantly higher blood lead levels among 130 hypertensive subjects than among matched normotensive subjects. Blood lead levels also were higher among normotensive individuals with elevated serum urea levels. Another study that included 20 men with essential hypertension and normal renal function observed significantly higher blood lead levels among hypertensive subjects than the normotensive controls.53 In each study, occult lead exposure was presumed to be a contributing factor to the development of hypertension. Duration of lead exposure may be more strongly associated with blood pressure than elevations in blood lead levels. Because of this, the EDTA lead mobilization test has been suggested as one potentially helpful test for evaluating selected patients with essential hypertension, but it is not broadly accepted as such. In the absence of other manifestations of lead intoxication, there is no indication for chelation treatment for hypertension associated with increased body burden of lead.

Cadmium. Cadmium has also been linked with hypertension. Some, but not all, animal experiments have demonstrated a cadmium hypertensive effect at levels of exposure below those producing evident renal damage. Studies of industrially and environmentally exposed individuals, however, have generally found no substantial increase in hypertension. In summary, the data regarding a possible cadmium hypertensive effect are equivocal, but this issue has not been addressed as closely for cadmium as for lead.54 Carbon disulfide. Chronic exposure to carbon disulfide is associated with an increased risk for hypertension and hypertensive diseases. The hypertensive effect is presumed to be the result of vascular nephropathy (discussed earlier).

Non-renal urinary tract disorders Urolithiasis The acute toxic nephropathy induced by ethylene glycol is mediated in part by the intratubular concentration and crystallization of calcium oxalate (described earlier). Although it is believed that tubular obstruction by these crystals is the major mechanism by which renal injury is produced, acute oxalate crystal formation is not associated with clinically significant obstruction distally in the urinary tract. There is no reported increase in risk for stone formation among individuals with chronic ethylene glycol exposure in the occupational setting. Urinary calculi have been reported with high frequency in some studies of long-term cadmium-exposed workers, with prevalence reported as high as 44% in one series involving Swedish workers. This risk for stone formation has not been observed in all series of cadmium-exposed workers, however, and the wide differences in prevalence remain unexplained. Hypercalciuria is usually present with cadmium-associated urolithiasis, and it is believed to play

584 Renal and Bladder Disorders a primary role in stone formation. Other possible contributory effects of cadmium include its interference with calcium-binding proteins and calcium deposition in bone. There are a variety of other uncommon or speculative causes of urolithiasis. Chronic beryllium exposure is rarely associated with chronic tubulointerstitial nephritis (described earlier), and this may be accompanied by passage of small calcium oxalate stones, which may contain beryllium. Silica stones are common in domestic grazing animals, particularly in regions where the amount of silicon in forage is high. However, although silicon is a common component of calcium oxalate, uric acid, and phosphate stones in humans, silica-predominant stones occur rarely in humans and are only known to occur with chronic magnesium trisilicate antacid use. It should be noted that, although primary gout may lead to urate nephrolithiasis, there is no identified risk for stone formation with saturnine (lead-induced) gout, even though it may produce hyperuricemia and hyperuricosuria. There is also no evidence that manganese causes urolithiasis, but it may be noteworthy that manganese can interfere with the synthesis of glycosaminoglycans, which have been hypothesized to play inhibitory roles in calcium oxalate crystal growth and aggregation in urine. Bilateral renal calculi and aminoaciduria have been reported with excessive intake of Worcestershire sauce, representing a suspected environmental lithogen. No specific component of the sauce was identified as the possible cause.

Bladder disorders Neurogenic bladder.

A number of pharmacologic agents are capable of causing bladder dysfunction. In the late 1970s, neurogenic bladder associated with the polyurethane foam catalytic agent dimethylaminoproprionitrile (DMAPN) occurred in epidemic proportions at two separate polyurethane manufacturing plants following introduction of the agent, and additional symptomatic cases were later identified at five other plants.55 Although DMAPN had already been used as a catalyst in an acrylamide waterproofing process for more than 10 years without any recognized problem, more than one-half of the exposed workers at the polyurethane plant had symptomatic bladder dysfunction, most often manifested as urinary retention. Of individuals referred for neurologic evaluation, seven of eight reported cases lacked either a detrusor reflex or normal sensation of bladder filling, and three had prolonged sacral latencies, with two of those three having evidence of partial denervation on external anal sphincter examination. Electrophysiologic findings of distal extremity peripheral neuropathy were also present in most members of this selected subgroup. Broad surveys at the affected plants also revealed symptomatic evidence of sexual dysfunction (partial impotence and increased libido) and non-specific symptoms of irritability, insomnia, and headaches. The symptoms of urinary retention occurred as early as 1 week after the first exposure, and persisted without improvement in most cases until the DMAPN exposure ceased. About

one-half of cases improved within 1–2 weeks after the agent was withdrawn, and of the 14% who were still symptomatic after 3 months, nearly all were still symptomatic 2 years later. Several individuals still had objective findings on cystometric and electrophysiologic examinations. Experimental animal studies and isolated reports of human pathologic studies of biopsied sural nerves indicate that the primary site of DMAPN action is probably the neuronal axon. The predominance of genitourinary over extremity dysfunction is unusual for toxic neuropathies and is still not fully explained. Urinary difficulties have also been described in a few cases of neuropathy from acrylamide, another nitrile compound, but these manifestations of autonomic involvement have only appeared late in the course of peripheral neuropathy. Rodents can develop urine retention in less than 12 hours after oral doses of DMAPN, and then bladder function returns to normal within 2 to 3 days. A study of rats exposed separately to equimolar doses of DMAPN, cyanoacetic acid, and dimethylamine found urinary retention in 100%, 50%, and 25%, of animals respectively. The relevance of these additional findings is not established for human exposures.

Cystitis. Acute hemorrhagic cystitis and urinary irritative symptoms (e.g., dysuria and frequent urination) have occurred in humans with exposures to formamidine pesticides, toluidines (methyl aniline homologues), and chlorotoluidines. The conditions have generally cleared promptly after exposure cessation. As an illustration, two individuals had abdominal pain, dysuria, and gross hematuria during the evening after having cleaned a water tank earlier that day. Cystoscopic bladder biopsies demonstrated hemorrhagic cystitis. Hematuria resolved within 2 days and dysuria within about 1 week. Based on the presence of a metabolite of the chlorotoluidine pesticide chlordimeform (chlorphenamidine) in the serum and urine specimens, it was determined that the tank had been used previously to haul chlordimeform, resulting in this acute exposure. Lower urinary tract symptoms have also been reported among jewelers exposed to cadmium in brazing materials. Consideration should always be given to the possibility that hematuria of bladder origin, particularly in instances of suspected chronic chemical exposures, could represent an initial manifestation of malignant or premalignant disease.

References 1. Mutti A. Biologic monitoring in occupational and environmental toxicology. Toxicol Lett 1999; 108:77–89. 2. Stengel B, Watier L, Chouquet C, Cenee S, Philippon C, Hemon D. Influence of renal biomarker variability on the design and interpretation of occupational or environmental studies. Toxicol Lett 1999; 106:69–77. 3. Roels HA, Van Assche FJ, Oversteyns M, De Groof M, Lauwerys RR, Lison D. Reversibility of microproteinuria in cadmium workers with incipient tubular dysfunction after reduction of exposure. Am J Ind Med 1997; 31:645–52. 4. Taylor SA, Chivers ID, Price RG, et al. The assessment of biomarkers to detect nephrotoxicity using an integrated database. Environ Res 1997; 75:23–33.

Disorders of the Kidney and Urinary Tract 585 5. Koffler A, Frieler RM, Massry SG. Acute renal failure due to non-traumatic rhabdomyolysis. Ann Intern Med 1976; 85:23–8. 6. Abuelo JG. Renal failure caused by chemicals, foods, plants, animal venoms, and misuse of drugs. Arch Intern Med 1990; 150:505–10. 7. Fowler BA, Weissberg JB. Arsine poisoning. N Engl J Med 1974; 291:1171–4. 8. Romeo L, Apostoli P, Kovacic M, Brugnone F. Acute arsine intoxication as a consequence of metal burnishing operations. Am J Ind Med 1997; 32:211–6. 9. Djerassi L. Hemolytic crisis in G6PD-deficient individuals in the occupational setting. Int Arch Occup Environ Health 1998; 71(Suppl):S26–28. 10. Prasad GV, Rossi NF. Arsenic intoxication associated with tubulointerstitial nephritis. Am J Kidney Dis 1995; 26:373–6. 11. Kelley C, Sargent DE, Uno JK. Cadmium therapeutic agents. Curr Pharm Des 1999; 5:229–40. 12. Wedeen RP, Qian L. Chromium-induced kidney disease. Environ Health Perspect 1991; 92:71–4. 13. Loghman-Adham M. Aminoaciduria and glycosuria following severe childhood lead poisoning. Pediatr Nephrol 1998; 12:218–21. 14. Zalups RK. Molecular interactions with mercury in the kidney. Pharm Rev 2000; 52:113–43. 15. Graeme KA, Pollack CV. Heavy metal toxicity, Part I: arsenic and mercury. J Emerg Med 1998; 16:45–56. 16. Rosa RM, Brown RS. Acute renal failure associated with heavy metals and organic solvents. In: Brenner BM, Lazarus MJ, eds. Acute renal failure, 2nd edn. Edinburgh: Churchill Livingstone, 1988;353–69. 17. Pavlakis N, Pollock CA, McLean G, Bartrop R. Deliberate overdose of uranium: toxicity and treatment. Nephron 1996; 72:313–7. 18. Zamora ML, Tracy BL, Zielinski JM, Meyerhof DP, Moss MA. Chronic ingestion of uranium in drinking water: a study of kidney bioeffects in humans. Toxicol Sci 1998; 43:68–77. 19. Manno M, Rezzadore M, Grossi M, Sbrana C. Potentiation of occupational carbon tetrachloride toxicity by ethanol abuse. Hum Exp Toxicol 1996; 15:294–300. 20. Verplanke A, Leummens M, Herber R. Occupational exposure to tetrachloroethene and its effects on the kidneys. J Occup Environ Med 1999; 41:11–16. 21. Letz GA, Pond SM, Osterloh JD, Wade RL, Becker CE. Two fatalities after acute occupational exposure to ethylene dibromide. JAMA 1984; 252:2428–31. 22. Jacobsen D, McMartin KE. Methanol and ethylene glycol poisonings: mechanisms of toxicity, clinical course, diagnosis and treatment. Med Toxicol 1986; 1:309–34. 23. Landry JF, Langlois S. Acute exposure to aliphatic hydrocarbons: an unusual cause of acute tubular necrosis. Arch Intern Med 1998; 158:1821–3. 24. Stengel B, Cenee S, Limasset JC, et al. Immunologic and renal markers among photogravure workers exposed to toluene. Scand J Work Environ Health 1998; 24:276–84. 25. Kamijima M, Nakazawa Y, Yamakawa M, et al. Metabolic acidosis and renal tubular injury due to pure toluene inhalation. Arch Environ Health 1994; 49:410–13. 26. Vaziri ND, Ness RL, Fairshter RD, Smith WR, Rosen SM. Nephrotoxicity of paraquat in man. Arch Intern Med 1979; 139:172–4. 27. Jones GM, Vale JA. Mechanisms of toxicity, clinical features, and management of diquat poisoning: a review. J Toxicol Clin Toxicol 2000; 38:123–8. 28. Daniell WE, Couser WG, Rosenstock L. Occupational solvent exposure and glomerulonephritis. JAMA 1988; 259:2280–3. 29. Jarup L, Berglund M, Elinder CG, Nordberg G, Vahter M. Health effects of cadmium exposure – a review of the literature and a risk estimate. Scand J Work Environ Health 1998; 24(Suppl 1):1–51. 30. http://www.osha-slc.gov/OshStd_data/1910_1027.html This site contains a complete description of the OSHA cadmium standard.

31. http://www.osha-slc.gov/dts/osta/oshasoft/gocad2.html This site provides resources to facilitate the implementation of OSHA’s biologic monitoring protocol for cadmium. 32. Nishijo M, Nakagawa H, Morikawa M, et al. Relationship between urinary cadmium and mortality among inhabitants in a cadmium polluted area in Japan. Toxicol Lett 1999; 108:321–7. 33. Loghman-Adman M. Renal effects of environmental and occupational lead exposure. Environ Health Perspect 1997; 105:928–39. 34. Wedeen RP. Nephrotoxicity secondary to environmental agents and heavy metals. In: Schrier RW, Gottschalk CW, eds. Diseases of the kidney, 6th edn. Boston: Little, Brown & Co., 1996;1231–47. 35. Cocco P, Hua F, Boffetta P, et al. Mortality of Italian lead smelter workers. Scand J Work Environ Health 1997; 23:15–23. 36. Steenland K, Selevan S, Landrigan P. The mortality of lead smelter workers: an update. Am J Public Health 1992; 82:1641–4. 37. Chia KS, Jeyaratnam J, Lee J, et al. Lead-induced nephropathy: relationship between various biologic exposure indices and early markers of nephrotoxicity. Am J Ind Med 1995; 27:883–95. 38. Kim R, Rotnitzky A, Sparrow D, Weiss ST, Wager, Hu H. Longitudinal study of low-level lead exposure and impairment of renal function. JAMA 1996; 275:1177–81. 39. Wedeen RP, Malik DK, Batuman V. Detection and treatment of occupational lead nephropathy. Arch Intern Med 1979; 139:53–7. 40. Wedeen RP, D’Haese P, Van de Vyver FL, Verpooten GA, DeBroe ME. Lead nephropathy. Am J Kidney Dis 1986; 8:380–3. 41. Stoeckle SD, Hardy HL, Weber AL. Chronic beryllium disease: long-term follow-up of sixty cases and selective review of the literature. Am J Med 1969; 46:545–61. 42. Hess B, Raisin J, Zimmermann A, Horber F, Bajo S, Wyttenback A, Jaeger P. Tubulointerstitial nephropathy persisting 20 months after discontinuation of chronic intake of germanium lactate citrate. Am J Kidney Dis 1993; 21:548–52. 43. Barregard L. Enestrom S, Ljunghusen O, Wieslander J, Hultman P. A study of autoantibodies and circulating immune complexes in mercury-exposed chloralkali workers. Int Arch Occup Environ Health 1997; 70: 101–6. 44. Ellingsen DG, Barregard L, Gaarder PI, Hultberg B, Kjuus H. Assessment of renal dysfunction in workers previously exposed to mercury vapor at a chloralkali plant. Br J Ind Med 1993; 50:881–7. 45. Rapiti E, Sperati A, Miceli M, et al. End-stage renal disease among ceramic workers exposed to silica. Occup Environ Med 1999; 56: 559–61. 46. Calvert JM, Steenland K, Palu S. End-stage renal disease among silica-exposed gold miners. JAMA 1997; 277:1219-23. 47. Gregorini G, Ferioli A, Donato F, et al. Association between silica exposure and necrotizing crescentic glomerulonephritis with P-ANCA and anti-MPO antibodies: a hospital based case-control study. In: Wolfgang G, ed. Advances in experimental medicine and biology. New York: Plenum Press, 1993. 48. Nuyts GD, VanVlem E, DeVos A, et al. Wegener granulomatosis is associated to exposure to silicon compounds: a case-control study. Nephrol Dial Transplant 1995; 10:1162–5. 49. Hotz P. Occupational hydrocarbon exposure and chronic nephropathy. Toxicology 1994; 90: 163–283. 50. Ravnskov U. Hydrocarbons may worsen renal function in glomerulonephritis: a meta-analysis of the case-control studies. Am J Ind Med 2000; 37:599–606. 51. Schwartz J. The relationship between blood lead and blood pressure in the NHANES-II survey. Environ Health Perspec 1988; 78:15–22.

586 Renal and Bladder Disorders 52. Pocock SJ, Shaper AG, Ashby D, Delves HT, Clayton BE. The relationship between blood lead, blood pressure, stroke, and heart attacks in middle-aged British men. Environ Health Perspec 1988; 78:139–55. 53. Granadillo VA, Tahan JE, Salgado O, et al. The influence of the blood levels of lead, aluminum and vanadium upon the arterial hypertension. Clin Chem Acta 1995; 233:47–59.

54. Staessen JA, Kuznetsova T, Roels HA, Emelianov D, Fagard R. Exposure to cadmium and conventional and ambulatory blood pressures in a prospective population study. Am J Hypertens 2000; 13:146-56. 55. Kreiss K, Wegman DH, Niles CA, Siroky MB, Krane RJ, Feldman RG. Neurological dysfunction of the bladder in workers exposed to dimethylaminoproprionitrile. JAMA 1980; 243:741–5.

Chapter 26 Gastrointestinal Disorders 26.1 Liver Diseases Carl A Brodkin, Stan Lee, Carrie A Redlich Occupational and environmental liver diseases include acute and chronic liver injury caused by workplace or environmental exposure(s). In the first half of the 20th century, a number of case reports and series documented liver injury following exposure to now well-recognized hepatotoxins such as carbon tetrachloride, trinitrotoluene, tetrachloroethane, and yellow phosphorus. The use of such agents has been reduced or eliminated in most workplace settings. However, liver injury associated with exposure to chemicals used in the workplace still occurs and may be underdiagnosed. Hepatotoxic exposures also can be encountered in the environment by contact with contaminated water, air, soil, or food, or with naturally occurring hepatotoxins such as aflatoxins. Occupational and environmental liver injury can be challenging to recognize, diagnose, and manage for several reasons. The incidence, etiologic agents, natural history, diagnostic criteria and management strategies for occupational liver disease have not been fully characterized. Acute toxic liver injury frequently has a non-specific clinical presentation, whereas chronic liver injury usually remains asymptomatic until it reaches an advanced stage. It is unknown how often occupational liver disease goes unrecognized, is attributed to other etiologies, or is misclassified as idiopathic. Once liver disease is recognized, it must be distinguished from non-occupational or non-environmental etiologies on the basis of clinical presentation, exposure history, laboratory findings, and pathologic findings when available. The existence of two common causes of liver disease, namely alcohol and hepatitis viruses, must be considered. Although exposure to alcohol and hepatitis viruses may be more frequent in certain occupations, the majority of liver disease caused by these two entities occurs outside the workplace. However, even the separation between occupational and non-occupational liver disease is not absolute. For example, a study of cirrhosis mortality and occupation in California from 1979 to 1981 found the highest cirrhosis mortality rates among "blue-collar" occupations (construction workers, carpenters, painters, machinists). Notably, these occupations often carry greater exposure to hepatotoxic chemicals that may directly contribute to cirrhosis or potentially interact with alcohol consumption, to increase risk.1 The diagnosis of occupational liver injury is facilitated by human and toxicologic data. The information available is based on the following sources: (1) isolated case reports of liver injury, usually following obvious overexposure; (2) epidemiologic studies demonstrating liver function test

abnormalities or increased incidence of chronic liver disease in certain groups (frequently with multiple, poorly characterized exposures); and (3) animal toxicology studies documenting hepatotoxic responses to specific agents. Because of limited human data, the hepatotoxic potential of a given substance frequently is determined based on animal studies. Such studies often focus on effects of a single acute high exposure rather than on chronic low-level or simultaneous multiple exposures that reflect most occupational exposure conditions. Significant interspecies differences in susceptibility, possibly related to variable hepatic metabolism, also limit the applicability of animal findings to the prediction of human hepatotoxicity. The diagnosis of occupational liver disease is complicated by several host and acquired factors, discussed later, that can potentiate hepatotoxic effects and may result in enhanced or unexpected toxicity. Thus, a high level of suspicion is required to recognize and diagnose liver disease of occupational or environmental origin; the diagnosis need not be dismissed because liver injury associated with a given substance has not been reported previously. This chapter reviews acute and chronic non-malignant occupational and environmental liver injury, including clinical features, known and suspected causative agents, strategies for diagnosis, screening for preclinical disease, and management.

ROLE OF THE LIVER IN METABOLISM The liver is the major site for the biotransformation of exogenous substances such as chemicals, drugs, and dietary factors, and it is a frequent site of toxicity. These exogenous substances, or xenobiotics, are generally lipophilic and not easily cleared from the body. They are metabolized primarily by the cytochrome P-450 mono-oxygenase enzyme system, a large family of related enzymes located in the smooth endoplasmic reticulum of the hepatocyte. There are additional hepatic enzyme systems that metabolize certain alcohols, nitrogen and sulfur compounds, oxides, and other xenobiotics (see Chapter 5). Hepatotoxicity may result from the xenobiotic directly, or from its metabolism to toxic intermediates – often electrophilic compounds or free radicals. Normally these intermediates are detoxified by conjugation or transferase reactions to less toxic, watersoluble compounds and excreted in the urine or bile.2 The toxicity of a given substance may be assessed by the activity of the specific hepatic enzyme system responsible

588 Liver Diseases for its bioactivation and detoxification, as well as by factors affecting the substance’s absorption, storage, and excretion. Significant interindividual variation in these enzyme systems can occur. Such differences are determined by (1) host factors, including genetically inherited polymorphisms, age, and sex, and (2) acquired or exogenous factors, such as medication, ethanol or cigarette use, diet, and preexisting liver disease. All of these factors may contribute to individual differences in susceptibility to exogenous chemicals and drugs. Numerous drugs (including barbiturates, phenytoin, and steroids), foreign chemicals (including organic solvents, polycyclic aromatic hydrocarbons in cigarette smoke, halogenated pesticides, dioxins, and polychlorinated biphenyls [PCBs]), and dietary factors (such as ethanol or vegetable indoles) are potent inducers of the cytochrome P-450 enzymes. Metals, antibiotics, cimetidine, carbon monoxide, ethanol, and altered nutritional status can inhibit the cytochrome P-450 system. The ability of many of these exogenous agents to potentiate hepatotoxicity has been well documented in animal studies and human case reports. In animals, the administration of alcohols (e.g., ethanol, isopropanol), ketones (e.g., acetone), trichloroethylene (TCE), medications (e.g., phenobarbital), and PCBs have been shown to potentiate the hepatotoxicity of carbon tetrachloride and other chlorinated hydrocarbons. In humans, ethanol administration has been shown to increase the toxicity of TCE, carbon tetrachloride, and drugs such as acetaminophen. Recent human exposure studies have shown that ethanol can alter the metabolism of a number of chemicals including styrene, methyl ethyl ketone, and toluene. Similar interactions probably occur with many occupational, environmental, and dietary substances, and these interactions may be important determinants of hepatic toxicity. These findings have an impact on the evaluation of occupational liver disease, because workers are frequently exposed to multiple chemicals, alcohol, medications, cigarettes, and varied diets. Interactions between these factors may potentiate hepatotoxic effects and must be considered in evaluating patients with possible occupational or environmental liver injury.

sic hepatotoxic agents, is illustrative. An etiologic model of alcohol-induced hepatocellular damage involving oxidative stress has been demonstrated in animal studies. In addition to ethanol,3 the microsomal ethanol oxidizing system (MEOS) has a high capacity to oxidize numerous solvent substrates including halogenated compounds such as carbon tetrachloride4 and non-chlorinated solvents including acetone,5 butanol, pentanol, and benzene.6 Potentiation of hepatotoxicity from solvent mixtures has been demonstrated with the induction of microsomal cytochrome P450IIE1 by alcohol consumption in the presence of bromobenzene.7 This solvent inducible enzyme system can produce highly reactive intermediates, leading to peroxidation of lipid membranes8 such as endoplasmic reticulum, and covalent binding to cellular macromolecules (DNA, RNA, and proteins). Support for a mechanism of direct biologic injury includes (1) the anatomic association between selective centrizonal injury and enhanced microsomal P450IIE1 activity in this region,9 and (2) animal models demonstrating marked enhancement of ethanol-induced hepatic injury with carbon tetrachloride, with striking activation of P450IIE1.10 In contrast to intrinsic hepatotoxins, a few hepatotoxins (e.g., beryllium) are idiosyncratic in that they cause liver injury that is sporadic and generally not dose related, possibly by a hypersensitivity or other immunologic reaction. Typically, granulomas or eosinophilic infiltrates are found on liver biopsy. Hepatotoxins can also be classified by the pathologic or clinical syndrome they cause, although exposure to a specific hepatotoxin can result in more than one type of liver injury. For example, acute exposure to the solvent carbon tetrachloride can result in acute necrosis, whereas chronic exposure results in steatosis, fibrosis, and cirrhosis in animals. In Table 26.1.1, occupational and environmental liver disorders are classified based on both their clinical

Type of injury Acute/subacute Acute or subacute hepatic necrosis (± steatosis)

CLASSIFICATION OF OCCUPATIONAL HEPATOTOXINS AND MECHANISMS Hepatotoxins can be classified as either intrinsic or idiosyncratic toxins. Most hepatotoxins are intrinsic toxins, that is, their hepatotoxicity is a predictable property of the substance itself, and most individuals will be affected if the dose is sufficient. Most intrinsic toxins (e.g., carbon tetrachloride) also are classified as direct toxins, namely, the substance or its metabolite directly injures the liver. Acute and subacute injury by such toxins usually produces varying degrees of dose-dependent hepatocellular injury with necrosis and steatosis. The mechanism of hepatic injury related to alcohol and other organic solvents, representing a large class of intrin-

Acute viral hepatitis Acute cholestatic hepatitis Chronic Steatosis Chronic hepatocellular disease (fibrosis/cirrhosis) Chronic hepatitis Granulomatous hepatitis Hepatoportal sclerosis

Selected agents Halogenated aliphatics (CCL4, tetrachloroethane, trichloroethylene, ethylchloroform) Nitro compounds (dimethylformamide, trinitrotoluene, 2-nitropropane) Aromatic hydrocarbons (toluene) Metals (arsenic, lead, phosphorus) Viral agents (hepatitis A, B, C) (see Chapter 22) Methylene dianiline See acute necrosis CCL4 and other chlorinated organic solvents, arsenic, mixed solvents Viral agents (hepatitis B, C) Beryllium, copper Vinyl chloride monomer, arsenic, thorium

Table 26.1.1 Classification of non-malignant occupational liver disease

General Evaluation 589 presentation and pathologic processes, with examples of proven or suspected etiologic agents. Table 26.1.2 lists the major known or suspected occupational and environmental hepatotoxins associated with non-malignant liver injury. The toxicity of certain agents is predicted on the basis of their chemical structure. An agent that has documented hepatotoxicity in animal studies should be considered a possible human hepatotoxin. Hepatotoxic substances may be encountered in a variety of occupations, including construction activities, electronic manufacturing, farming, painting, and textile and dye manufacturing. Epidemiologic studies of workers in such occupations have shown variable evidence of liver injury. Contributing to this variability is the lack of sensi-

Halogenated hydrocarbons Carbon tetrachloride Tetrachloroethane Tetrachloroethylene Trichloroethylene 1,1,1,-Trichloroethane Chloroform Vinyl chloride Anesthetic gases Hydrochlorofluorocarbons (HCFCs) Halothane Methoxyflurane Alcohols Ethyl alcohol Aromatic hydrocarbons Styrene Toluene Xylene Plasticizers Methylenedianiline Nitro compounds Trinitrotoluene Dinitrobenzene 2-Nitropropane N,N-Dimethylformamide Pesticides Organochlorine insecticides Chlordecone Fungicides Hexachlorobenzene Herbicides Paraquat Chlorphenoxy acids (2, 4, 5-T, Agent Orange) Chlorinated aromatic compounds Polychlorinated biphenyls Chloronaphthalenes 2, 3, 7, 8-Tetrachlorodibenzo-p-dioxin (TCDD) Dibenzofurans Metallic compounds Arsenic Beryllium Cadmium Copper Iron Thallium Lead Phosphorus Environmental agents Aflatoxin Amanita phalloides toxin Table 26.1.2 Partial list of known or suspected human hepatotoxins associated with non-malignant liver disease

tivity and specificity of current screening tests for subclinical liver disease and imprecise exposure assessment. Exposure to hepatotoxic substances also can occur in a number of home or environmental settings such as with cleaning agents, paints or paint removers, chemical contaminants, or natural toxins in the environment. Dramatic instances of liver disease associated with massive environmental contamination have been reported, such as cooking oil heavily contaminated with PCBs (Japan, 1968), wheat with hexachlorobenzene (Turkey, 1955 to 1957), and flour with 4,4′-diaminodiphenyl-methane (England, 1965). There is often little information available on the effects of low-level exposure to the numerous natural and man-made substances present in the environment and food chain.

GENERAL EVALUATION Clinical history As with most environmental illnesses, a careful medical, occupational, and exposure history is the key to suspecting and diagnosing work-related or environmental hepatotoxicity. The clinical presentation of acute and chronic toxic liver injury can range from no symptoms to acute nausea, abdominal pain, and jaundice. Occasionally, it presents insidiously as end-stage liver disease. Most hepatotoxins also affect other organs, including the central and peripheral nervous systems, kidney, and mucous membranes. Symptoms related to these organs may predominate, and information on these symptoms should be elicited from the patient and coworkers. Central nervous system symptoms (e.g., euphoria, headaches, and dizziness), mucosal irritation, and disulfiram-like reactions following alcohol ingestion suggest excessive solvent exposure. The temporal relationship between the exposure and onset of symptoms is essential to diagnosing acute hepatotoxicity but is less helpful in diagnosing chronic disorders. Symptoms usually develop within days to weeks following acute solvent exposure, depending on the extent of poisoning. Exposure to solvents, pesticides, and heavy metals should be specifically assessed. Further information about potential hepatotoxic exposures can be obtained from material data safety sheets, the employer, Occupational Safety and Health Administration (OSHA), and industrial hygiene monitoring. A description of the workplace, including the use of protective equipment and ventilation systems, is helpful in assessing the mode (dermal, inhalation, or ingestion) and extent of exposure. The introduction of any new chemicals and the occurrence of unusual or accidental exposures also should be investigated. A prior history of liver and biliary disease, medication and alcohol use, obesity, diabetes, and risk factors for viral hepatitis (blood transfusions, sexual practices, and intravenous drug abuse) should be ascertained to rule out other contributing causes of hepatic injury. Alternative hepatotoxic exposures, such as hobby or other home chemical use, and additional jobs, should also be excluded.

590 Liver Diseases

Physical examination Although typical signs of hepatic injury or dysfunction may be present (right upper quadrant tenderness, hepatosplenomegaly, jaundice), the physical examination is not a sensitive indicator of liver disease. The examination should not be limited to the abdomen; evidence of toxicity to other organ systems, such as mucous membranes and nervous system, and extra-abdominal manifestations of liver disease should be evaluated.

Laboratory and radiographic assessment of hepatotoxicity Tests for evaluating hepatotoxicity fall into four general categories: (1) serum markers of hepatobiliary disease, (2) biochemical tests of liver function, (3) imaging tests, and (4) liver biopsy.

Serum markers of hepatobiliary disease Most useful in the evaluation of hepatotoxicity are markers of hepatocellular necrosis, which reflect the release of intracellular enzymes into serum following cytotoxic injury.11 Enzyme release results from increased hepatocyte membrane permeability. The most important of these tests are aspartate aminotransferase (AST or SGOT), a mitochondrial and cytosolic enzyme, and alanine aminotransferase (ALT or SGPT), a cytosolic enzyme. AST also is present in muscle, heart, and kidney, whereas ALT is more specific for liver. Both markers have been validated in a wide variety of clinical and experimental settings. Severe elevation (> 8–10 times normal) may occur in acute toxic and viral injury. Mild elevation (< 2–3 times normal) is generally seen in chronic or resolving hepatocellular injury or less severe acute injury. ALT activity is reduced in alcohol-related liver disease;12 levels of greater than 300 are uncommon in alcoholic liver injury. The pattern of transaminase elevation also can be helpful in distinguishing alcohol from other hepatotoxic injury. Since alcohol selectively inhibits ALT activity, a high AST/ALT ratio (greater than one) is often seen in this setting.13 In contrast, other toxin- and viralinduced hepatic injury usually results in an AST/ALT ratio of less than one.14 It is important to recognise that subclinical disease can occur despite normal aminotransferases, especially with chronic liver injury. A second group of serum markers include enzymes whose synthesis is induced by cholestasis. The most important of these are alkaline phosphatase (AP) and γ-glutamyl transpeptidase (GGT). The specificity of AP for liver disease is reduced by the presence of other AP isoenzymes in bone, intestine, and placenta. AP is useful in assessing hepatotoxic exposures associated with cholestasis, such as methylenediamine. Although GGT is sensitive for hepatocellular injury, it lacks sufficient specificity to be useful diagnostically. This lack of specificity is due to its presence in several organs (kidney, seminal vesicles, pancreas, spleen, heart, brain, and common bile duct), its inducibility by alcohol and other medications, and its long half-life in serum (26 days).

Biochemical tests of liver function Measurements to assess various functions of the liver include tests of hepatic clearance, tests of hepatic metabolism, and tests of synthetic function.

Tests of hepatic clearance Clearance tests dynamically assess three levels of hepatocyte function simultaneously: uptake, metabolism, and excretion. These tests assess the liver’s ability to clear exogenously introduced or endogenously produced organic anions. Hepatic clearance was traditionally assessed by the sulfobromophthalein sodium test (BSP). BSP is eliminated by biliary excretion following conjugation with glutathione in the liver. The irritative effects of BSP during infusion have limited its clinical utility. Indocyanine green (ICG) dye is excreted without conjugation by the liver, reducing biologic variability due to enzyme induction (e.g., from diet and medications). Increased retention of ICG thus reflects diminished hepatic clearance. The utility of ICG clearance in detecting occupational liver injury has been evaluated primarily in workers exposed to vinyl chloride monomer. ICG clearance is believed to be the most specific test in detecting early liver injury in this population. The utility of the ICG test in other exposure settings is unclear. The test also is limited by cost, availability, and cumbersome administration. Cholic and chenodeoxycholic bile acids, which are included in the bile, are endogenously produced organic anions. Bile acids are cleared exclusively by the liver; thus, their serum concentrations reflect hepatic function, analogous to serum creatinine in the assessment of renal function. Measurement of fasting serum bile acids (SBA) avoids the administration of exogenous agents, and has been advocated as a sensitive screening test for hepatobiliary disease. Conjugated cholic acid (CCA) appears to provide high positive predictive value for hepatobiliary disease.15 Serum bile acids more specifically reflect excretory function than serum bilirubin, and in occupational settings, several studies have revealed increased serum bile acid levels in workers exposed to mixed solvents as well as styrene.16–19 Bilirubin levels may be elevated in cholestasis, severe acute hepatitis, and chronic liver injury; conjugated (direct) bilirubin and direct/total bilirubin may be useful in assessing early hepatic dysfunction on a population basis in solvent-exposed workers.20,21 Tests of hepatic metabolism The hepatic biotransformation system, composed predominantly of cytochrome P-450 (microsomal) enzymes, is the major site of metabolism for exogenous compounds such as drugs or chemicals. Measurement of an exogenous substance’s clearance can be used to assess hepatic metabolic enzyme activity. Changes in metabolism may reflect (1) advanced hepatocellular injury, resulting in reduced enzyme activity, or (2) the induction or inhibition of certain hepatic metabolic enzymes by xenobiotic agents. The metabolism of radiolabeled antipyrine and aminopyrine, both of which can be measured non-invasively, most frequently has been used in research settings. Several studies have reported altered antipyrine metabolism

Acute Occupational and Environmental Liver Disorders 591 in workers exposed to solvents and pesticides. However, the significance of such findings is unclear. The liver and erythropoietic systems are organs of heme synthesis, via porphyrin intermediates. Accumulation of several porphyrin metabolites, which are measurable in urine, can occur with hepatic dysfunction. Coproporphyrinuria has been demonstrated in the setting of alcohol, lead, and hexachlorobenzene exposure. Although urinary porphyrin levels have not been validated clinically, they may provide a useful biologic marker of exposure in the future.

Tests of hepatic synthetic function Hepatic function can be assessed grossly by serum measurement of proteins synthesized and secreted by the liver, such as serum albumin, ferritin, urea, and coagulation factors, measured indirectly by prothrombin time (PT). These tests are limited clinically by poor sensitivity. This poor sensitivity is due to the large functional reserve of the liver; generally more than 90% of hepatic parenchyma must be destroyed before measurable changes in protein synthesis occur.

Imaging tests Imaging studies provide a clinically useful, non-invasive means of hepatic tissue characterization. Ultrasonography compares sonic penetration of hepatic and renal parenchyma: steatosis and fibrosis increase liver echogenicity relative to the kidney. In a prospective, controlled study of 85 patients with biopsy-proven hepatic disease, Saverymuttu observed a sensitivity of 94%, and specificity of 84% in detecting hepatic parenchymal changes.22 A significant exposure–response relationship between sonographic parenchymal changes and hepatotoxic exposure to vinyl chloride monomer and perchloroethylene has also been observed in occupational settings.23–25 Neither ultrasound nor CT distinguishes steatosis from fibrosis with precision. Ultrasound is preferred in the initial evaluation of biochemical abnormalities, to assess biliary as well as parenchymal disease. In patients with chronic hepatic injury, magnetic resonance imaging (MRI) may be useful in distinguishing inflammatory and necrotic changes (patchy enhancement) from fibrosis (linear enhancement) using a gadoliniumenhanced technique.26

Liver biopsy Histopathologic examination offers the most precise assessment of hepatotoxic parenchymal changes, such as hepatocellular injury, inflammation, steatosis, and fibrosis. Thus, it is considered the definitive "gold standard" test for liver disease and can be helpful both diagnostically and prognostically in characterizing the type, extent, and activity of liver injury. In experienced hands, percutaneous liver biopsy is a safe procedure that can be performed on an outpatient basis. However, as with any invasive procedure, it does carry some risks (e.g., hemorrhage, infection). Indications for liver biopsy include persistently elevated aminotransferases of unclear etiology, unexplained hepatomegaly, and anatomic abnormalities. Histologic evidence of hepatocellular injury such as necrosis or regen-

eration and steatosis (microvesicular, macrovesicular, or both), although not unique for toxic liver injury, are suggestive of the condition. Centrizonal (perivenular or Zone 3) distribution of injury within the hepatic lobule is characteristic of solvent-induced injury, though this may be diffuse in more severe cases. Such findings may be subtle, especially if the biopsy is performed weeks to months after the exposure of concern, as frequently occurs, and they are not specific for a given substance. Ultrastructural examination may provide further, or confirmatory, findings suggestive of a toxic injury, such as small droplet fat formation and prominent vesicular endoplasmic reticulum. If environmental liver injury is suspected prior to biopsy, part of the tissue should be fixed in glutaraldehyde, because it enables the evaluation of both fat content and ultrastructure. The lack of certain findings on biopsy, such as Mallory’s hyaline and ground glass hepatocytes, also can be helpful in ruling out alternative diagnoses such as alcohol- or viral-induced injury.

Overview of laboratory tests In summary, of the numerous laboratory and imaging tests available, serum aminotransferases (ALT and AST) are most helpful in the evaluation of patients with hepatocellular injury and in screening exposed populations. However, such tests may not be sensitive enough to detect mild acute injury, and are much less useful in the detection of chronic liver injury. They also are not specific for chemical or occupational-induced injury. Although liver biopsy is useful in evaluating individual cases of suspected hepatotoxicity, it is an invasive procedure that carries cost and some risk for morbidity (and in rare instances, mortality), making it impractical as a screening test. See Table 26.1.3 for further comments regarding these diagnostic tests.

ACUTE OCCUPATIONAL AND ENVIRONMENTAL LIVER DISORDERS Acute and subacute hepatocellular diseases Acute and subacute hepatocellular injuries are the most commonly recognized occupational liver disorders. They include a spectrum of liver injury ranging from acute injury following a single massive hepatotoxic exposure, as occurs in accidental poisoning, to repeated hepatotoxic exposures over days to weeks. Although numerous suspected and known animal hepatotoxins, primarily organic solvents, are widely used throughout industry today, there is a limited amount of human data on the effects of such exposures. The incidence of acute and subacute occupational and environmental hepatotoxicity is unknown.

Pathogenesis The typical pathologic findings are varying degrees of hepatocellular necrosis and steatosis. Neither of these

592 Liver Diseases Test

Type of injury

Comments

Serum markers Aminotransferases (AST, ALT)

Acute/subacute hepatocellular injury.

Alkaline phosphatase

Cholestatic injury (i.e., methylenediamine)

γ-glutamyl transpeptidase (GGT) Tests for function Tests of hepatic clearance Indocyanine green (ICG)

Cholestatic and hepatocellular injury

Bile acids (SBA) (cholic, chenodeoxycholic acid)

Early hepatobiliary and hepatocellular injury

Bilirubin

Cholestatic with hepatocellular injury

Tests of hepatic metabolism Antipyrine clearance

Early hepatocellular injury

Not readily available, high test individual variability. May be marker of exposure rather than toxicity

Late hepatocellular injury

Low sensitivity for early disease

Biliary disease, fatty change, hepatocellular injury

High sensitivity and specificity for biliary disease. Efficacy not established for parenchymal disease Efficacy not established for parenchymal disease Fibrosis observed by linear enhancement using gadolinium Gold standard for the diagnosis of parenchymal disease. Centrizonal (Zone 3) changes in solvent-induced injury

Tests of synthetic function Albumin, PT Imaging tests Ultrasound

Early hepatobiliary injury

CT scan

Fatty change, hepatocellular injury

MRI scan

Chronic hepatocellular injury and fibrosis

Liver biopsy

Subacute/chronic hepatocellular injury

Best available test for acute/subacute hepatocellular injury Not sensitive for fatty liver High sensitivity for cholestatic injury; not very specific Elevated with ethanol consumption, hepatobiliary disease, low specificity Useful for vinyl chloride toxicity Requires IV infusion and multiple blood samples Not readily available. Improved sensitivity compared to AST, ALT for hepatobiliary and hepatocellular disease in some settings (e.g., vinyl chloride) Elevated conjugated (direct) bilirubin, conjugated (total bilirubin ratio observed with styrene and other solvent exposure) Less specific than SBA or AST, ALT

Table 26.1.3 Diagnostic tests for occupational liver disease

conditions is specific for occupational or environmental toxins. Necrosis can occur with either a zonal pattern (usually centrizonal) or more diffusely, depending on the toxic agent and degree of exposure. Some degree of inflammatory response, with a lymphocytic infiltrate, also may be seen. Agents causing acute hepatocellular injury also may result in more chronic hepatocellular disease following repeated exposure, and are discussed in the following section. Some established and suspected human hepatotoxic agents, grouped by classes of substance, are listed in Table 26.1.2.

Clinical features The clinical effects of acute and subacute hepatic necrosis may vary greatly in severity from asymptomatic histologic or functional abnormalities to manifestations of severe hepatic necrosis, depending on the intrinsic toxicity of the hepatotoxic agent, the degree of exposure, and host susceptibility factors (either genetic or acquired). Fulminant hepatic injury, with abdominal pain, nausea, and jaundice, may occur following heavy acute. Historically, exposures have frequently involved chlorinated organic compounds (e.g., carbon tetrachloride).27 Most recently reported cases have occurred following accidental overexposure, ingestion,

abuse (e.g., sniffing), extensive dermal contact, or respiratory exposure in a closed or poorly ventilated area. Agents have included dimethylformamide, 2-nitropropane, 1,1,1trichloroethane, or trichloroethylene (TCE). Such acute hepatic injury usually is part of a multisystem disease, with prominent central nervous system and renal involvement. With less extensive exposure, asymptomatic elevations in liver aminotransferases may be noted. A clinical presentation between these extremes, with non-specific gastrointestinal symptoms, also occurs. Individual host factors, or additional exposures such as alcohol, medications, or other chemicals in the workplace may alter the severity of the toxic hepatic injury. Physical findings depend on the severity and duration of the hepatocellular injury. They range from a normal examination, to right upper quadrant tenderness, hepatomegaly, jaundice, and signs of hepatic failure in extreme cases.

Diagnosis In acute fulminant cases, the diagnosis of hepatitis and the specific etiology usually are not difficult to determine. Often, there is a history of an unusual overexposure occurring immediately prior to the onset of symptoms (1–3 days). Markedly elevated liver aminotransferase and bilirubin

Chronic (Non-Malignant) Occupational and Environmental Liver Disorders 593 levels are noted, along with other findings of hepatic failure, such as coagulopathy, renal, and neurologic dysfunction. The specific hepatotoxin(s) usually can be determined from occupational and environmental history. Occasionally, blood or urine testing can be performed to identify a particular chemical or its metabolite. Less severe clinical presentations, such as that following subacute poisoning, can be more difficult to recognize and diagnose. Symptoms and physical findings may be unremarkable so liver function tests are not obtained. The clinician must therefore have a high level of suspicion for occupational liver disease. Elevated aminotransferase levels (ALT and/or AST) are the best available indicator of acute hepatic injury, but other causes for hepatic injury should be excluded before attributing the injury to a workplace exposure. A thorough exposure history is essential to recognising an occupational etiology. A medical history, physical examination, and blood tests help exclude other causes of elevated levels of liver enzymes, including medications; hypervitaminosis (due to megavitamins); obesity; metabolic disorders, including diabetes hypertriglyceridemia, and hemochromatosis; pregnancy; viral hepatitis; and alcohol. Serologic tests for hepatitis A and B are useful in detecting acute viral hepatitis; in hepatitis C, the serologic test may take up to 6 months after infection to become positive. Polymerase chain reaction (PCR) tests for both hepatitis B DNA and hepatitis C RNA are extremely sensitive for viral hepatitis in both the acute and chronic setting (see Chapter 22). The possibility of unrecognized alcohol use or idiopathic liver injury are usually the most difficult differential diagnoses to definitively rule out. With alcohol-induced liver injury, AST levels greater than 300 are uncommon and the AST/ALT ratio usually is greater than one. An occupational or environmental etiology is strongly suggested if aminotransferase levels improve within 2–6 weeks following removal from the exposure(s), but it is not ruled out if the aminotransferase levels remain elevated (see later). If liver aminotransferases remain elevated, a liver biopsy should be considered. Findings of hepatocellular injury and steatosis are consistent with a hepatotoxic etiology, whereas marked inflammation is more suggestive of a viral one.

Natural history and management Acute occupational hepatitis is a severe, potentially fatal disease that usually occurs 1–3 days following a massive accidental hepatotoxic exposure. Multiorgan system failure, including central nervous system depression (i.e., coma) and renal failure, is frequently present, particularly in fulminant cases. Acute occupational hepatitis should be treated in a similar manner to any other type of severe hepatitis, with appropriate supportive measures. Referral to a liver transplant center may be required in fulminant cases. If the patient recovers from the acute episode, the prognosis is usually good. Aminotransferase elevations and symptoms generally improve within days, with resolution over weeks, but the time course is variable. Subacute hepatocellular injury may develop weeks to months following the hepatotoxic exposure, depending on the nature and extent of the exposure. If the patient is

diagnosed or strongly suspected of having undergone toxic exposure, he or she should be removed from the suspected exposure and observed. Exposure to other risk factors such as alcohol consumption should be minimized. The workplace and environment should be evaluated for hepatotoxins, and appropriate measures taken to reduce exposures. Elevated aminotransferase levels and symptoms generally improve 1–4 weeks following removal from exposure, but the time course is variable, depending on the specific hepatotoxin; persistent elevations that last for months have been documented in outbreaks. Whether chronic liver disease may result following subacute hepatotoxic injuries is unclear.

Acute cholestatic injury Acute cholestatic liver injury is rare following occupational or environmental exposures but has been reported following exposure to methylene dianiline, an aromatic amine used as an epoxy resin hardener.28 An epidemic of cholestatic jaundice occurred in Epping, England (so-called Epping jaundice) in 1965 after bread made from flour contaminated with methylene dianiline was ingested. Similar cases have been reported following occupational exposure during the manufacture or handling of methylene dianiline. The clinical, laboratory, and pathologic findings are consistent with a mixed cholestatic-hepatocellular injury,29 that is dose dependent rather than idiosyncratic. Symptoms may include abdominal pain, pruritus, fever, and jaundice. Laboratory studies show elevated bilirubin, alkaline phosphatase, and aminotransferase levels. A consistent pattern of injury with bile stasis, portal inflammation, and variable hepatic necrosis is found on liver biopsy. The diagnosis depends on an appropriate exposure history, clinical and laboratory presentation, and elimination of other causes of acute cholestatic injury, such as drugs or biliary obstruction. Resolution of liver injury following removal typically occurs from exposure has not been reported.

CHRONIC (NON-MALIGNANT) OCCUPATIONAL AND ENVIRONMENTAL LIVER DISORDERS Fatty liver (steatosis) and steatohepatitis Fatty change in the liver, termed steatosis, was first characterized in alcohol-related liver disease. Steatosis is defined morphologically as greater than 5% hepatocytes containing fat, or quantitatively as greater than 5 g lipid per 100 g hepatic tissue. Steatosis occurs in other disorders, including diabetes mellitus, hypertriglyceridemia, obesity30 (with 90% prevalence in the morbidly obese); with various medications; and as a normal variant, with up to 20% prevalence in some series.31 Some degree of steatosis usually is found accompanying acute hepatocellular necrosis. Marked steatosis is more commonly seen in chronic toxin-induced

594 Liver Diseases liver injury, frequently as the predominant finding. Steatosis may be accompanied by varying degrees of inflammation, termed steatohepatitis. There is good evidence that this can be multifactorial, thus industrial hepatotoxins may interact with underlying metabolic disorders and other causes of non-alcoholic steatohepatitis (NASH).32

Pathogenesis Steatosis results from the pathologic alteration of hepatic fat metabolism. Dietary fat normally is transported to the liver via the portal vein as chylomicrons containing fatty acids. Hepatocytes convert free fatty acids to a transportable form, very low-density lipoprotein (VLDL), which is secreted into the circulation. Hepatotoxins can block fat metabolism at several steps, resulting in intrahepatic accumulation of free fatty acids and triglycerides, producing lipid droplet formation at the cellular level and diffuse or focal steatosis at the tissue level.

Clinical presentation Fatty liver associated with occupational exposure was first described with yellow phosphorus poisoning in the 19th century, with pronounced steatosis and necrosis found at autopsy. Similar cases of acute massive necrosis and steatosis have been described with trinitrotoluene in munitions industries, arsenical pesticide use in vintners, and the use of certain chlorinated aliphatic solvents (such as carbon tetrachloride, methyl chloroform, and tetrachloroethane), usually following accidental massive exposures. More subtle microsteatosis was recently described following routine short-term exposure to dimethylformamide in a fabric-coating factory. The prevalence of fatty changes related to chronic, lowlevel, chemical exposure is not known. Patients are usually asymptomatic. Screening tests (AST, ALT) may not detect steatosis in the absence of inflammation and necrosis. Investigation of steatosis is usually limited by strong reliance on histologic diagnosis, which is invasive and may not be available. Diagnosis is further complicated by the confounding etiologies of fatty liver, including alcohol consumption, obesity, diabetes, medications, and their interactions with suspected toxins. Chronic exposure to chlorinated solvents such as carbon tetrachloride can cause varying degrees of steatosis and hepatocellular injury. Although the evidence is less consistent with non-chlorinated solvents, several studies have found steatosis in workers exposed to non-chlorinated solvents, including dimethylformamide, toluene, and mixed aliphatic and aromatic solvents. The relative degree and severity of steatosis and necrosis found on histopathologic examination is likely related to the chronicity and extent of the exposure, as well as the hepatotoxic characteristics of the particular exposure. Acute high-level exposures tend to be associated with greater amounts of necrosis, whereas steatosis tends to predominate with more chronic exposures.

animal models indicate that fatty change occurs as part of a continuum of reversible, morphologic changes: cloudy swelling is followed by hydropic changes and, finally, steatosis. In humans, the reversibility of alcohol-related fatty change with abstinence has been well documented. Acute solvent-related steatosis is generally associated with necrosis and elevated aminotransferase levels. The elevated aminotransferase levels generally resolve within weeks to months of removal from exposure. Because of the need for histologic evidence, the natural history of the accompanying acute steatosis is less clear, but the condition most likely also resolves over time. Steatosis following acute exposures is thus likely to represent subclinical injury at a reversible stage. At present, there are no human longitudinal studies addressing the sequelae of isolated steatosis with prolonged or ongoing occupational exposure. Necrosis, steatosis, and fibrosis can be induced in animals by the chronic administration of carbon tetrachloride. Evidence exists in both animal and human studies that steatosis can occur in the absence of elevated serum hepatic transaminase levels. Balazs demonstrated that rats exposed to low doses of carbon tetrachloride had no ALT elevation, despite marked fatty and hydropic changes histologically.33 Similarly, longitudinal follow-up of individuals with alcohol-induced or metabolically induced (diabetes and obesity) hepatic injury have demonstrated significant progression of steatosis to fibrosis histologically, termed ‘steatocirrhosis’,34 often in the absence of an inflammatory response and associated transaminase elevation.35,36 In one study, seven workers exposed to dimethylformamide over a 3-month to 10-year period, with persistently elevated aminotransferase levels, underwent liver biopsy. Variable degrees of steatosis, which was greatest in those with longer periods of exposure, were found; there was no evidence of progression to fibrosis or cirrhosis at the time of biopsy.

Diagnosis Laboratory tests may not be helpful in diagnosing fatty liver, because they frequently do not detect steatosis in the absence of inflammation. Non-invasive anatomic evaluation of the liver by ultrasonography and CT scan can suggest hepatic steatosis. The definitive diagnosis of steatosis depends on histopathologic examination of a liver biopsy specimen. When significant steatosis is found, either incidentally or in the work-up of elevated serum aminotransferases, evaluation should attempt to differentiate occupational from other known causes of steatosis, as discussed earlier. The history should include a review of the following: medications associated with steatosis (such as phenytoin [Dilantin], tetracycline, isoniazid, nitrofurantoin, and phenylbutazone) and the presence of hyperlipidemia, diabetes mellitus, obesity or pregnancy, and substance abuse, such as glue sniffing. Laboratory evaluation should include measurements of fasting blood sugar and triglyceride levels.

Natural history

Management

The natural history of chemically induced hepatic steatosis has not been well characterized. Histopathologic data from

The presence of steatosis without other obvious etiologies, associated with exposure to hepatotoxic agents, is suggestive

Chronic (Non-Malignant) Occupational and Environmental Liver Disorders 595 of occupational or environmental liver disease. Although the natural history of steatosis in such settings is unclear, further toxic exposure should be minimized, and removal of the person from the workplace should be considered. Resolution of any concomitant elevation of aminotransferase levels after removal from exposure supports an occupational etiology.

Chronic hepatocellular injury Chronic hepatocellular injury can occur after prolonged exposure to agents that cause acute and subacute hepatic injury or steatosis. Human data are limited to a few case reports of liver injury following chronic exposure to various hepatotoxic agents (including dimethylformamide, carbon tetrachloride, and mixed solvents) and cross-sectional surveys reporting elevated liver enzyme levels in exposed cohorts such as painters, print or shoe repair workers20 compared with controls. The clinical presentation may range from minimal to pronounced symptoms and aminotransferase elevations. Liver biopsy may show varying degrees of necrosis, regeneration, inflammation, and steatosis. The diagnosis of chronic hepatocellular injury is made based on the guidelines given earlier for acute and subacute liver injury. However, the diagnosis frequently is more difficult to determine than that of acute or subacute injury. Exposure assessment frequently is complicated by multiple, poorly defined exposures over many years, and it must be differentiated from other etiologies. Resolution of enzyme elevations after removal of the person from exposure is helpful. However, the natural history is not well defined, and aminotransferase abnormalities appear to resolve more slowly than with acute hepatocellular injury. It is unclear whether progression to cirrhosis or increased risk for hepatoma occurs.

Granulomatous hepatitis Exposure to pathogenic dusts that cause pulmonary granulomatous lesions also may result in hepatic granulomas. Hepatic granulomas also may result from drug-induced hepatic injury or may be a non-specific finding in almost any setting. Hepatic granulomas have been described in chronic beryllium disease, silicosis, vineyard sprayer’s lung (presumed secondary to copper exposure), and following mica and cement exposure. The hepatic lesions usually are asymptomatic and functionally not important, but rarely can be accompanied by hepatomegaly, necrosis, or fibrosis. Diagnosis depends on the appropriate exposure history and finding granulomatous changes on liver biopsy. The inhaled foreign particles may be detected in the biopsy specimen. The natural history of such lesions is not well defined but is most likely benign. As with the underlying granulomatous disorder, the lesions may progress after exposure has ceased. It is unclear whether or not the administration of steroids is beneficial. Exposure to additional hepatotoxic agents should be minimized.

Hepatoportal sclerosis Hepatoportal sclerosis is a rare form of non-cirrhotic periportal fibrosis, which can lead to portal hypertension. It has been described most commonly as a consequence of exposure to vinyl chloride monomer,37,38 inorganic arsenicals, and thorium compounds. Inorganic arsenicals and thorium compounds are no longer in common use. Vinyl chloride monomer, a chlorinated hydrocarbon, is used in the production of polyvinyl chloride (PVC), a widely manufactured plastic. Hepatoportal sclerosis has been documented in several studies of workers exposed to vinyl chloride, primarily in PVC polymerization manufacturing plants.39 Liver histology has shown hyperplasia of hepatocytes and sinusoidal cells, with dilatation of sinusoids and progressive subcapsular, portal, perisinusoidal, and, occasionally, intralobular fibrosis, which is accompanied by portal hypertension and splenomegaly. Workers with hepatoportal sclerosis have a markedly increased risk of developing angiosarcoma of the liver, a rare liver cancer seen almost exclusively in the setting of vinyl chloride or arsenic exposure;40 greater than threefold increases in liver cancer (SMR 333; 90% CI 202–521) have been observed in vinyl chloride polymerization plant workers.41 Hepatoportal sclerosis usually is found in tumor-free areas of the liver in patients with angiosarcoma and may be a premalignant promoting factor (see Chapter 30.6). Vinyl chloride-induced liver injury presents insidiously, with the diagnosis made frequently 15–20 years after the first exposure.42 Portal hypertension with periportal fibrosis may develop eventually. The diagnosis is made on the basis of a consistent exposure history and characteristic findings on liver biopsy. Standard liver function tests usually are of limited value in detecting vinyl chlorideinduced liver injury. Several studies have indicated that ICG clearance and serum bile acids provide improved sensitivity and specificity in screening exposed workers.43 There have been a few case reports suggesting that exposure to polymeric vinyl chloride also may be associated with hepatic angiosarcoma. Whether exposure to PVC also can cause non-malignant liver injury is unclear. There generally is little release of vinyl chloride monomer from solid PVC materials, although significant amounts may be released with combustion of PVC products and possibly in the processing or some applications of PVC materials.

Cirrhosis Cirrhosis, or end-stage liver disease, is defined as a chronic, irreversible condition in which the normal lobular architecture is replaced by fibrous tissue and regenerating nodules derived from the remaining hepatocytes. Cirrhosis is most commonly due to alcohol or viral infection but also may be caused by chronic biliary disease, metabolic diseases such as hemochromatosis, and congestive heart failure. Cases of cirrhosis sometimes are idiopathic. Animal studies have shown that hepatotoxic exposures such as carbon tetrachloride, arsenic, and aflatoxins can result in cirrhosis. The role of occupational or environmental

596 Liver Diseases substances in the development of cirrhosis, either alone or interacting synergistically with other etiologic agents has not been well characterized in humans. Whatever the initial cause, the pathogenesis involves hepatocellular necrosis, with subsequent deposition of large amounts of connective tissue and nodular regeneration of hepatocytes. There have been isolated, often not well-documented, case reports of cirrhosis associated with repeated exposure to several occupational agents, most of them no longer in use, including carbon tetrachloride, arsenicals, tetrachloroethane, trinitrotoluene, and trichloroethylene (TCE).44 Increased mortality due to cirrhosis has been noted in several cohorts of workers with known or suspected hepatotoxic exposures (primarily solvents), including pressmen, shipyard workers, marine inspectors, metal fabrication employees, anesthesiologists, vineyard workers exposed to arsenical pesticides, and PCB workers. Mortality studies have also suggested increased risk for chronic hepatic injury in occupational groups with longterm daily exposure to organic solvents. These studies have observed relative risks for cirrhosis ranging from 1.6 to 2.1 in automobile spray painters and newspaper printers.45–48 However, in many of these studies, confounding factors, such as ethanol or viral hepatitis, were not accounted for. In patients with idiopathic cirrhosis, the diagnosis of an occupational or environmental etiology should be considered if there has been a history of repeated exposure to known or suspected hepatotoxins. Further hepatotoxic exposures should be minimized.

MAJOR HUMAN HEPATOTOXINS Several specific environmental agents that have been reported to cause liver injury in humans are discussed in this section.

Solvents

icity still are reported. Tetrachloroethane and chloroform also are no longer in general use because of their recognized hepatotoxicity. 1,1,1-Trichloroethane and TCE are widely used chlorinated solvents with documented animal hepatotoxicity. Although they are less hepatotoxic than carbon tetrachloride, exposure to both agents has been associated with case reports of hepatic necrosis and steatosis.50–53 Perchloroethylene, a popular dry-cleaning solvent, is also hepatotoxic in animal studies and historical case reports.54 Elevations in GGT, in the absence of other liver function test abnormalities have been observed in surveillance of exposed workers.55 A cross-sectional field investigation of 29 dry-cleaning operators observed mild to moderate hepatic parenchymal changes in echogenicity, as determined by hepatic ultrasonography, in two-thirds of dry cleaners compared with only one-third of referents (OR 3.2, 95% CI 1.04–9.8),25 with minimal associated changes in transaminase levels.

Dimethylformamide Dimethylformamide is a widely used hepatotoxic solvent. Several recent reports have documented acute and chronic liver injury in workers exposed to dimethylformamide.56 In addition to typical gastrointestinal symptoms, a disulfiramtype reaction (acquired alcohol intolerance) is frequently reported following acute exposure. Elevated serum aminotransferase levels, with an increased ALT/AST ratio, have been found. Liver biopsy following acute exposure (less than 2 months) has shown variable degrees of hepatic necrosis, regeneration, and microvesicular and macrovesicular steatosis, whereas marked steatosis without fibrosis has been found on biopsy following longer exposures57 (Fig. 26.1.1). Elevated liver enzyme levels may persist weeks to months after removal of the person from further exposure. The long-term effects of dimethylformamide exposure are not known.

Carbon tetrachloride and other chlorinated solvents Carbon tetrachloride is the classic example of an intrinsic occupational hepatotoxin. The hepatotoxicity of carbon tetrachloride was first recognized in the 1920s, when it was frequently used as a solvent, dry-cleaning agent, fire extinguisher, and anthelmintic. As with most hepatotoxic solvents, acute intoxication results in multisystem disease, with central nervous system depression and acute necrosis of the renal tubules and the liver. Symptoms and liver aminotransferase abnormalities usually occur 1–2 days following the acute exposure and resolve in 1–2 weeks, but the time course may vary depending on the dose and length of exposure. Histology reveals centrilobular necrosis and steatosis; cirrhosis following carbon tetrachloride exposure is well established in animal models, and there are a few case reports in humans. Carbon tetrachloriderelated hepatotoxicity is potentiated by ethanol consumption.49 Because of renal and hepatic toxicity, use of carbon tetrachloride has decreased greatly, but cases of hepatotox-

2 1

Figure 26.1.1: Liver histology from a patient who had been exposed to the solvent demethylformamide for several years. Liver architecture is normal, but there is moderate to severe steatosis, mainly in zones 2 and 3. 1, central vein; 2, a portal tract. (Carnoy’s fixation, H & E, original magnification × 130.) (From Redlich CA, West AB, Fleming LE, et al. Clinical and pathological characteristics associated with occupational exposure to dimethylformamide. Gastroenterology 1990; 99:748–57, with permission from American Gastroenterological Association.)

Major Human Hepatotoxins 597

Aromatic and other solvents Toluene is one of the most frequently used solvents in industry, and it is a known animal hepatotoxin. Elevated serum transaminase and GGT levels have been found in print workers exposed to toluene58,59 and after exposure to glues.60 Liver biopsies reveal fatty liver with mild nonspecific inflammation. The long-term hepatic effects of toluene exposure are not known. Several investigations of workers exposed to relatively high levels of styrene (e.g., 50–300 ppm) observed elevations in serum hepatic transaminase61 and GGT levels,62–64 suggesting hepatic necrosis and cholestasis, respectively. Two independent cross-sectional studies of workers exposed to lower levels of styrene, including fiberglass reinforced plastics workers and boat and tank fabricators, observed increased direct bilirubin and AP levels compared to non-exposed referent groups, consistent with diminished hepatic clearance of conjugated bilirubin with associated cholestasis.21 The absence of similar elevations in hepatic transaminase levels in these studies suggests that low-level styrene exposure results in mild metabolic dysfunction without significant hepatic parenchymal necrosis. There are isolated case reports of acute hepatotoxic injury following acute exposure to several other solvents, including 2-nitropane,65,66 methyl chloride, tetrahydrofuran, carbon tetrabromide, and xylene.67

Mixed solvents Exposure to solvent mixtures is more common than isolated single exposures. Several epidemiologic studies have reported elevated liver enzyme levels in workers such as painters, print workers, and chemical workers exposed to various solvents including toluene, acetone, benzene, methyl ethyl ketone, and styrene. Typically, aminotransferase levels have been elevated compared with control groups, with AST/ALT ratio less than one. Dossing observed that 13 of 156 (8.3%) career painters with chronic encephalopathy due to ‘white spirit’ (mixed aliphatic and aromatic hydrocarbons) exposure had elevated serum transaminases.68 Liver biopsy revealed that 11 of these 13 painters (85%) had moderate steatosis and focal hepatic necrosis. Evidence of hepatotoxicity has also been demonstrated in painters exposed chronically to solvents, with elevated alkaline phosphatase levels observed in a crosssectional study of Swedish house painters.69 Elevated GGT levels also have been reported in solvent-exposed groups such as paint manufacturers and sprayers,70 as well as necrosis and steatosis on liver biopsies of selected subjects.71 Severe liver injury with centrilobular and hemorrhagic necrosis has been observed in chemical plant workers exposed to mixtures of carbon disulfide, isopranolol, toluene, and acrylonitrile.72 In some of these studies, other competing etiologies, such as alcohol, viral hepatitis, or obesity, were not well controlled. Solvent mixtures may interact to enhance hepatotoxicity, as demonstrated by Charbonneau and colleagues in rodents exposed to CCl4, acetone, and corn oil.73,74 In human volunteers, altered oxidative hepatic metabolism

during mixed solvent exposure has also been demonstrated with toluene,75,76 as well as methyl ethyl ketone (MEK) and various alcohols,77 and may explain the variable hepatotoxicity observed with mixed solvent exposures.11

Halothane and other anesthetic solvents Halothane is a general anesthetic agent that can cause hepatitis in patients undergoing general anesthesia, and in occupationally exposed workers, such as anesthetists and operating room personnel.78,79 Halothane-induced hepatitis is not dose dependent, and significant liver injury is rare, usually occurring following repeated exposures. Serum aminotransferase levels are elevated, and liver biopsy shows spotty to massive necrosis, frequently with steatosis and eosinophilic infiltration. Fever and eosinophilia usually are present. There are rare case reports of cirrhosis and chronic active hepatitis following halothane exposure. A few epidemiologic studies have reported an increased frequency of liver disease among operating room personnel.80 Methoxyflurane, another widely used anesthetic, also has been reported to cause liver injury similar to halothaneinduced hepatitis.

Organochlorine pesticides and related polychlorinated compounds Several non-specific liver abnormalities have been associated with exposure to pesticides. A massive outbreak of acquired porphyria cutanea tarda and hepatomegaly occurred in Turkey following the ingestion of grain contaminated with the fungicide hexachlorobenzene. Chronic hepatic sequelae were not well documented. The organochlorine pesticide chlordecone has been reported to cause hepatomegaly, non-specific changes on liver biopsy, and marked proliferation of smooth endoplasmic reticulum on electron microscopy, consistent with induction of P-450 enzymes.81 These changes resolved following removal of the affected individuals from exposure. The use of organochlorine pesticides has been limited because of potential carcinogenicity, including liver and biliary tract cancer among DDTexposed workers.82 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD), or dioxin, is a contaminant of chlorophenoxy herbicides that can cause chloracne and liver injury.83 Mildly elevated liver aminotransferase and GGT levels have been reported following TCDD exposure, with necrosis and steatosis on liver biopsy. Persistent liver injury has been reported by a European group who found elevated transaminases, GGT, and urinary porphyrins in Austrian chemical workers 25 years after their main exposure to dioxin.84 Because of its storage in body fat, dioxin has an extremely long half-life of many years. Persistently elevated transaminase levels have also been documented among Vietnam veterans exposed to Agent Orange.83,85 PCBs and chlorinated naphthalenes have been used widely as insulating liquids in electrical systems. While no longer manufactured because of concerns about their biopersistence and carcinogenicity,86 exposure continues to be

598 Liver Diseases documented in inhabitants living near electrochemical factories.87 Heavy occupational exposure or accidental oral ingestion of these agents has been reported to cause hepatic necrosis and, in some cases, subsequent subacute hepatic necrosis and cirrhosis. Among workers manufacturing capacitors and transformers, elevations in LDH and GGT correlated with PCB levels.88 Halogenated pesticides such as DDT and hexachlorobenzene, chlorinated dioxins such as TCDD, and polychlorinated biphenyls (PCBs) are all potent inducers of various cytochrome P-450 mono-oxygenases, and they may potentiate the hepatotoxicity of other exposures by enhancing metabolism of toxic intermediates.

Metals Arsenic Accidental arsenic ingestion can result in acute hepatic injury with necrosis and steatosis. Chronic exposure to arsenic has been seen in vintners,89 farmers, and miners, with variable hepatotoxcity. Arsenic in insecticide sprays has been associated with an increased incidence of cirrhosis and hepatoportal sclerosis. Such arsenical compounds are no longer in use. Besides its hepatic effects, arsenic also affects the gastrointestinal tract, central nervous system, and cardiovascular system. Hemorrhagic necrosis of the GI tract, nausea, vomiting, CNS depression, and vascular collapse are frequently seen.90 Environmental arsenic poisoning from ground-water sources has been an important public health problem in endemic areas such as Bangladesh and West Bengal, where hepatic abnormalities are frequently seen as part of a multisystemic illness.91

Lead Elevated aminotransferase levels have been reported with acute inorganic lead intoxication, usually with lead levels in the range of 80 μg/dl or greater. Levels promptly improve following chelation treatment and removal from further exposure. The presence of liver injury may explain some of the gastrointestinal toxicity found in acute poisoning cases.

Other environmental agents Although they are primarily industrial hepatotoxins, many of the agents discussed earlier and listed in Table 26.1.1 also are potential environmental toxins. Hepatotoxic exposures could potentially occur through contact with contaminated water, air, soil, or food. Such environmental contamination is of particular concern with chemicals such as DDT or PCBs, which are very stable, resist biodegradation, and accumulate in the environment and food chain. Notably, the substitution of hydrochlorofluorocarbons (HCFC) for ozone-depleting chlorofluorocarbons has resulted in prominent hepatic injury among groups of exposed workers.92 In addition, several naturally occurring toxins produce marked hepatotoxicity on ingestion.

Aflatoxins The fungus Aspergillus flavus produces toxins, termed aflatoxins, that can contaminate food products, including nuts, corn, and wheat. The fungus is abundant in tropical and subtropical regions, and thrives under moist, warm conditions. Aflatoxin B1 is a potent inhibitor of RNA synthesis and the most hepatotoxic of these compounds.90 Acute exposure has been reported to cause necrosis, steatosis, and inflammation. Chronic exposure is associated with an increased risk of hepatic carcinoma; the risk of malignancy is directly proportional to the amount of toxin ingested.90

Mushroom poisoning Ingestion of Amanita phalloides (death cap) mushrooms can result in fatal poisoning, with massive hepatic necrosis. Fulminant hepatic failure is seen, with coma and renal failure common. These mushrooms contain several toxins, with α-amanitin being the most toxic. Amatoxin is thermostable, and can remain active for years. A fatal dose has been estimated to be in the range of 7 mg, or three mushrooms.90

Viral agents Hepatitis A, B, or C can occur as an occupational disease in healthcare workers and others in contact with the excreta or blood from patients or carriers infected with these viruses. Hepatitis A causes an acute hepatitis, typically with no long-term sequelae. Hepatitis B and hepatitis C may cause both acute and chronic hepatitis. A full discussion of viral hepatitis is provided in Chapter 22.

MEDICAL SURVEILLANCE FOR OCCUPATIONAL LIVER DISEASE Surveillance strategies Strategies for controlling occupational liver disease follow the general principles of primary and secondary prevention. Primary preventive strategies attempt to identify and remove (or reduce) hepatotoxic exposures. Hepatotoxic substances are identified by review of the chemicals used and assessment of industrial hygiene to limit exposure. Hepatotoxic exposures can be minimized by substituting less toxic agents and by using engineering controls (e.g., improved ventilation) and personal protective equipment. Secondary preventive strategies involve the screening of workers actively exposed to known or suspected hepatotoxins. This approach is appropriate when exposure is unpredictable or unavoidable (i.e., not amenable to primary prevention). Such strategies attempt to identify hepatic disease at an early, reversible stage.

Screening tests for occupational hepatotoxicity Screening tests for occupational liver disease include the biochemical, clearance, and metabolic tests described earlier.93 Despite high sensitivity and specificity in particu-

Medical Surveillance for Occupational Liver Disease 599 lar settings, all tests are potentially limited by low predictive value.11,94 Because of a low prevalence of hepatic disease in the general population, only a small percentage of individuals who have positive test results actually have liver disease. For example, serum aminotransferase levels, which are more than 90% accurate in the diagnosis of acute viral hepatitis, have a predictive value of less than 10% in screening for chronic liver disease in the general population. Determining the predictive value of screening tests is complicated by the lack of prevalence data for occupational liver disease. A notable exception is vinyl chloride monomer, with a 2–3% prevalence of chronic liver disease in exposed workers. The most accurate test for such disease is ICG clearance, with a predictive value of 20%. In most industries, however, the prevalence of hepatic disease is unknown and probably variable; the efficacy of a screening test in one setting cannot be assumed to be the same in another. Screening of healthy workers is recommended only when there is exposure to known or strongly suspected hepatotoxic agents. Because no ideal screening test for occupational liver injury exists, serum aminotransferase tests (AST and ALT) remain the best practical choice at present. They are inexpensive, available, and clinically validated. Combination testing, namely requiring positive results on two different tests, may enhance predictive value, but this approach has not been validated and may result in high rates of false-negative results. Metabolic tests such as antipyrine clearance should be considered markers for specific exposures rather than for actual hepatic injury.

Clinical management of abnormal liver function tests Management of abnormal liver function tests depends on the clinical presentation (the extent and type of liver injury) and exposure setting (the likelihood and extent of hepatotoxic exposure).95 In acute and subacute settings, with well-documented exposure to a known or suspected hepatotoxin, an association between the liver injury and hepatotoxic exposure frequently can be determined, once other possible etiologies have been excluded. Prompt removal of the affected individual from exposure to any workplace or environmental hepatotoxin is indicated. Transaminase levels are generally elevated several fold and should be monitored closely, with the use of early hepatic ultrasound to rule out biliary obstruction. Supportive care, including hospitalization, may be required in severe cases. On the other hand, in the setting of chronic low-level exposures, as often occurs in mass screening of healthy workers, the relationship between elevated hepatic transaminase levels and specific exposures may be much more difficult to determine. Because low predictive values, in the range of 5–20% for available tests, generate high rates of false-positive findings, emphasis should be placed on distinguishing true from false-positive results. Removal of the affected individual from work should be considered only after alternative etiologies, such as medications, non-occupational liver disorders, and excessive alcohol use

have been excluded.96 For example, mild hyperbilirubinemia in an otherwise healthy worker likely represents Gilbert’s syndrome and usually does not warrant further evaluation. There are no strict guidelines for the management of abnormal serum aminotransferase levels, which are often minimally elevated. In situations in which significant exposure is unlikely and transaminase levels are between one and two times normal, repeat testing in 4 weeks has been advocated, with further investigation reserved only for persistent elevations. A general approach for evaluating possible occupational or environmental liver disease is summarized in the following list, although modifications are needed, depending on the particular clinical and exposure setting. 1. Assess both occupational and non-occupational causes of liver disease, including significant alcohol use, viral hepatitis (A, B, and C), biliary disease, medications, blood transfusions, and hepatotoxic exposures from work, hobbies, the home, and second jobs (see item 3). Potential hepatotoxic medications are numerous97 (e.g., acetaminophen, isoniazid, erythromycin, estrogens, phenytoin, and megavitamins such as vitamin A) and should be discontinued, when possible. The abnormal liver tests should be repeated in 2–4 weeks. Metabolic etiologies including diabetes mellitus, hemochromatosis, hypertriglyceridemia and obesity should be considered. 2. Consider the AST/ALT ratio. Ratios less than one are suggestive of viral or toxic exposures rather than alcohol use. 3. Attempt to determine the presence of any known or suspected hepatotoxins in the workplace or environment. Sources of such information may be obtained from material data safety sheets, the worker, the employer, the manufacturer, and industrial hygiene sampling, if available. 4. If a significant hepatotoxic exposure is identified, the individual should be removed from further exposure and his or her AST and ALT should be rechecked in 2–4 weeks. Recovery suggests an occupational or environmental etiology, but persistent elevations do not rule it out. Such persons may return to work or the same environmental setting on a trial basis if appropriate controls or exposure modifications have been implemented. Close follow-up, with continued monitoring of AST and ALT levels, is necessary in these situations. 5. Persistent elevation of AST and ALT levels greater than twice normal for over 2 months warrants further investigation; referral to a hepatologist should be considered. Such individuals have an increased likelihood of chronic active hepatitis, steatohepatitis, and fibrotic changes.98 Hepatic ultrasonography, liver biopsy, or both, may be indicated. Findings of steatosis and necrosis are particularly suggestive of an occupational etiology. A trial entailing removal of the affected individual from hepatotoxic exposures, even if they are only suspected, should be considered. Because various hepatotoxins may interact, alcohol use and potentially hepatotoxic medications should be minimized. In the end, the diagnosis of occupational or environmental liver injury must integrate the clinical data discussed

600 Liver Diseases above with knowledge regarding the intrinsic hepatotoxicity and nature of the exposure(s). Limited knowledge regarding hepatotoxicity, particularly for new or poorly characterized chemicals, as well as a determination of the nature and extent of exposure, especially with more chronic presentations, frequently represent challenges to the clinician.

References 1. Harford TC, Brooks SD. Cirrhosis mortality and occupation. J Stud Alcohol 1992; 53:463–8. 2. Batt A-M, Ferrari L. Manifestations of chemically induced liver damage. Clin Chem 1995; 41:1882–7. 3. Reinke LA, McCay PB. Free radicals and alcohol liver injury. In: Watson RR, ed. Drug and alcohol abuse reviews, vol. 2: liver pathology and alcohol. Totawa: The Humana Press, 1991;133–68. 4. Johansson I, Ingelman-Sundberg M. Carbon tetrachlorideinduced lipid peroxidation dependent on an ethanolinducible from of rabbit liver microsomal cytochrome P-450. FEBS Lett 1985; 183:265–9. 5. Koop DR, Casazza JP. Identification of ethanol-inducible P-450 isozyme 3A as the acetone and acetol monooxygenase of rabbit microsomes. J Biol Chem 1985; 260:13607–12. 6. Koop DR, Orlandi CL, Schnier GG. Ethanol-inducible P450 isozyme 3a is a benzene hydroxylase. FASEB J 1988; 2:A1012(abstract). 7. Hetu C, Dumont A, Joly JG. Effect of chronic ethanol administration on bromobenzene liver toxicity in the rat. Toxicol Appl Pharmacol 1983; 67:166–77. 8. Ingelman-Sundberg M, Johansson I. Mechanisms of hydroxyl radical formation and ethanol oxidation by ethanol-inducible and other forms of rabbit liver microsomal cytochromes P-450. J Biol Chem 1984; 259:6447–58. 9. Tsutsumi M, Shimizu M, Lasker MJ, Lieber CS. Intralobular distribution of ethanol-inducible cytochrome P450IIE1 in liver. Hepatology 1988; 8:1237A. 10. Hasumara Y, Teschke R, Lieber CS. Increased carbon tetrachloride hepatotoxicity, and its mechanism, after chronic ethanol consumption. Gastroenterology 1974; 66:415–22. 11. Harrison R. Medical surveillance for workplace hepatotoxins. State Art Rev 1990; 5:515–30. 12. Matloff DS, Selinger MJ, Kaplan MM. Hepatic transaminase activity in alcoholic liver disease. Gastroenterology 1980; 78:1389–92. 13. Cohen JA, Kaplan MM. The SGOT/SGPT ratio – an indicator of alcoholic liver disease. Dig Dis Sci 1979; 24:835–8. 14. Fleming LE, Shalat SL, Redlich CA. Liver injury in workers exposed to dimethylformamide. Scand J Work Environ Health 1990; 16:289–92. 15. Ferraris R, Fiorentini M, Galatola G, et al. Diagnostic value of serum immunoreactive conjugated cholic or chenodeoxycholic acids in detecting hepatobiliary disease. Dig Dis Sci 1987; 32:817–23. 16. Edling C, Tagesson C. Raised serum bile acid concentrations after occupational exposure to styrene: a possible sign of hepatotoxicity. Br J Ind Med 1984; 41:257–9. 17. Franco G, Fonte R, Tempini G, Candura F. Serum bile acid concentrations as a liver function test in workers occupationally exposed to organic solvents. Int Arch Occup Environ Health 1986; 58:157–64. 18. Franco G, Santagostino G, Lorena M, Imbriani M. Conjugated serum bile acid concentrations in workers exposed to low doses of toluene and xylene. Br J Ind Med 1989; 46:141–2. 19. Franco G. New perspectives in biomonitoring liver function by means of serum bile acids; experimental and hypothetical biochemical basis. Br J Ind Med 1991; 48:557–61.

20. Tomei F, Giuntoli P, Biagi M, Baccolo TP, Tomao E, Rosati MV. Liver damage among shoe repairers. Am J Ind Med 1999; 36:541–7. 21. Brodkin CA, Moon JD, Camp J, Echeverria D, Redlich C, Checkoway H. Serum hepatic biochemical activity in two populations of styrene-exposed workers. Occup Environ Med 2001; 58:95–102. 22. Savertymuttu SH, Joseph AEA, Maxwell JD. Ultrasound scanning in the detection of hepatic fibrosis and steatosis. Br Med J 1986; 292:13. 23. Williams DMJ, Smith PM, Taylor KJW, Crossley IR, Duck BW. Monitoring liver disorders in vinyl chloride monomer workers using greyscale ultrasonography. Br J Ind Med 1976; 33:152–7. 24. Taylor KJW, Williams DMJ, Smith PM, Duck BW. Grey-scale ultrasonography for monitoring industrial exposure to hepatotoxic agents. Lancet 1975; i:1222–4. 25. Brodkin CA, Daniell W, Checkoway H, et al. Ultrasonic detection of occult hepatotoxicity in perchloroethylene-exposed workers. Occup Environ Med 1995; 52:679–85. 26. Semelka RC, Chung JJ, Hussain SM, et al. Chronic hepatitis: correlation of early patchy and late linear enhancement patterns on gaddinium-enhanced MR images with histopathology initial experience. J Mag Res Imag 2001; 13:385–91. 27. Brautbar N, Williams J. Industrial solvents and liver toxicity: risk assessment, risk factors, and mechanisms. Int J Hyg Environ Health 2002; 205:479–91. 28. Bastian PG. Occupational hepatitis caused by methylenedianiline. Med J Aust 1984; 141:533–5. 29. Kopelman H, Scheuer PJ, Williams R. The liver lesion of the Epping jaundice. Q J Med 1977; 35:553–67. 30. Wanless IR, Lentz JS. Fatty liver hepatitis (steatohepatitis) and obesity: an autopsy study with analysis of risk factors. Hepatology 1990; 12:1106–10. 31. Nomura H, Kashiwagi S, Hayashi J, et al. Prevalence of fatty liver in a general population of Okinawa, Japan. Jpn J Med 1988; 27:142–9. 32. Chitturis, Farrel GC. Etiopathogenesis of nonalcoholic steatohepatitis. Semin Liver Dis 2001; 21:27–41. 33. Balazs T, Murray TK, McLaughlan JM, Grice HC. Hepatic tests in toxicity studies on rats. Toxicol Appl Pharmacol 1961; 3:71–9. 34. Harinasuta U, Chomet B, Ishak KG, Zimmerman HJ. Steatonecrosis-mallory body type. Medicine 1967; 46:141–61. 35. Diehl AM, Goodman Z, Ishak KG. Alcohol-like liver disease in nonalcoholics: a clinical and histologic comparison with alcohol-induced liver injury. Gastroenterology 1988; 195:1056–62. 36. Powell EE, Cooksley WGE, Hanson R, Searle J, Halliday JW, Powell LW. The natural history of nonalcoholic steatohepatitis: a follow-up study of forty-two patients for up to 21 years. Hepatology 1990; 11:74–80. 37. Thomas LB, Popper H, Berk PD, et al. Vinyl chloride induced liver disease. N Engl J Med 1975; 292:17–22. 38. Tamburro CH, Mack L, Popper H. Early hepatic histologic alterations among chemical (vinyl monomer) workers. Hepatology 1984; 4:413–8. 39. Falk H, Creech JL, Heath CW, et al. Hepatic disease among workers at a vinyl chloride polymerization plant. JAMA 1974; 230:59–63. 40. Wagoner JK. Toxicity of vinyl chloride and poly (vinyl chloride): a critical review. Environ Health Persp 1983; 52:61–6. 41. Wu W, Steenland K, Brown D, et al. Cohort and case-control analyses of workers exposed to vinyl chloride: an update. J Occup Med 1989; 31:518–23. 42. ATSDR (Agency for Toxic Substances and Disease Registry) case studies in environmental medicine. Vinyl chloride toxicity. Clin Toxicol 1990; 28:267–86. 43. Liss GM, Greenburg RA, Tamburro CH. Use of serum bile acids in the identification of vinyl chloride hepatotoxicity. Am J Med 1985; 78:68–76.

Medical Surveillance for Occupational Liver Disease 601 44. Thiele DL, Eigenbrodt EH, Ware AJ. Cirrhosis after repeated trichloroethylene and 1,1,1-trichloroethane exposure. Gastroenterology 1982; 83:926–9. 45. Dossing M, Skinhoj P. Occupational liver injury: present state of knowledge and future perspective. Int Arch Occup Environ Health 1985; 56:1–21. 46. Lloyd JW, Decoufle P, Salvin LG. Unusual mortality experience of printing pressman. J Occup Med 1977; 19:543–50. 47. Paganini-Hill A, Glazer E, Henderson BE, Ross RK. Cause-specific mortality among newspaper web pressmen. J Occup Med 1980; 22:542–44. 48. Chiazze L, Ference LD, Wolf PH. Mortality among automobile assembly workers. J Occup Med 1980; 22:520–6. 49. Hasumura Y, Teschke R, Lieber CS. Increased carbon tetrachloride hepatotoxicity, and its mechanism, after chronic ethanol consumption. Gastroenterology 1974; 66:415. 50. McCunney RJ. Diverse manifestations of trichlorethylene. Br J Ind Med 1988; 45:122–6. 51. Nakayama H, Kobayashi M, Takahashi M, et al. Generalized eruption with severe liver dysfunction associated with occupational exposure to trichloroethylene. Contact Derm 1988; 19:48–51. 52. Hodgson MJ, Heyl AE, Van Thiel DH. Liver disease associated with exposure to 1,1,1-trichloroethane. Arch Int Med 1989; 149:1793–8. 53. Texter EC Jr, Grunow WA, Zimmerman HJ. Massive centrizonal necrosis of the liver due to inhalation of 1,1,1-trichloroethane. Gastroenterology 1979; 76:1260. 54. Meckler LC, Phelps DK. Liver disease secondary to tetrachloroethylene exposure. A case report. JAMA 1966; 197:662. 55. Gennari P, Massimo N, Motta R, et al. Gammaglutamyltransferase isoenzyme pattern in workers exposed to tetracholorethylene. Am J Ind Med 1992; 21:661–71. 56. Redlich CA, Beckett WS, Sparer JS. Liver disease associated with occupational exposure to the solvent dimethylformamide. Ann Intern Med 1988; 108:680–6. 57. Redlich CA, West AB, Fleming LE, et al. Clinical and pathological characteristics associated with occupational exposure to dimethylformamide. Gastroenterology 1990; 99:748–57. 58. Guzelian P, Mills S, Fallon HJ. Liver structure and function in print workers exposed to toluene. J Occup Med 1988; 30:791–6. 59. Boewer C, Enderlein G, Wollgast U, Nawka S. Epidemiological study on the hepatotoxicity of occupational toluene exposure. Int Arch Environ Health 1988; 60:181–6. 60. Knight AT, Pawsey CG, Aroney RS, et al. Upholsterers’ glue associated with myocarditis, hepatitis, acute renal failure and lymphoma. Med J Aust 1991; 154:360–2. 61. Axelson O, Gustavson J. Some hygienic and clinical obersvation on styrene exposure. Scand J Work Environ Health 1978; 4(Suppl 2):215–9. 62. Thiess AM, Friedheim M. Morbidity among persons employed in styrene production, polymerization and processing plants. Scand J Work Environ Health 1978; 4(Suppl 2):203–14. 63. Triebig G, Lehrl S, Weltle D, Schaller KH, Valentin H. Clinical and neurobehavioural study of the acute and chronic neurotoxicity of styrene. Br J Ind Med 1989; 46:799–804. 64. Lorimer WV, Lilis R, Fischbein A, et al. Health status of styrene-polystyrene polymerization workers. Scand J Work Environ Health 1978; 4(Suppl 2):220–6. 65. Harrison R, Letz G, Pasternak G, Blanc P. Fulminant hepatic failure after occupational exposure to 2-nitropropane. Ann Int Med 1987; 107:466–8. 66. Hine CH, Pasi A, Stephens BG. Fatalities following exposure to 2-nitropropane. J Occup Med 1978; 20:333–7. 67. Morely R, Eccleston DW, Douglas CP, et al. Xylene poisoning: a report on one fatal case and two cases of recovery after prolonged unconsciousness. Br Med J 1970; 3:442–3.

68. Dossing M, Arloen-Soberg P, Peterson LM, et al. Liver damage associated with occupational exposure to organic solvents in house painters. Eur J Clin Invest 1983; 13:151–7. 69. Lundberg I, Nise G, Hedenborg G, Hogberg M, Vesterberg O. Liver function tests and urinary albumin in house painters with previous heavy exposure to organic solvents. Occup Environ Med 1994; 51:347–53. 70. Chen JD, Wang JD, Jang, JP, Chen YY. Exposure to mixtures of solvents among paint workers and biochemical alternations of liver function. Br J Ind Med 1991; 48:696–701. 71. Sotaniemi EA, Sutinen S, Arranto AJ, Pelkonen RO. Liver injury in subjects exposed to chemicals in low doses. Acta Med Scand 1982; 212:207–15. 72. Dossing M. Noninvasive assessment of microsomal enzyme activity in occupational medicine: present state of knowledge and future perspectives. Int Arch Environ Health 1984; 53:205–18. 73. Charbonneau M, Couture J, Plaa GL. Inhalation versus oral administration of acetone: Effect of the vehicle on the potentiation of CC14-induced liver injury. Toxicol Lett 1991; 57:47–54. 74. Charbonneau M, Tuchweber B, Plaa GL. Acetone potentiation of chronic liver injury induced by repetitive administration of carbon tetrachloride. Hepatology 1986; 6:694–700. 75. Waldron HA, Cherry N, Johnston JD. The effects of ethanol on blood toluene concentrations. Int Arch Environ Health 1983; 51:365–9. 76. Wallen M, Naslund PH, Nordqvist MB. The effects of ethanol on the kinetics of toluene in man. Toxicol Appl Pharmacol 1984; 76:414–9. 77. Liira J, Riihimaki V, Engstrom K. Effects of ethanol on the kinetics of methyl ethyl ketone in man. Br J Ind Med 1990; 47:235–30. 78. Edling C. Anesthetic gases as an occupational hazards: a review. Scand J Work Environ Health 1980; 6:85–93. 79. Neuberger J, Vergani D, Mieli-Vergani G, Davis M, Williams R. Hepatic damage after exposure to halothane in medical personnel. Br J Anaesth 1981; 53:1173–7. 80. Dahlgren B-E. Hepatic and renal effects of low concentrations of methoxyflurane in exposed delivery ward personnel. J Occup Med 1980; 22:817–9. 81. Guzelian PS, Vranian G, Boylan JJ, et al. Liver structure and function in patients poisoned with chlordecone (kepone). Gastroenterology 1980; 78:206–13. 82. Cocco P, Blair A, Congia P, Saba G, Ecca AR, Palmas C. Long-term health effects of the occupational exposure to DDT. A preliminary report. Ann NY Acad Sci 1997; 837:246–56. 83. Tamburro CH. Chronic liver injury in phenoxy herbicide-exposed Vietnam veterans. Environ Res 1992; 59:175–88. 84. Neuberger M, Rappe C, Bergek S, et al. Persistent health effects of dioxin contamination in herbicide production. Environ Res Sect A 1999; 81:206–14. 85. Michalek JE, Ketchum NS, Longnecker MP. Serum dioxin and hepatic abnormalities in veterans of Operation Ranch Hand. Ann Epidemiol 2001; 11:304–11. 86. Brown DP, Jones M. Mortality and industrial hygiene study of workers exposed to PCBs. Arch Environ Health 1981; 36:120–9. 87. Sala M, Sunyer J, Otero R, Santiago-Silva M. Organochlorine in the serum of inhabitants living near an electrochemical factory. Occup Environ Med 1999; 56:152–8. 88. Fischbein A. Liver function tests in workers with occupational exposure to polychlorinated biphenyls (PCBs): comparison with Yusho and Yu-Cheng. Environ Health Perspect 1985; 60:145–50. 89. Pimentel JC, Menezes AP. Liver disease in vineyard sprayers. Gastroenterology 1977; 72:275–83. 90. Zimmerman HJ, Lewis JH. Chemical- and toxin-induced hepatotoxicity. Gastroenterol Clin North Am 1995; 24:1027–45.

602 Liver Diseases 91. Rahman MM, Chowdhury UK, Mukherjee SC, et al. Chronic arsenic toxicity in Bangladesh and West Bengal, India – a review and commentary. J Toxicol Clin Toxicol 2001; 39:683–700. 92. Hoet P, Graf ML, Bourdi M, et al. Epidemic of liver disease caused by hydro chlorofluorocarbons using ozone-sparing substitutes of chlorofluorocarbons. Lancet 1997; 350(9077):556–9. 93. Tamburro CH, Liss GM. Tests for hepatotoxicity: usefulness in screening workers. J Occup Med 1986; 28:1034–44. 94. Wright C, Rivera J, Baetz J. Liver function testing in a working population: three strategies to reduce false-positive results. J Occup Med 1988; 30:693–7.

95. Herip DS. Recommendations for the investigation of abnormal hepatic function in asymptomatic workers. Am J Ind Med 1992; 21:331–9. 96. Hodgson M, Goodman-Klein B, Van Thiel D. Evaluating the liver in hazardous waste workers. Occup Med State Art Rev 1990; 5:67–78. 97. Lee WM. Drug-induced hepatotoxicity. N Engl J Med 2003; 349:474–85. 98. Hay JE, Czaja AJ, Rakela J, Ludwig J. The nature of unexplained chronic aminotransferase elevations of a mild to moderate degree in asymptomatic patients. Hepatology 1989; 9:193–7.

26.2 Disorders of the Gut and Pancreas Stan Lee, Lora E Fleming, Carl A Brodkin

INTRODUCTION AND APPROACH TO THE PATIENT Occupational and environmental diseases of the gastrointestinal (GI) tract often are unrecognized. A variety of toxic exposures in the workplace may cause GI symptoms (see Table 26.2.1). Organic solvents may cause nausea and vomiting by acute solvent intoxication, presumably mediated through their effects on the central nervous system. In addition, exposure to a variety of toxins may cause nonspecific GI symptoms, such as nausea among workers exposed to lead. The prevalence of non-malignant GI diseases of occupational and environmental etiology is unknown. This is due in part to lack of clinical recognition: because GI symptoms are so common, clinicians often fail to consider workplace exposures in the differential diagnosis. The clinical evaluation of GI disease of suspected environmental origin follows the same general principles as that in general medical practice. The history and physical examination should not be limited to the GI system. Toxins that cause GI symptoms, such as metals, solvents, and pesticides, often cause more direct effects on other systems such as the central nervous system or skin. Laboratory tests of GI system anatomy and function are rarely diagnostic of etiology. In contract to infectious diarrhea in an agricultural worker, with a finding of ova and parasites in the stool, organophosphate poisoning in the same worker would have no specific GI finding. As with the physical examination, the laboratory investigation should never be limited to tests of the GI system and should be directed by the history. Nevertheless, acute GI symptoms, such as vomiting, abdominal pain, and severe diarrhea, have to be ameliorated because these symptoms often are what brought the patient to medical attention, not the fact of a toxic exposure. The suspected or established causes of common GI disturbances are summarized in Table 26.2.2 and discussed in detail in the following section.

UPPER GASTROINTESTINAL DISORDERS Esophagitis and gastritis Occupational and environmental causes Most cases of occupationally related esophagitis and gastritis are associated with corrosives and solvents. Accidental exposures to corrosives, which include strong acids (such as sulfuric and chromic acids), fixers (such as formaldehyde), and strong bases (such as lye) cause direct tissue injury. Ingestion injury is the classic route of exposure for the aforementioned toxins. However, alternative routes of exposure, including dermal absorption and inhalation of high concentrations of irritant vapors, may be important with other agents. Grade 2 esophagitis has been reported

in a plastics industry worker accidentally exposed to dimethylacetamide and 1,2-ethanediamine.1 Occupationally related gastritis has also been reported without an ingestion exposure. Solvents often have excellent penetration through both dermal and inhalational routes; for example, up to 10% of topical and 2% of inhaled dimethylacetamide is absorbed into the blood stream.2 Excessive exposure to solvents frequently causes nausea and anorexia, through the central nervous system (see Chapter 40). Solvents additionally exert direct irritational effects on gastric (as well as bowel) mucosae, contributing to symptoms. More severe or localized effects are most likely to occur following ingestion. Gastritis has been observed in worker populations exposed to the solvent dimethylformamide, another solvent with high skin penetration. Chronic gastritis has been reported in a laboratory technician using a solution containing 80% methylmethacrylate for tissue preparation; concomitant skin changes were also present, with contact dermatitis. Interestingly, GI symptoms were reproduced with positive skin patch testing with methylmethacrylate.3 Chronic gastritis has also been observed in several workers following an industrial exposure to methylmethacrylate; the symptoms continued for several years despite withdrawal from exposure.

Clinical course, diagnosis and management Except for a history of ingestion and/or inhalation, occupationally related esophagitis and gastritis present similarly to non-environmentally related cases. Symptoms of esophagitis may include throat pain and dysphagia. Symptoms of gastritis may include upper- to mid-epigastric abdominal pain, with or without food intolerance. The diagnosis and treatment is also similar to nonoccupational cases, but must emphasize elimination or protection from exposure. This can serve as a diagnostic trial as well as a therapeutic intervention.

Peptic ulcer disease (gastric and duodenal) Occupational and environmental causes Peptic ulcer disease, encompassing both gastric and duodenal ulcers, has been reported in higher prevalence among certain occupations, with increased risk observed in a number of epidemic studies. There is a controversial association between occupational stress (either physical or emotional) and an increased prevalence of peptic ulcers; it is a controversy partially due to inconsistent and nonspecific definitions of stress. Traditionally, a number of sedentary, professional jobs with high stress, such as air traffic controllers and executives, have been found to have a higher prevalence of ulcers among workers.4 However, more recent work has shown an increased prevalence of peptic ulcers in unskilled workers as well. In particular, immigrants and manual labor workers

604 Disorders of the Gut and Pancreas Symptom Nausea and vomiting

Constipation Diarrhea

Jaundice

Type of exposure (examples) Central nervous system depressant (organic solvents with acute solvent intoxication) Cholinesterase inhibitors (organophosphates, carbamates) Metabolic Methemoglobin formers (organic nitrogen compounds) Caustic Irritants (epoxy resins, copper and tin fumes) Heavy metals (lead, barium, thallium) Infections (parasites, bacteria) Cholinesterase inhibitors (organophosphates) Metals (arsenic, phosphorus) Hepatotoxins (organic solvents) Hemolytic agents (arsine, naphthalene, phenylhydrazine, stilbene)

Table 26.2.1 Major acute gastrointestinal symptoms and their common occupational and environmental causes

Disorder Upper GI tract disorders Esophagitis/gastritis Peptic ulcer disease Pancreatitis

Lower GI tract disorders Infectious gastroenteritis

Celiac disease Pneumatosis cystoides intestinalis Non-specific toxic GI syndromes

Exposure or occupation (examples) Dimethylformamide Methylmethacrylate Dimethylacetamide Shift work Chlorinated naphthalenes Organophosphate pesticides Organic solvents Scorpion toxin Hospital workers Laboratory workers Food producers Abbatoir workers Farm workers Sewage workers Allergens associated with hypersensitivity pneumonitis (see Chapter 19.6) Trichloroethylene Lead Other heavy metals Organophosphate pesticides

Table 26.2.2 Environmental agents/occupations associated with gastrointestinal disorders

try in Italy, an increased prevalence of peptic ulcer disease and chronic gastritis was noted, which correlated in a dose-related fashion with number of years worked, noise, temperature, vibration, workshift, and workload.8 Ulcer prevalence is higher in shift workers than in daytime workers;9 this increased prevalence is believed to be due to sleep disturbance and altered eating patterns in the shift worker. A study from Denmark found that low socioeconomic status and non-daytime work were associated with an increased risk of gastric ulcer.10 The authors hypothesized that shift workers experienced physiologic and psychologic stress from disruption in circadian rhythms, limitations in family and social life, and sleep difficulties. During deep (e.g., rapid eye movement [REM]) sleep, there is a decrease in gastric acid secretion and an increase in gastric motility. Disruption of sleep processes may, therefore, theoretically increase the risk of ulcer disease.10 In addition to stress, food and smoking may also play a role in peptic ulcer disease. A study in a Japanese plastic processing plant demonstrated an association between peptic ulcers and smoking, as well as family history.5 In another study, fishermen and transportation workers in Norway had an increased prevalence of both gastric and duodenal ulcers compared with 11 other occupational groups. This prevalence was believed to be due to irregular meals, high coffee intake, and heavy smoking.6 Similar findings in a Russian study of railroad workers found an increased prevalence of peptic ulcer disease that correlated with irregular hurried meals of both cold and fried foods.11

Clinical course, diagnosis, and management The diagnosis and treatment do not differ from that in the general medical setting. Management of shift work may be challenging, but regular meals and sleep habits should be emphasized. Acid suppression should be prescribed to effect ulcer healing or symptom resolution. Helicobater pylori should be treated if present. Use of non-steroidal anti-inflammatory drugs and aspirin should be eliminated if possible. Behavioral changes such as smoking cessation and decreased coffee and alcohol consumption should be encouraged.

Pancreatitis Occupational and environmental causes

have a higher prevalence and mortality from peptic ulcer disease (gastric rather than duodenal) than sedentary workers. Among some groups of miners, such as the copper miners in Chile, the incidence of peptic ulcer disease is much higher than in mine administrators or mechanics. Of note, the increases among manual and migrant workers parallel the increased mortality rate among lower socioeconomic classes, and may be a confounding factor in these studies.5,6 The role of physical work, energy expenditure, and other physical factors appears to be important in ulcer prevalence and mortality rates.7 In a study in the metalworking indus-

Toxin-induced pancreatitis may be more common than is now recognized, especially given the high proportion of cases now considered to be idiopathic or drug related. Extremely high exposures to a variety of occupational toxins (such as the chlorinated naphthalenes), that are also associated with severe liver damage and systemic toxicity, have been associated with diffuse pancreatic damage.12 Isolated islet cell injury has been reported with accidental ingestion of the rodenticide pyriminil (Vacor). The organophosphate pesticides have been associated with pancreatitis and other pancreatic disorders. Accidental ingestion of the organophosphate pesticide O-ethyl-S-

Lower Gastrointestinal Disorders 605 phenylethylphosphophenodithioate leads to pancreatitis, frequently with pancreatic pseudocyst formations, during the acute stages of organophosphate poisoning. A male farm worker exposed to the organophosphate Q Dimethoate developed pancreatitis with acute organophosphate poisoning.13 Two out of nine patients who ingested parathion developed painless acute hemorrhagic pancreatitis.14 Acute and chronic pancreatitis have also been associated with occupational exposures to organic solvents.15,16 Numerous cases have now been described in association with occupational exposure to perchloroethylene, trichloroethylene (TCE), mineral spirits, solvent-based paints, and diesel fuel. There have been at least two case reports of pancreatitis associated with occupational exposure to the solvent dimethylformamide (a known hepatotoxin) with no history of concurrent alcohol use; full recovery was apparent following removal from exposure.17 Experimental animal evidence also indicates that this solvent causes pancreatic damage. Acute pancreatitis has also been reported with scorpion bites (Tityus trintatis), both occupationally and environmentally, in Trinidad.

Pathogenesis Experimental models of poisoning in pigs and dogs with organophosphates reveal increases in amylase levels and increased intraductal pressures within the pancreas. The postulated mechanism is the stimulation of parasympathetic pathways to the pancreas through cholinesterase inhibition, which, in turn, augments the secretory flow, increasing intraductal pressure. This response can be attenuated experimentally through pretreatment with atropine. Pancreatitis associated with scorpion toxin appears to be related to increased cholinergic stimulation, as well as a direct toxic effect.18,19

other occupational GI diseases, removal of the affected individual can be a useful diagnostic as well as therapeutic intervention.

LOWER GASTROINTESTINAL DISORDERS Infectious gastroenteritis Occupational and environmental causes Occupationally related gastroenteritis has been reported in workers and their families from a number of occupations (Table 26.2.3). In particular, parasitic and bacterial forms of gastroenteritis have been noted in specific occupations (such as healthcare workers, laboratory technicians, and sewage workers) that involve contact with infected materials.20–24 Zoonotic illness resulting in gastroenteritis has been reported among animal handlers and laboratory technicians; agricultural workers and other rural manual laborers (and their families) have been noted to have increased parasitic and bacterial infestations, probably due to infected water supplies, poor hygienic conditions, and the use of night soil (human waste) as a fertilizer.

Clinical course, diagnosis, and management While the signs and symptoms, as well as the diagnosis and treatment, of occupationally related gastroenteritis are the same as for any gastroenteritis, management should emphasize examination of family members who may be infected from common sources.

Celiac disease

Clinical course, diagnosis, and management

Occupational and environmental causes

Clinical features of pancreatitis of environmental or occupational etiology do not appear to differ from those of other established causes. The key difference in treatment is prevention of re-exposure to the suspected toxin. As with

Hypersensitivity pneumonitides such as farmer’s lung and bird fancier’s lung have been associated with celiac disease in some cases. Celiac disease involves a malabsorption syndrome, with duodenal or jejunal villous changes

Occupation Occupations with infected materials Laboratory workers Healthcare workers (patient care) Food production Sewage workers

Pathogens Shigella Brucellosis, typhoid Tularemia, tuberculosis Salmonella typhimurium enteritis Cryptosporidiosis Campylobacter jejuni enteritis Parasitic infections (Entamoeba histolytica, Giardia lamblia) Unclassified

Occupations with animal contact Laboratory worker (working with coyotes) Campylobacter jejuni enteritis Abattoir workers Salmonella Agricultural and rural manual occupations Agricultural workers Gastrointestinal parasitoses Road workers Gastrointestinal parasitoses Table 26.2.3 Occupational gastroenteritis: occupations and pathogens

Reference Kolavic et al., 199737 Pike, 197922; Pike, 197623 Steckelberg et al., 198838; Standaert et al., 199439 Koch et al., 198524 Jones, 197940 Clark et al., 198421; Hays, 197741 Hickey, 197542 Khuder et al., 199820 Fox et al., 198943 Deseö & Engeli, 197944 Sterba et al., 198845; Ungar et al., 198646; Ortiz, 198047 Latham et al., 198348

606 Disorders of the Gut and Pancreas observed on biopsy; it can present clinically, prior to or concomitant with the pulmonary disease.25 There is still considerable controversy regarding the actual existence of this disease entity.26 Berrill’s investigation of 42 patients with bird fancier’s lung (including exposure history, diffuse lung disease, and precipitins to bird serum) showed that eight persons had villous atrophy on jejunal biopsy.25 In another large series, four of 57 patients with farmer’s lung had celiac disease on biopsy; low red cell folate levels and multiple food antibodies were also observed.27 Other investigations have failed to show celiac disease on biopsy of patients who had proven bird fancier’s lung, and this may represent as infrequent complication.

Pathogenesis This environmentally related enteropathy appears to be distinct from traditional celiac disease. There often are specific antibodies to birds or molds that are distinct from the avian antibodies seen with common celiac disease (these antibodies are believed to be antigens to hen’s egg yolk).28 Specific respiratory responses to provocation inhalation tests are observed with these patients that are not seen in patients with common celiac disease. In addition, although these patients have antibodies to gluten, often these are not antibodies to the gliadin fraction of gluten (seen with common celiac disease). Of interest, this enteropathy seems to respond to removal of the affected individual from exposure, in combination with a glutenfree diet. One etiologic theory posits that there may be a genetic predisposition that is triggered by exposure to specific antigens; notably, both celiac disease and some types of fibrosing alveolitis, including farmer’s lung disease, are more common in persons of the HLA B8 genotype.

Clinical course, diagnosis, and management The signs and symptoms of this disease entity are due to the concurrent involvement of the GI and respiratory systems, including mouth ulcers, food intolerance, abdominal pain, diarrhea, constipation, as well as weight loss, dyspnea, general malaise, and bronchospasm. These symptoms may coexist or can precede one another. A history of occupational or heavy environmental exposure to hay, straw, or birds should raise clinical suspicion in an individual with lower GI symptoms. A duodenal or jejunal biopsy showing villous atrophy as well as antigen testing (e.g., avian precipitins [bird fancier’s lung] or Micropolyspora faeni [farmer’s lung]) are diagnostic. For diagnostic issues regarding the respiratory component, please refer to Chapter 19.6. In addition, there should be no evidence for malignancy or collagen vascular disease. Of note, there are reports of co-existing celiac disease in spouses, supporting a common source exposure. Removal of the individual from exposure in combination with dietary restrictions (e.g., gluten free, egg free) appear to be key factors in management of the condition. Sometimes, more aggressive treatment, such as corticosteroids, is necessary, particularly for the lung disease. The prognosis is good, especially with removal of the individual from exposure.

Pneumatosis cystoides intestinalis Occupational and environmental causes Pneumatosis cystoides intestinalis (PCI) is a relatively rare, usually benign condition, characterized by the formation of multiple intramural gas-filled cysts along some portion of the GI tract involving most commonly, though not invariably, the lower gastrointestinal tract. Over 350 cases have been reported in Japan, many of them associated with occupational exposure to TCE.29–31 There appear to exist two etiologic groups of PCI. One is primary or idiopathic PCI, with no particular responsible or associated abnormalities; it is a relatively benign condition that occurs more frequently in women. The other group, so-called secondary PCI, is seen with a broad variety of associated lesions, including intestinal obstruction (with concurrent pyloric stenosis related to peptic ulcer disease), chronic obstructive pulmonary disease, and connective tissue diseases. Since 1952, when the first cases were reported, the majority of patients have fallen in the second group, with lesions predominantly in the small intestine. More recently, in Japan, there have been a series of reports of primary PCI, predominantly in the large intestine (especially the sigmoid colon); this condition also has been called pneumatosis cystoides coli (PCC). Many of these PCI (or PCC) patients have a history of occupational exposure to TCE, for degreasing metal parts and products. In cases of TCE-associated PCI, TCE has been found in gas through endoscopic collection in the cystic spaces, and the metabolite trichloroacetic acid (TCA) has been found in the urine and bile of patients. However, the etiologic mechanism of TCE is not well understood.32 Another possible mechanism is the inhibition of bacterial hydrogen consumption by alkyl halides such as TCE. This may lead to increased net hydrogen gas production, resulting in ‘counterperfusion supersaturation’ and formation of gas cysts.33

Clinical course, diagnosis, and management The symptoms of PCI are non-specific and include abdominal fullness, tenderness, and pain; constipation; and a frothy mucous discharge from the rectum (described by patients as foamy tomato juice). Pneumoperitoneum occurs frequently, but because of the sterility of the gas collections, patients rarely have complications of peritonitis.34 A period of up to 10 years may be required before symptoms develop. On barium enema, polypoid changes are noted, with multiple elevated lesions found on sigmoidoscopy. Needle aspiration of these areas reveals gas. No malignancy or other source for the pain and hematochezia is identified in the workup. Associated etiologic conditions should be sought to exclude secondary PCI. Removal of the individual from exposure to TCE and, in some cases, oxygen therapy are the major forms of treatment for PCI.35 Some studies have examined the use of metronidazole to decrease anaerobic bacteria and reduce production of luminal gas. Primary PCI often disappears spontaneously, with or without other interventions. For secondary PCI, treatment of the underlying disease process

Lower Gastrointestinal Disorders 607 is generally necessary. With pneumatosis secondary to TCE exposure, removal of the individual from further exposure is associated with a good prognosis.

Non-specific toxic gastrointestinal syndromes Lead poisoning Lower gastrointestinal symptoms have been considered a hallmark of lead intoxication, both in children and in adults. Symptoms of colic with either constipation or diarrhea, sometimes severe enough to mimic an acute abdomen, are particularly dominant in acute poisoning, in which a rapid rise in tissue lead levels occurs. In more chronic cases, GI symptoms are less apparent. The effect on the GI tract is probably, in part, due to inhibition of autonomic function of intestinal smooth muscle. In its most extreme form, this condition may lead to toxic megacolon, which is well described in children. Management of lead toxicity involves removal of the individual from exposure in every case, and careful evaluation to exclude other remediable GI lesions such as ulcers or malignancy. In acute lead poisoning, chelation therapy usually results in prompt relief of GI symptoms. Chelation may be accomplished by intravenous calcium disodium EDTA therapy or oral meso-2,3-dimercaptosuccinic acid (DMSA; Succimer) therapy.36

Other metals Many other metals, including aluminum, arsenic, barium, copper, iron, mercury, and thallium, can cause gastroenteritis with anorexia, and a combination of upper GI (e.g., nausea, vomiting) and lower GI changes in bowel habits. In general, these toxicities are related to acute and subacute exposures, especially through the oral route, and less frequently with long-term exposures. The pathogenesis is not well understood in most cases, but, in general, it appears to involve some disruption of epithelial cell function, leading to cellular necrosis in severe cases.

Cholinesterase inhibitors Methylcarbamates and organophosphates inhibit cholinesterase, and may cause anorexia, vomiting, cramps, and diarrhea owing to their cholinergic (muscarinic) effects on bowel function. Increased motility and intestinal secretion, and impaired salt and water reabsorption probably account for the major lower GI symptoms that may occur in acute poisoning. The treatment is removal of the individual from exposure; atropine inhibits the muscarinic effects acutely (see Chapter 48).

References 1. Marino G, Anastopoulos H, Woolf AD. Toxicity associated with severe inhalational and dermal exposure to dimethylacetamide and 1,2-ethanediamine. J Occup Med 1994; 36:637-41. 2. Kennedy GL, Pruett JW. Biologic monitoring for dimethyl acetamide: measurement for 4 consecutive weeks in a workplace. J Occup Med 1989; 31:47-50.

3. Mathias CGT, Caldwell TM, Maibach HI. Contact dermatitis and gastrointestinal symptoms from hydroxyethylmethacrylate. Br J Dermatol 1979; 100:447-9. 4. Dunn JP, Cobb S. Frequency of peptic ulcer among executives, craftsmen and foremen. J Occup Med 1962; 4:343-8. 5. Araki S, Goto Y. Peptic ulcer in male factory workers: a survey of prevalence, incidence and aetiological factors. J Epidemiol Community Health 1985; 39:82-5. 6. Ostensen H, Burhol PG, Stormer J, Bonnevie O. The incidence of peptic ulcer disease related to occupation in the northern part of Norway. Scand J Gastroenterol 1985; 20:79-82. 7. Sonnenberg A, Sonnenberg GS, Withers W. Historic changes of occupational work load and mortality from peptic ulcer in Germany. J Occup Med 1987; 28:756-61. 8. Magni G, Rizzardo R, De Leo D, Salmi A. Adverse environmental factors, peptic ulcer, and chronic gastritis in a metalworking industry. Med Lav 1984; 75:215-20. 9. Segawa K, Nakazawa S, Tsukamoto Y, et al. Peptic ulcer is prevalent among shift workers. Dig Dis Sci 1987; 32:449-53. 10. Tuchsen F, Jeppesen HJ, Bach E. Employment status, nondaytime work and gastric ulcer in men. Int J Epidemiol 1994; 23:365-70. 11. Zhangabylov AK, Bekisheva AS. [Analysis of the etiologic significance of nutrition factors in the occurrence of peptic ulcer in railroad workers.] Vopr Pitan 1989; (3):22-5 (article in Russian). 12. Braganza JM, Jolley JE, Lec WR. Occupational chemicals and pancreatitis. Int J Pancreatol 1986; 1:9-19. 13. Marsh WH, Vukov GA, Conradi EC. Acute pancreatitis after cutaneous exposure to an organophosphate insecticide. Am J Gastroenterol 1988; 83:1158-60. 14. Lankisch PG, Muller CH, Niederstadt H, Brand A. Painless acute pancreatitis subsequent to anticholinesterase insecticide (parathion) intoxication. Am J Gastroenterol 1990; 85:872-5. 15. McNamee R, Braganza JM, Hogg J, et al. Occupational exposure to hydrocarbons and chronic pancreatitis: a case-referent study. Occup Environ Med 1994; 51:631-7. 16. Hotz P, Pilliod J, Bourgeois R, Boillat MA. Hydrocarbon exposure, pancreatitis and bile acids. Br J Ind Med 1990; 47:833-7. 17. Chary S. Dimethylformamide: a cause of acute pancreatitis? Lancet 1974; ii:356 (letter). 18. Dressel TD, Goodale RL, Zweber B, et al. The effect of atropine and duct decompression on the evaluation of diazinoninduced acute canine pancreatitis. Ann Surg 1982; 195:424-34. 19. Gallagher S, Sankaran H, Williams J. Mechanism of scorpion toxin-induced enzyme secretion in rat pancreas. Gastroenterology 1981; 80:970-3. 20. Khuder SA, Arthur T, Bisesi MS, et al. Prevalence of infectious diseases and associated symptoms in waste water treatment workers. Am J Ind Med 1998; 33:571-7. 21. Clark CS, Linneman CC, Clark JG, Gartside PS. Enteric parasites in workers occupationally exposed to sewage. J Occup Med 1984; 26:273-5. 22. Pike RM. Laboratory-associated infections: incidence, fatalities, causes, and prevention. Ann Rev Microbiol 1979; 44:41-66. 23. Pike RM. Laboratory-associated infections: summary and analysis of 3,921 cases. Health Lab Sci 1976; 13:105-14. 24. Koch KL, Phillips DJ, Aber RC, Current WL. Cryptosporidiosis in hospital personnel. Ann Intern Med 1985; 102:593-6. 25. Berrill WT, Fitzpatrick PF, Macleod WM, et al. Bird fancier’s lung and jejunal villous atrophy. Lancet 1975; 2:1006-8. 26. Hendrick DJ, Faux JA, Anand B, et al. Is bird fancier’s lung associated with coeliac disease? Thorax 1978; 33:425-8. 27. Turton CW, Turner-Warwick M, Owens R, et al. Red cell folate levels, food antibodies and reticulin antibodies in farmer’s lung – is there an association with coeliac disease? Br J Dis Chest 1983; 77:397-402. 28. Faux JA, Hendrick DJ, Anand B. Precipitins to different avian serum antigens in bird fancier’s lung and coeliac disease. Clin Allergy 1978; 8:101-8.

608 Disorders of the Gut and Pancreas 29. Hosomi N, Yoshioka H, Kuroda C, et al. Pneumatosis cystoides intestinalis: CT findings. Abdom Imaging 1994; 19:137-9. 30. Ogata M, Kihara T, Kamoo R, et al. A report of a worker suffering from pneumatosis cystoides intestinalis following trichloroethylene exposure. Ind Health 1988; 26:179-82. 31. Yamaguchi K, Shirai T, Shimakura K, et al. Pneumatosis cystoides intestinalis and trichloroethylene exposure. Am J Gastroenterol 1985; 80:753-7. 32. Kaneko T, Saegusa M, Tasaka K, et al. Immunotoxicity of trichloroethylene: a study with MRL -1pr/1pr mice. J Appl Toxicol 2000; 20:471-5. 33. Florin THJ. Alkyl halides, super hydrogen production and the pathogenesis of pneumatosis cystoides coli. Gut 1997; 41:778-84. 34. Boerner RM, Fried DB, Warshauer DM, Isaacs K. Pneumatosis intestinalis: two case reports and a retrospective review of the literature from 1985 to 1995. Dig Dis Sci 1996; 41:2272-85. 35. Forgacs P, Wright PH, Wyatt AP. Pneumatosis cystoides intestinalis treated by oxygen breathing. Lancet 1979; i:579-82. 36. Levin SM, Goldberg M. Clinical evaluation and management of lead-exposed construction workers. Am J Ind Med 2000; 37:23-43. 37. Kolavic SA, Kimura A, Simons SL, et al. An outbreak of Shigella dysenteriae type 2 among laboratory workers due to intentional food contamination. JAMA 1997; 278:396-8. 38. Steckelberg JM, Terrell C, Edson RS. Laboratory-acquired Salmonella typhimurium enteritis: association with erythema nodosum and reactive arthritis. Am J Med 1988; 85:705-7. 39. Standaert SM, Hutcheson RH, Schaffner W. Nosocomial transmission of Salmonella gastroenteritis to laundry workers

40. 41.

42.

43.

44.

45.

46.

47.

48.

in a nursing home. Infect Control Hosp Epidemiol 1994; 15:22-6. Jones A. Campylobacter enteritis in a food factory. Lancet 1979; i:618-9. Hays BD. Potential for parasitic disease transmission with land application of sewage plant effluents and sludge, review paper. Water Res 1977; 11:583-95. Hickey JL, Reist PC. Health significance of air-borne microorganisms from waste water treatment processes. J Water Pollut Control Fed 1975; 47:2741-57. Fox JG, Taylor NS, Penner JL, et al. Investigation of zoonotically acquired Campylobacter jejuni enteritis with serotyping and restriction endonuclease DNA analysis. J Clin Microbiol 1989; 27:2423-5. Deseö L, Engeli P. [Symptomless enteritis-salmonella excreta in an abbatoir.] Schweiz Med Wochenschr 1979; 109:1995-9 (article in German). Sterba J, Ditrich O, Prokopic J, Kadlcik K. Gastrointestinal parasitoses discovered in agricultural workers in South Bohemia, Czechoslovakia. Folia Parasitol 1988; 35:169-73. Ungar BLP, Iscoe MHS, Bartlett JG. Intestinal parasites in a migrant farmworker population. Arch Intern Med 1986; 146:513-15. Ortiz JS. The prevalence of intestinal parasites in Puerto Rican farm workers in western Massachusetts. Am J Public Health 1980; 70:1103-5. Latham MC, Wolgemuth JC, Hall A. Nutritional status, parasitic infections and health of roadworkers in 4 areas of Kenya: Part II Kirinyaga and Murang’a districts, the Highlands. East Afr Med J 1983; 60:75-81.

Chapter 27 Endocrine and Reproductive Disorders 27.1 Endocrine Disorders Ulrike Luderer, Mark R Cullen During the past decade, interest in the effects of environmental agents on the endocrine system has burgeoned. This is, in part, due to the concern that environmental contaminants, acting via hormonal mechanisms, may be responsible for a number of adverse health outcomes in humans and wildlife. It has been hypothesized that these environmental endocrine disrupters may be playing roles in the etiology of breast, testicular, and prostate cancer, decreased fertility, endometriosis, and abnormalities of male reproductive system development such as hypospadias and cryptorchidism.1–3 Although much of the impetus has come from concern about the reproductive effects of these agents, new knowledge has also begun to accumulate regarding their impact on thyroid, adrenal, and pituitary function. At the same time, study of the effects of the environment on the endocrine functions of gastrointestinal organs, and bone and calcium metabolism has begun, albeit in a more limited way to date. Collectively, these aspects are the subjects of this chapter. Related discussions of the reproductive effects appear in Chapter 27.2, while the subject of neoplasms of the endocrine and reproductive organs is covered in Chapters 30.7 and 30.8. Although understanding of this potentially enormous subject remains relatively limited compared with other targets of environmental factors, certain principles regarding non-reproductive endocrine disorders can be stated that simplify the practitioner’s task. First, with the possible exception of the potent rodenticide Vacor, it appears that agents that affect endocrine functions other than reproduction tend to be ones with diffuse effects on other organs as well, such as heavy metals, pesticides, and solvents. In general, therefore, the search for environmental factors is most relevant to patients with medical problems in addition to those problems that might be attributable to endocrinopathy, such as central nervous system (CNS), hepatic, respiratory, dermal, and, importantly, reproductive disorders. In other words, it would be uncommon, given our present understanding, to find an endocrine disorder as the sole manifestation of environmental toxicity in adults. This is in marked contrast to agents with primary reproductive effects that may often dominate the clinical picture of toxicity, obscuring other less salient effects. Subclinical effects, however, could precede other measurable effects, because even subtle alterations may be potentially useful as biomarkers in exposed populations. The advantage they provide over merely measuring exposure levels is that changes in these endpoints indicate that sufficient absorption has occurred to cause a physiologic effect. Interventions to reduce expo-

sure can then be instituted before the onset of clinically significant effects. The corollary of the first principle is that environmental agents typically cause subtle, rather than profound disturbances of endocrine function, except in cases of massive exposures in adults. Classic presentations of organ failure or hyper-reactivity, such as myxedema, thyroid storm, pituitary apoplexy, and addisonian crisis (adrenal insufficiency) are not part of the spectrum of endocrinopathies described in human cases or predicted by dose–response testing in animals or cell cultures. Rather, the effects of environmental agents on endocrine functions other than reproduction tend to be subtle and easily masked, unless they are specifically considered and tested for. However, given our increasing understanding of the importance of endocrine balance for a wide range of crucial functions, such as growth and development, response to stress, and risk for longterm cardiovascular disease, these relatively subtle effects of the environment may be more important than previously appreciated, particularly on a population basis. Indeed, recent work suggests that even small disruptions of thyroid homeostasis in utero and during early postnatal development may have significant adverse effects on the developing brain. For example, it has been long known that infants with congenital hypothyroidism will develop mental retardation and severe motor abnormalities if not treated from birth with thyroxine; however, only recently has it begun to be appreciated that even with thyroxine replacement from birth, these children may manifest subtle, but measurable, deficits of IQ, motor function, and coordination.4

MECHANISMS OF ENDOCRINE TOXICITY There are a variety of mechanisms by which toxicants can affect endocrine function (Table 27.1.1). Induction of cell death within an endocrine gland or tissue is used therapeutically in the treatment of Graves’ disease by means of ablative therapy with radioactive iodine. Induction of apoptosis or programmed cell death in ovarian follicle cells also appears to be one mechanism by which the chemotherapeutic drug cyclophosphamide causes ovarian failure. Many diverse compounds have been shown to disrupt endocrine function by binding to hormone receptors, acting as receptor agonists or antagonists (see Chapter 45). The organochlorine insecticide methoxychlor is an estrogen receptor agonist, while

610 Endocrine Disorders Mechanism Induction of cell death Hormone receptor binding Transport protein binding Altering hormone synthesis or secretion Altering hormone metabolism Altering hormone receptor levels Modulating intracellular signaling pathways

Example

References

Cyclophosphamide, radioactive iodine Methoxychlor, p,p′-DDE* Polychlorinated biphenyls Perchlorates, thiocyanates, styrene Heptachlor, chlordane Gonadotropin-releasing hormone agonists Lithium

69 70,71 16 38,47 72 73 38

*1,1-dichloro-2,2′-bis(p-chlorophenyl) ethylene

Table 27.1.1 Mechanisms of endocrine toxicity

1,1-dichloro-2,2′-bis(p-chlorophenyl) ethylene (p,p′-DDE), a metabolite of another organochlorine insecticide, 1,1,1trichloro-2,2′-bis(p-chlorophenyl) ethane (DDT), is a potent androgen receptor antagonist. Yet other compounds compete with endogenous ligands for non-receptor transport proteins. Hydroxylated metabolites of various polychlorinated biphenyls have been demonstrated to potentially bind transthyretin, disrupting thyroid homeostasis. Some endocrine toxicants act by altering hormone synthesis or secretion. For example, the aromatic solvent styrene increases pituitary prolactin secretion, and thiocyanates and perchlorates inhibit thyroid hormone synthesis by blocking iodine uptake into the thyroid. Endocrine homeostasis can also be disturbed by compounds that alter hormone metabolism. The cyclodiene insecticides heptachlor and chlordane are potent inducers of hydroxylases that metabolize sex steroids, resulting in reduced circulating levels of these hormones. Additional mechanisms by which toxicants might disrupt endocrine function are by up- or down-regulating hormone receptor levels or by interfering with hormonal intracellular signaling pathways. For example, lithium suppresses thyroid hormone release by inhibiting cyclic AMP-mediated effects of thyroid-stimulating hormone (TSH). A useful mechanistic distinction for understanding the developmental effects of endocrine toxicants is that of organizational versus activational effects of hormones.5 Organizational effects occur early during development, usually before birth, and are permanent. In contrast, activational effects usually occur during adult life and are transient. Organizational effects of hormonally active toxicants usually require lower doses and shorter exposures during critical windows of development than do activational effects. Examples of organizational effects of hormones include the stimulation by androgens of male reproductive duct and gland development, and the masculinization of the CNS by exposure to estrogens in utero. Examples of activational effects include the stimulation of thyroid gland hyperplasia by TSH or of uterine endometrial proliferation by estrogens.

CLINICAL EVALUATION History and physical examination For the reasons outlined earlier, the typical historic and physical findings associated with primary endocrine

diseases are not likely to be found in patients with endocrine effects resulting from environmental factors. In fact, where classic findings such as frontal bossing, lid lag, galactorrhea, and hyperpigmentation are uncovered, serious consideration must be given to non-environmental causes such as autoimmune disease or tumors. Nonetheless, any suspicion of endocrine dysfunction should prompt a thorough review of systems for each of the classic historic features of primary endocrine disturbance and a careful physical examination that includes a search for the well-known physical findings. Far more relevant to uncovering endocrine effects of environmental agents are the subtle and less specific consequences of hormonal imbalances. The client’s medical history should include information on each of the following topics: • sleep disturbance or changes in energy level or mood; • alterations in weight, appetite, and bowel function; • sexual interest and function, and, in women, menstrual changes; • changes in temperature perception, sweating, or flushing; and • alterations of body habitus, hair growth, and skin texture. In addition to these specific inquiries, the history should also include a careful search for other toxic effects of suspected causal agents, for the reasons mentioned earlier. Often, these effects may overlap with or obscure endocrinerelated phenomena, especially those that involve the CNS, gastrointestinal, or reproductive symptoms. The presence of such symptoms, although initially confusing, should heighten rather than diminish suspicion of environmentally induced endocrinopathy. These non-specific symptoms mandate a search, usually requiring laboratory tests, to identify effects on each organ that may be contributing to the symptoms. It should not be presumed that identification of one possible disorder, such as mild encephalopathy, precludes the need to evaluate the integrity of endocrine function when exposure to a suspect agent has occurred. The same general principle applies to the physical examination. Although careful examination may uncover a very revealing and localizing sign, such as a goiter or change in a secondary sexual characteristic, the greater usefulness of the examination is in identification of more typical, non-endocrine markers of toxicity from the environmental agent in question, such as Mees lines (skin

Disorders of Thyroid Function 611 lesions associated with heavy metal exposure) or neurologic changes due to solvents, organophosphates, organochlorine, or heavy metals. A second important role for the physical examination is to carefully evaluate evidence of non-environmental causes for endocrinopathy such as a tumor (e.g., if milk can be expressed from the breast of a non-lactating woman).

Laboratory evaluation Given the subtle, non-specific nature of historic clues and physical findings for endocrinopathy of environmental origin, the appropriate use of laboratory tests is essential. There are four specific goals in the use of these tests: • detection of mild, often subclinical alterations of function; • localization of the injury within the hierarchy of endocrine regulation (e.g., hypothalamus or pituitary versus end organ); • exclusion of tumor or other primary endocrine organ failure due to non-environmental causes; and • establishment of markers of exposure to a causal agent or evidence of other organ system damage from the agent. Satisfactory accomplishment of the first three goals requires a detailed understanding of the available strategies for testing and localizing endocrine lesions, which is beyond the scope of this text. Certain generalization is worth noting here regarding available tests.

Tests of end-organ function The circulating levels of virtually every hormone can now be measured directly, including thyroid hormones, adrenal steroids and amines, anterior and posterior pituitary secretions, parathormone, and 1,25-dihydroxycholecalciferol. For some of these hormones, levels are sufficiently stable over time so that a single measurement accurately reflects organ function. Examples of these include thyroid hormones and cholecalciferols. For others, such as cortisol and insulin, blood levels fluctuate so rapidly that baseline function cannot be adequately assessed from a single or even from multiple random measures; instead, some integrated measure is needed, such as collection of steroid metabolites in a timed urine collection, or the hormone must be measured under specified conditions known to maximize or suppress glandular activity, such as fasting or glucose ingestion for insulin secretion. For some hormones, such as insulin, indirect measurements of hormonal function are often more readily available, such as blood glucose levels. However baseline function of the end organs is assessed, it must be remembered that the endocrine system as a whole is finely balanced to ensure homeostasis even in the face of injury. For example, a partially diseased thyroid gland may still be capable of secreting an adequate (i.e., normal) level of thyroxine (T4) by stimulating the pituitary compensatory secretion of TSH. Thus, a subtle lesion in the thyroid axis would not necessarily be revealed by simple measurement of the end product, but would

become apparent if TSH were simultaneously measured. For other hormones, even simultaneous measurement of trophic hormone and its end product may mask a lesion. For example, mild adrenal insufficiency may not be apparent from cortisol levels or urinary excretion of adrenal metabolites under normal conditions, but they may be significant or even life threatening under stresses such as infection or bleeding. An array of stress tests have been devised that can allow detection of more subtle lesions that are insufficient to alter function under normal conditions. The best known example of such is the glucose tolerance test, which indirectly assesses the adequacy of insulin regulation by measurement of timed glucose levels after a challenge. In the reverse direction, administration of insulin to induce mild hypoglycemia is an excellent way to assess whether the adrenal axis is intact, because lowering of blood sugar is a very potent stimulus for cortisol secretion. The same stress should also induce an outpouring of growth hormone, even if baseline levels are low. It is important to recognize, however, that many of these stresses are inherently dangerous, especially in patients with suspected defects of endocrine function. The tests should be performed under tightly controlled circumstances and always by individuals experienced in their administration and interpretation.

DISORDERS OF THYROID FUNCTION The ability of environmental factors to affect the function of the thyroid gland has been appreciated for centuries because of the occurrence of endemic goiters. Although iodine deficiency was recognized as an important cause of goiter before the days of dietary supplementation, the presence of various goitrogens in medicines, food, and drinking water was also recognized. These compounds include perchlorates and thiocyanate, which inhibit iodine transport into the thyroid, and a number of compounds that inhibit organification of iodine, including the metal cobalt and several classes of cyclic organic compounds, such as substituted phenols. Particularly concerning are the developmental effects of such alterations, given the important role of T4 in CNS development. A recent study of electroplating workers with chronic cyanide exposure, which causes thiocyanate accumulation, demonstrated reduced serum T4 and tri-iodothyronine (T3), and elevated TSH concentrations relative to controls.6 Perchlorate has historically been used to treat Graves’ disease, effectively lowering T4 levels; however, it was discontinued due to a number of reports of aplastic anemia. Environmental perchlorate contamination of drinking water has been linked to reduced T4 and increased TSH in newborns.7 While increased rates of congenital hypothyroidism have not been observed, this endpoint is not sensitive for detecting subclinical thyroid dysfunction.8 Two studies of workers with inhalational perchlorate exposure, in contrast, did not find evidence of altered thyroid function,9,10 although the applicability of these findings to

612 Endocrine Disorders pediatric populations is unclear. More recent investigations of orally administered perchlorate in human volunteers have documented inhibition of I-123 uptake by the thyroid at low doses (0.01–0.043 mg/kg/day).11–13 These findings indicate inhibition of the sodium-iodide symporter, with mild changes in thyroid hormone levels at higher doses (0.5 mg/kg/day) in the short term.3 Additional health concerns regarding perchlorate have focused on rodent models, demonstrating a lowest observable adverse effect level (LOAEL) of 0.01 mg/kg/day based on brain morphometric and thyroid hormonal changes in rat pups, with thyroid follicular epithelial cell hyperplasia observed at 0.1 mg/kg/day.14 In aggregate, these investigations have led to recommendations for lowered reference doses, with proposed reference doses as low as 1 ppb for drinking water (Drinking Water Equivalent Level [DWEL]).14 The effect of ionizing radiation on the thyroid gland has also been appreciated for many years.15 Low to moderate doses of local irradiation to the gland, either through an external X-ray study or by absorption of radioactive iodine, have been responsible for the development of benign and malignant tumors (see Chapter 30.7). High doses of ionizing radiation are ablative and are the basis for one of the standard treatments for Graves’ disease. To date, clinical studies do not suggest that the low, carcinogenic doses are associated with functional changes; hypothyroidism has not been described as a relevant effect of radiation exposure except in the setting of very high therapeutic doses. More recently, the effects of environmental contaminants such as heavy metals, solvents, pesticides, and related compounds have been recognized as causing functional thyroid effects either clinically or in experimental animals. In humans, depressed levels of thyroid function have been correlated with exposures to lead, carbon disulfide, polybrominated biphenyls (PBBs), polychlorinated biphenyls (PCBs), dioxins, and furans. These agents have been shown to act by a variety of mechanisms in laboratory studies, including enhancement of T4 glucuronidation, and thus excretion, by PCBs, DDT, and chloroacetanilides; binding to serum thyroid hormone transport proteins by PCBs;16 and inhibition of peripheral T4 deiodination by methoxychlor.17 As with the goitrogens discussed above, the effects of disruption of thyroid homeostasis by these compounds during prenatal and postnatal development have been of particular concern because of the importance of thyroid hormone to normal brain development.18 During the first trimester of pregnancy, the developing embryo is totally dependent on maternal T4, and even subclinical decreases in the maternal free thyroxine index during early gestation have been associated with impaired psychomotor development and reduced IQ.18 The fetus begins to synthesize T4 around week 10 to 12 of gestation, but some T4 is still transported from the mother throughout gestation. Thus, the severe sequelae of congenital hypothyroidism are seen after birth, with the withdrawal of the maternal source of T4.4 Thyroid function in lead workers has been studied quite extensively. Results suggest mild depression of function in

chronically and heavily exposed adults,19,20 without substantial effects at lower levels.21 Challenge testing has suggested that the level of the lesion is at the hypothalamus, although neither pituitary nor direct thyroid effects, as predicted by animal studies, could be excluded. Significantly, a careful study of young inner-city children failed to reveal any relationship between lead exposure and thyroid function, although screening for such an effect in the face of childhood lead poisoning is recommended in view of the ramifications of even mild hypothyroidism on child development.22 Early reports of depressed thyroid function in carbon disulfide-exposed workers have been neither clinically studied nor confirmed in more recent cohorts.23 As is generally true for endocrine effects, other clinical effects of carbon disulfide poisoning, including neuropathy and cardiovascular disease, tend to dominate the clinical picture in the only available reports. The study of the agricultural communities in Michigan accidentally exposed to PBBs in the 1970s provides the only report of an outbreak of non-goitrogenic thyroid dysfunction attributable to environmental contamination by chemicals.24 Although sophisticated testing of those with depressed T4 levels was not conducted to establish the level of injury in the thyroid–pituitary–hypothalamic axis, the report of increased antithyroid antibodies serves as indirect evidence of a primary effect on the thyroid gland itself. Studies of adult humans exposed to relatively high levels of PCBs provide conflicting data about the effects of these compounds on thyroid function. In adults who 16 years earlier had consumed cooking oil (‘Yusho rice oil’) contaminated with PCBs, serum T3 and T4 were elevated, but TSH and antithyroid antibodies were not changed compared to controls.25 Serum PCBs did not correlate with T4 or T3 in that study. In contrast, a history of current or former occupational PCB exposure among transformer repairmen was associated with a lower T4 and T4RT3 index compared to unexposed controls.26 In the same study, adipose or serum PCB concentrations were not significantly associated with exposure, suggesting that exposure history may have been spurious or that the relationship between PCB body burden and thyroid function is not a linear one. A recent study of employees in a former PCB plant found increased thyroid volumes and increased prevalence of antithyroid antibodies in the employees compared to controls, but no differences in T4 or TSH levels.27 Unlike the conflicting human data, experimental studies in adult rats demonstrate suppression of serum T4 by PCBs.16 Exposure to another class of chlorinated polyaromatic hydrocarbons, the dioxins, has also been associated with subclinical changes in thyroid hormone secretion in some human studies. There was a positive relationship between serum 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) concentration and T4 and free T4 index, but not TSH or thyroid disease, in industrial workers exposed to TCDD contamination 15 years previously.28 Similarly, another study of chemical workers also found a positive associa-

Disorders of Adrenal Function 613 tion between serum TCDD concentration and T4, but not TSH.29 In contrast, free T4 index did not differ between Operation Ranch Hand Vietnam veterans exposed to TCDD in Agent Orange and controls.28 PCBs, like dioxins and furans, are highly persistent, lipophilic, organochlorine pollutants for which more than 90% of human exposure occurs via food intake, primarily meat, fish, and dairy products (see Chapter 45).30 Breastfeeding can be an important source of exposure, accounting for 12% to 14% of the total PCB and dioxin intake in one study.30 One recent study in children found a negative relationship between serum PCB levels and serum levels of free T3, and a positive correlation with TSH.31 A study of breast-fed Japanese infants found significant negative correlations between total dioxin equivalents contributed by PCBs, dioxins, and furans in the breast milk, and serum T4 and T3 levels.32 A Dutch study found a positive correlation between breast milk PCBs and dioxins and neonatal TSH, but no relationship with T4.33 Another study of background level PCB exposure in the US, in contrast, found no significant correlation between maternal PCB levels and neonatal T4, T3, or TSH levels at birth.34 The significance of small reductions in thyroid hormone levels observed in the former studies on the developing brain are not yet fully characterized. However, studies of human populations with prenatal and perinatal environmental PCB exposures demonstrate neurologic deficits that are consistent with a mechanism involving disruption of thyroid hormone homeostasis. These studies reported inverse relationships between various indices of PCB exposure and motor function, attention, IQ, and memory deficits.4,35 Moreover, animal studies have shown that maternal PCB exposure causes reduced T4 and T3 levels in fetal brains, as well as motor deficits and impaired learning.4,35 Thyroid function has been studied in women exposed to cobalt-based paints.36,37 Cross-sectional blood tests suggest that one effect is dose-related elevation of T4, free T4, and the T4-to-T3 ratio, without a change in TSH levels or pituitary function. The results suggest reduced peripheral deiodination of T4 to T3, but the clinical significance of this observation is unclear. For none of the settings in which human effects have been observed and reported are there adequate longitudinal data on the natural history or response to any intervention. Lead workers removed from further exposure appear to have gradual amelioration of depressed T4 levels. The effects of chelation or thyroid supplementation are unknown. The potential reversibility of lesions due to other exposures described above is also unknown, but at a minimum, removal of the affected individual from any further exposure would seem advisable once non-environmental causes have been excluded. In addition to the effects that have been reported in workers and communities, experimental studies in animals suggest that the thyroid is potentially susceptible to injury from widely used chemicals. These include insecticides such as organophosphates, carbamates and organochlorines, fungicides, food colorants, and mercury.38 It is premature to

recommend that exposed individuals be screened or otherwise evaluated for these effects outside research protocols. Nonetheless, exposure to these agents should be considered as potentially causal in patients who are identified with thyroid functional abnormalities when other established factors have been excluded.

DISORDERS OF ADRENAL FUNCTION Very little is known about the effects of environmental agents on adrenal cortical or medullary function in humans. Occupationally, only workers heavily exposed to lead and PCBs have been studied. Results of studies of lead-exposed workers suggest that secretion of corticosteroids is depressed in response to insulin-induced hypoglycemia and vasopressin; baseline levels of 17-hydroxy (glucocorticoid) and 17-keto (androgenic) steroids were low normal. Because response to administration of adrenocorticotropic hormone (ACTH) was normal, it was presumed that the lesion was at the hypothalamic-pituitary level, but primary toxicity on the adrenal itself could not be excluded. Neither the clinical relevance nor the appropriate treatment for these lesions has been characterized. Importantly, all workers demonstrating these effects also had other evidence of lead toxicity, requiring all the workers to be removed from exposure and chelation therapy to be instituted in those more severely affected. In a study of humans massively exposed to the pesticide gramoxone by ingestion, levels of adrenocortical hormones were measured and found to be markedly elevated in those who later died, but normal or low in survivors.39 Whether this factor represents toxicity to the adrenal gland or axis or whether it is a non-specific response to systemic poisoning could not be ascertained; the effects of lower doses of this and related agents are unknown. Current and former transformer repairmen exposed to PCBs were found to have lower urinary 17-hydroxycortisone excretion than controls.26 Moreover, urinary 17hydroxycortisone was negatively correlated with adipose tissue PCB concentration.26 It is not known whether there was a concurrent effect on serum glucocorticoid concentrations, as they were not measured. Reduced urinary 17hydroxycortisone may reflect a suppressive effect of PCBs on adrenal glucocorticoid secretion or synthesis, resulting in reduced serum levels of glucocorticoids. Animal studies suggest that the adrenal gland may be more sensitive to environmental effects than the limited human data would suggest. A broad range of organochlorine, organophosphate, and carbamate pesticides has been tested, and evidence of histologic alterations in adrenocortical cells has been found; similar effects have been shown with ammonium sulfate fertilizer. The herbicide paraquat has been established to be a potent inhibitor of aldosterone synthesis in the adrenal gland, with action similar to the drugs spironolactone and metyrapone;40 depression

614 Endocrine Disorders of aldosterone secretion has been demonstrated in animals. Several organophosphates, including malathion and diazinon, have been shown to interfere with adrenomedullary function, causing enhanced secretion of epinephrine and norepinephrine, with resulting hyperglycemia, glycogen deposition in the liver, and exhaustion of adrenal stores of these amines.41,42 Mirex, toxaphene, and dioxin (2,3,7,8 TCDD) have been shown to cause direct suppression of glucocorticoid synthesis by the adrenal, with resultant hypoglycemia that could be reversed by cortisone administration.43–45 The solvent 1,1,1-trichloroethane suppresses plasma corticosterone and ACTH concentrations, possibly via an effect on hypothalamic corticotropin-releasing hormone (CRH) in rodents.46 High doses relative to likely human environmental exposures have been used in these studies, so the relevance to humans remains unclear.

DISORDERS OF THE PITUITARY GLAND Lead, styrene, and beryllium have been demonstrated to affect the pituitary gland of occupationally exposed workers. The data on lead have already been summarized earlier in the sections on thyroid and adrenal function. There is further evidence for an effect on the pituitary–gonadal axis (see Chapter 27.2). However, sophisticated testing of these workers failed to demonstrate any effects on growth hormone or prolactin secretion from the anterior pituitary or any evidence of posterior pituitary dysfunction. The effect of styrene on hypothalamic–pituitary function has been studied in several groups of exposed workers. Although thyroid and gonadotropic hormones appeared to be unaffected by the exposure, baseline prolactin and growth hormone levels were elevated in a dose-dependent fashion.47 Significantly, there was a very marked and dose-dependent enhancement of the prolactin response to thyrotropinreleasing hormone (TRH), suggesting the possibility of a defect in the normal counter-regulation of prolactin, which is controlled by hypothalamic dopamine.48 These results suggest the possibility that styrene specifically depletes dopamine from the tuberoinfundibular portion of the hypothalamus. Depletion of dopamine in this portion of the hypothalamus following styrene inhalation has been demonstrated in experimental studies in the rabbit.49 Notably, in occupational studies of pituitary function, workers experienced prominent, albeit vague, CNS complaints, suggesting that endocrine disturbance was not the first or sole toxicity resulting from the exposure. Although the clinical significance remains unclear, levels of prolactin observed in this condition could be a cause of galactorrhea, secondary infertility, or both in the appropriate exposure setting. Despite the hypothetical possibility of reduced fertility caused by hyperprolactinemia, a recent study found no effect of styrene exposure on time-to-pregnancy, an indicator of fertility, in men.50 Patients with chronic beryllium disease due to beryllium exposure may exhibit granulomas in the pituitary gland

similar to those seen in sarcoidosis. However, functional disorders of the anterior and posterior pituitary gland similar to those seen occasionally in sarcoidosis have not been described in beryllium disease. Nonetheless, the occurrence of granulomas in this organ raises the possibility of functional defects that should be considered when they are clinically suggested. In addition to the above-mentioned studies, limited human data suggest that pituitary secretion of the reproductive hormones luteinizing hormone (LH) and folliclestimulating hormone (FSH) may be suppressed by occupational exposure to toluene51,52 and herbicides53 and enhanced by occupational exposure to carbon disulfide.23 Very little experimental work on pituitary responses to environmental agents has been conducted to date. Animal studies have demonstrated effects of the aromatic solvents toluene and xylene on pituitary hormone secretion; however, the results of these studies have been inconsistent. One study reported effects of toluene exposure on prolactin levels, but not on LH, FSH, TSH, or cortisosterone concentrations,54 while another reported effects on LH and FSH levels.55 Xylenes were found to suppress both serum prolactin and corticosterone in rats in the third study.56

DIABETES MELLITUS It is evident from the above-mentioned effects of xenobiotics on the adrenal and pituitary glands that agents that enhance production or release of growth hormone or adrenomedullary amines could induce some degree of glucose intolerance. However, no human case nor animal model of diabetes that has developed via such a pathway has yet been described. One single agent, the rodenticide Vacor (pyridyl N-pnitrophenyl urea), has been implicated as a cause of diabetes mellitus after human ingestion.57 Notably, glucose intolerance was the major clinical toxicity in reported cases, although autonomic and peripheral neuropathy with orthostatic hypotension were concomitantly observed. Animal studies have now documented the very strong binding of Vacor to pancreatic islet beta cells by a mechanism similar to that of the drug streptozocin. Whether this inhibition of insulin release may be reversible is unclear. Studied cases have demonstrated subsequent cytotoxic effects on the islet cells, and chronic insulin-dependent diabetes has resulted. Anti-islet cell antibodies have been measured in serum from some patients. It is unclear whether specific treatment early in toxicity, such as with the competitive binder streptozocin, may lead to a better outcome than supportive care alone. Positive relationships have been observed between serum TCDD levels and both fasting serum glucose levels and risk of diabetes in workers exposed to chemicals contaminated with TCDD.58 An association between TCDD levels and risk of diabetes mellitus has also been observed among US veterans of Operation Ranch Hand, who were exposed to TCDD in Agent Orange.59 Interestingly, even among nondiabetic Ranch Hand veterans, there was a positive relationship between serum insulin and TCDD exposure category,

Disorders of Calcium and Bone Metabolism 615 suggesting hyperinsulinemia as a possible mechanism for the development of diabetes in dioxin-exposed individuals.59 Increased rates of diabetes-related mortality have been reported in female TCDD-exposed individuals in one study,60 but not in two other studies.61,62 A large international study that combined 36 dioxin-exposed cohorts from 12 countries found an increased risk of mortalit from diabetes, but did not reach statistical significance.63 However, because most people with diabetes mellitus now survive for years after diagnosis, one would not necessarily expect to see an association with diabetes-related mortality even if there were a true association between dioxin exposure and risk of diabetes.

DISORDERS OF CALCIUM AND BONE METABOLISM Calcium homeostasis and bone metabolism are under the control of the parathyroid hormone and the hepatic and renal metabolites of vitamin D. As well, the gastrointestinal tract, bone, and kidney are important target organs that determine the systemic impact of these hormones. Virtually no data are available regarding the effect of toxins on the function of the parathyroid gland; one study of lead-exposed workers showed no effect. Although a variety of chemicals and ionizing radiation are capable of inducing parathyroid adenomas under appropriate experimental conditions, the relevance to human exposure is questionable. Hypoparathyroid states due to toxins have not been described in either humans or animals. Vitamin D metabolism has been investigated in cases involving childhood lead poisoning because of the clinical observation of poor growth and abnormal bone structure in affected children.64 The best evidence currently suggests that lead directly interferes with renal metabolism of 25hydroxycholecalciferol to the more active 1,25-dihydroxy derivative. Whether supplementation with this metabolite could reverse some of the adverse sequelae of lead poisoning in this population has not been established. Beryllium disease has been associated with hypercalcemia in a small number of patients. Although reports of vitamin D metabolism in this specific setting have not yet been published, it is likely that the hypercalcemia is related to that seen in other granulomatous disorders – namely, increased conversion of 25-hydroxycholecalciferol to the active moiety by hydroxylase in the granulomas, leading effectively to hypervitaminosis D. The administration of corticosteroids generally is satisfactory treatment for this metabolic complication. Another environmental and occupational disease state associated with abnormalities of calcium metabolism in humans is chronic cadmium intoxication. Osteopenia is frequently seen in advanced cases, and it is often the first clinical manifestation (i.e., pathologic fractures) which is responsible for the Japanese name itai-itai (literally meaning ‘ouch-ouch’) for this condition. It is presumed that the bone disease occurs as a consequence of the renal tubular defect induced by prolonged exposure to cadmium, with subsequent metabolic acidosis, calcium loss, and bony

demineralization.65 Recently, lower levels of cadmium exposure have been associated with reduced bone mineral density and osteoporosis in manufacturers of heat exchangers and in battery workers.66,67 Recently, serum levels of p,p′-DDE, a metabolite of the organochlorine insecticide DDT, have been found to be negative predictors of bone mineral density in postmenopausal women.68 The authors speculated that p,p′DDE may affect bone density via its ability to potently antagonize androgen receptor binding. Beyond these reports, there are few experimental data by which to predict which other classes of toxins may have effects on vitamin D metabolism short of widespread systemic toxicity.

References 1. Colborn T, vom Saal FS, Soto AM. Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ Health Perspect 1993; 101:378-84. 2. Safe DH. Environmental and dietary estrogens and human health: is there a problem? Environ Health Perspect 1994; 103:346. 3. Toppari J, Larsen JC, Christiansen P, et al. Male reproductive health and environmental xenoestrogens. Environ Health Perspect 1996; 104(Suppl 4):741-803. 4. Sher ES, Xu XM, Adams PM, et al. The effects of thyroid hormone level and action in developing brain: are there targets for the actions of polychlorinated biphenyls and dioxins? Toxicol Ind Health 1998; 14:121-58. 5. Guillette LJJ, Crain DA, Rooney AA, et al. Organization versus activation: the role of endocrine-disrupting contaminants (EDCs) during embryonic development in wildlife. Environ Health Perspect 1995; 103(Suppl 7):157-64. 6. Banerjee KK, Bishayee A, Marimuthu P. Evaluation of cyanide exposure and its effects on thyroid function of workers in a cable industry. J Occup Environ Med 1997; 39:258-60. 7. Brechner RJ, Parkhurst GD, Humble WO, et al. Ammonium perchlorate contamination of Colorado river drinking water is associated with abnormal thyroid function in newborns in Arizona. J Occup Environ Med 2000; 42:777-82. 8. Lamm SH, Doemland M. Has perchlorate in drinking water increased the rate of congenital hypothyroidism? J Occup Environ Med 1999; 41:409-11. 9. Gibbs JP, Ahmad R, Crump KS, et al. Evaluation of a population with occupational exposure to air borne ammonium perchlorate for possible acute or chronic effects on thyroid function. J Occup Environ Med 1998; 40:1072-82. 10. Lamm SH, Braverman LE, Li FX, et al. Thyroid health status of ammonium perchlorate workers: a cross-sectional occupational health study. J Occup Environ Med 1999; 41:248-60. 11. Lawrence JE, Lamm SH, Pino K, Richman K, Braverman LE. The effect of short-term low-dose perchlorate on various aspects of thyroid function. Thyroid 2000; 10:659-63. 12. Lawrence JE, Lamm SH, Braverman LE. Low dose perchlorate (3 mg daily) and thyroid function. Thyroid 2001; 11:295. 13. Greer MA, Goodman G, Pleus RC, Greer SE. Health effects assessment for environmental perchlorate contamination: the dose-response for inhibition of thyroidal radioiodine uptake in humans. Environ Health Perspect 2002; 110:927-37. 14. US Environmental Protection Agency. Perchlorate environmental contamination: toxicologic review and risk characterization based on emerging information. NCEA-10503, January 16, 2002. Washington, DC: US Environmental Protection Agency.

616 Endocrine Disorders 15. Van Middlesworth L. Effects of radiation on the thyroid gland. Adv Intern Med 1989; 34:265-84. 16. Cheek AO, Kow K, Chen J, et al. Potential mechanisms of action of thyroid disruption in humans: interaction of organochlorine compounds with thyroid receptor, transthyretin, and thyroid-binding globulin. Environ Health Perspect 1999; 107:273-8. 17. Zhou L-X, Dehal SS, Kupfer D, et al. Cytochrome P450 catalyzed binding of methoxychlor to rat hepatic microsomal iodothyronine 5‘-monodeiodinase, type 1: does exposure to methoxychlor disrupt thyroid hormone metabolism? Arch Biochem Biophys 1995; 322:390-4. 18. Vulsma T. Impact of exposure to maternal PCBs and dioxins on the neonate’s thyroid hormone status. Epidemiology 2000; 11:239-41. 19. Cullen MR, Robins JM, Kayne RD. Endocrine and reproductive dysfunction in men associated with occupational inorganic lead intoxication. Arch Environ Health 1984; 39:431-40. 20. Robins JM, Cullen MR, Connors BB, et al. Depressed thyroid indices associated with occupational exposure to lead. Arch Intern Med 1983; 143:220-4. 21. Schumacher C, Brodkin CA, Alexander B, et al. Thyroid function in lead smelter workers: absence of subacute or cumulative effects with moderate lead burdens. Int Arch Occup Environ Health 1998; 71:453-8. 22. Siegel M, Forsyth B, Siegel L, et al. The effect of lead on thyroid function in children. Environ Res 1989; 49:190-6. 23. Wagar G, Tolonen M, Stenman UH, et al. Endocrinologic studies in men exposed occupationally to carbon disulfide. J Toxicol Environ Health 1981; 7:363-71. 24. Bahn AK, Mills JL, Snyder PJ, et al. Hypothyroidism in workers exposed to polybrominated biphenyls. N Engl J Med 1980; 302:31-3. 25. Murai K, Okamura K, Tsuji H, et al. Thyroid function in “yusho” patients exposed to polychlorinated biphenyls (PCB). Environ Res 1987; 44:179-87. 26. Emmett EA, Maroni M, Jefferys J, et al. Studies of transformer repair workers exposed to PCBs: II. Results of clinical laboratory investigations. Am J Ind Med 1988; 14:47-62. 27. Langer P, Tajtakova M, Fodor G, et al. Increased thyroid volume and prevalence of thyroid disorders in an area heavily polluted by polychlorinated biphenyls. Eur J Endocrinol 1998; 139:402-9. 28. Calvert GM, Sweeney MH, Deddens J, et al. Evaluation of diabetes mellitus, serum glucose, and thyroid function among United States workers exposed to 2,3,7,8-tetrachlorodibenzop-dioxin. Occup Environ Med 1999; 56:270-6. 29. Suskind RR, Hertzberg VS. Human health effects of 2,4,5-T and its toxic contaminants. JAMA 1984; 251:2372-80. 30. Patandin S, Dagnelie PC, Mulder PGH, et al. Dietary exposure to polychlorinated biphenyls and dioxins from infancy until adulthood: a comparison between breast-feeding, toddler, and long-term exposure. Environ Health Perspect 1999; 107:45-51. 31. Osius N, Karmaus W, Kruse H, et al. Exposure to polychlorinated biphenyls and levels of thyroid hormones in children. Environ Health Perspect 1999; 107:843-9. 32. Nagayama J, Okamura K, Iida T, et al. Postnatal exposure to chlorinated dioxins and related chemicals on thyroid hormone status in Japanese breast-fed infants. Chemosphere 1998; 37:1789-93. 33. Koopman-Esseboom C, Weisglas-Kuperus N, de Ridder MA, et al. Effects of polychlorinated biphenyl/dioxin exposure and feeding type on infants’ mental and psychomotor development. Pediatrics 1996; 97:700-6. 34. Longnecker MP, Gladen BC, Patterson DG, et al. Polychlorinated biphenyl (PCB) exposure in relation to thyroid hormone levels in neonates. Epidemiology 2000; 11:249-54. 35. Porterfield SP, Hendry LB. Impact of PCBs on thyroid hormone directed brain development. Toxicol Ind Health 1998; 14:103-20.

36. Prescott E, Netterstrom B, Faber J, et al. Effect of occupational exposure to cobalt blue dyes on the thyroid volume and function of female plate painters. Scand J Work Environ Health 1992; 18:101-4. 37. Christensen JM, Poulsen OM. A 1982-1992 surveillance programme on Danish pottery painters. Biologic levels and health effects following exposure to soluble or insoluble cobalt compounds in cobalt blue dyes. Sci Total Environ 1994; 150:95-104. 38. Capen CC. Mechanisms of chemical injury to the thyroid gland. Prog Clin Biol Res 1994; 387:173-91. 39. Barsony J, Kertesz F. Investigation of adrenal steroids and 25-hydroxy-cholecalcipherol in human gramoxone poisoning. Arch Toxicol Suppl 1985; 8:280-3. 40. Blanchouin-Emeric N, Defaye G, Toury R, et al. The reoxidation of cytochrome P-450 by paraquat inhibits aldosterone biosynthesis from 18-hydroxycorticosterone. J Steroid Biochem 1988; 31:331-5. 41. Matin MA, Husain K, Khan SN. Modification of diazinon-induced changes in carbohydrate metabolism by adrenalectomy in rats. Biochem Pharmacol 1990; 39:1781-6. 42. Liu PS, Kao LS, Lin MK. Organophosphates inhibit catecholamine secretion and calcium influx in bovine adrenal chromaffin cells. Toxicology 1994; 90:81-91. 43. Bestervelt LL, Cai Y, Piper DW, et al. TCDD alters pituitaryadrenal function. I: adrenal responsiveness to exogenous ACTH. Neurotoxicol Teratol 1993; 15:365-7. 44. Jovanovich L, Levin S, Khan MA. Significance of mirex-caused hypoglycemia and hyperlipidemia in rats. J Biochem Toxicol 1987; 2:203-13. 45. Mohammed A, Hallberg E, Rydstrom J, et al. Toxaphene: accumulation in the adrenal cortex and effect on ACTHstimulated corticosteroid synthesis in the rat. Toxicol Lett 1985; 24:137-43. 46. Pise VM, Reigle TG, Muralidhara S, et al. Effects of acute inhalation exposure to 1,1,1-trichloroethane on the hypothalamo-pituitary-adrenal axis in male Sprague-Dawley rats. J Toxicol Environ Health 1998; 54(Part A):193-208. 47. Mutti A, Vescovi PP, Falzoi M, et al. Neuroendocrine effects of styrene on occupationally exposed workers. Scand J Work Environ Health 1984; 10:225-8. 48. Arfini G, Mutti A, Vescovi P. Impaired dopaminergic modulation of pituitary secretion in workers occupationally exposed to styrene: further evidence from PRL response to TRH stimulation. J Occup Med 1987; 129:826-30. 49. Mutti A, Falzoi M, Romanelli A, et al. Regional alterations of brain catecholamines by styrene exposure in rabbits. Arch Toxicol 1984; 55:173-7. 50. Kolstad HA, Bisanti L, Roeleveld N, et al. Time to pregnancy for men occupationally exposed to styrene in several European reinforced plastics companies. Scand J Work Environ Health 1999; 25(Suppl 1):66-9. 51. Svensson B-G, Nise G, Erfurth EM, et al. Neuroendocrine effects in printing workers exposed to toluene. Br J Ind Med 1992; 49:402-8. 52. Svensson B-G, Nise G, Erfurth EM, et al. Hormone status in occupational toluene exposure. Am J Ind Med 1992; 22:99-107. 53. Garry VF, Burroughs B, Tarone R, et al. Herbicides and adjuvants: an evolving view. Toxicol Ind Health 1999; 15:159-67. 54. von Euler G, Fuxe K, Hansson T, et al. Effects of chronic toluene exposure on central monoamine and peptide receptors and their interactions in the adult male rat. Toxicology 1988; 52:103-26. 55. Andersson K, Fuxe K, Toftgard R, et al. Toluene-induced activation of certain hypothalamic and median eminence catecholamine nerve terminal systems of the male rat and its effects on anterior pituitary hormone secretion. Toxicol Lett 1980; 5:393-8. 56. Andersson K, Fuxe K, Nilsen OD, et al. Production of discrete changes in dopamine and noradrenaline levels and turnover

Disorders of Calcium and Bone Metabolism 617

57. 58.

59.

60.

61.

62.

63.

64.

in various parts of the rat brain following exposure to xylene, ortho-, meta- and para-xylene and ethylbenzene. Toxicol Appl Pharmacol 1981; 60:535-48. Karam JH, Lewitt PA, Young CW, et al. Insulinopenic diabetes after rodenticide (Vacor) ingestion. Diabetes 1980; 29:971-8. Sweeney MH, Calvert GM, Egeland GA, et al. Review and update of the results of the NIOSH medical study of workers exposed to chemicals contaminated with 2,3,7,8tetrachlorodibenzodioxin. Teratog Carcinog Mutagen 1997-98; 17:241-7. Michalek JE, Akhtar FZ, Kiel JL. Serum dioxin, insulin, fasting glucose, and sex hormone-binding globulin in veterans of Operation Ranch Hand. J Clin Endocrinol Metab 1999; 84:1540-3. Pesatori AC, Zocchetti C, Guercilena S, et al. Dioxin exposure and non-malignant health effects: a mortality study. Occup Environ Med 1998; 55:126-31. Ott MG, Olson RA, Cook RR. Cohort mortality study of chemical workers with potential exposure to the higher chlorinated dioxins. J Occup Med 1987; 29:422-9. Steenland K, Piacitelli L, Deddens J, et al. Cancer, heart disease, and diabetes in workers exposed to 2,3,7,8tetrachlorodibenzo-p-dioxin. J Natl Cancer Inst 1999; 91:779-86. Vena J, Boffetta P, Becher H, et al. Exposure to dioxin and non-neoplastic mortality in the expanded IARC International Cohort Study of Phenoxy Herbicide and Chlorophenol Production Workers and Sprayers. Environ Health Perspect 1998; 106(Suppl 2):645-53. Osterloh JD. Observations on the effect of parathyroid hormone on environmental blood lead concentrations in humans. Environ Res 1991; 54:8-16.

65. Nogawa K, Tsuritwni I, Kido T, Honda R, Yamada Y, Ishizaki M. Mechanism for bone disease found in inhabitants environmentally exposed to cadmium: decreased serum 1-alpha 25 dihydroxy vitamin D levels. Int Arch Occup Environ Health 1987; 59:21-30. 66. Jarup L, Alfven T, Persson B, et al. Cadmium may be a risk factor for osteoporosis. Occup Environ Med 1998; 55:435-9. 67. Alfven T, Elinder CG, Carlsson MD, et al. Low-level cadmium exposure and osteoporosis. J Bone Miner Res 2000; 15:1579-86. 68. Beard J, Marshall S, Jong K, et al. 1,1,1-trichloro-2,2-bis(pchlorophenyl)-ethane (DDT) and reduced bone mineral density. Arch Environ Health 2000; 55:177-80. 69. Davis BJ, Heindel JJ. Ovarian toxicants: multiple mechanisms of action. In: Korach KS, ed. Reproductive and developmental toxicology. New York: Marcel Dekker; 1998:373. 70. Gray LE Jr, Ostby J, Ferrell J, et al. A dose-response analysis of methoxychlor-induced alterations of reproductive development and function in the rat. Fundam Appl Toxicol 1989; 12:92-108. 71. Kelce WR, Stone CR, Laws SC, et al. Persistent DDT metabolite p,p‘-DDE is a potent androgen receptor antagonist. Nature 1995; 375:581-5. 72. Haake J, Kelley M, Keys B, et al. The effects of organochlorine pesticides as inducers of benz[a]pyrene hydroxylases. Gen Pharmacol 1987; 18:165-9. 73. Han YG, Kang SS, Seong JY, et al. Negative regulation of gonadotropin-releasing hormone and gonadotropin-releasing hormone receptor gene expression by a gonadotrophinreleasing hormone agonist in the rat hypothalamus. J Neuroendocrinol 1999; 11:195-201.

27.2 Disorders of Reproduction and Development Ulrike Luderer, Mark R Cullen, Donald R Mattison Extraordinary awareness, interest, and concern have arisen in the past three decades regarding the effects of physical activity and occupational and environmental exposures on human reproductive and developmental health. These factors have paralleled increased interest in reproductive health, motivated in part by an explosion of technology to enhance fertility and minimize adverse pregnancy outcomes. At the same time, the focus on reproduction and development has followed increased concerns about the effects of chemical, biologic, and physical hazards on health generally. Certain unique social, scientific, and medical circumstances have contributed further to these concerns. Most obvious has been the enormous shift of women into the workforces of developed countries. Most women now work before, during, and after their pregnancies. Furthermore, the range of occupations undertaken by women has expanded to include virtually all of those formerly limited to men. Not surprisingly, issues such as the safety of occupational exposure standards for women of reproductive age, pregnant women and their fetuses, and the appropriate modifications of work practices and the work environment for the pregnant woman have challenged traditional approaches to occupational safety, which were developed in an earlier era. More recently, the concern has been raised that environmental pollutants, at levels to which the general public is exposed, may be having adverse effects on human reproduction and development. Exposure to chemical pollutants that disrupt endocrine function by mechanisms such as binding to intracellular steroid receptors has been associated with reproductive failure and abnormal sexual differentiation in wildlife and laboratory species.1 Such observations led to hypotheses that similar abnormalities in humans, including declining sperm counts and increased incidence of abnormal urogenital development, breast cancer, and testicular cancer, may be caused by environmental exposures to endocrinedisrupting chemicals.1-4 These hypotheses have generated considerable controversy,5,6 and are currently under active investigation. Relevant research will be discussed in later sections of this chapter. Unfortunately, these challenges uncovered a second aspect of the reproductive-environmental health problem, namely, the extraordinary paucity of useful information. Even today, information is available on reproductive and developmental effects for a relatively small number of physical, chemical, and biologic agents, and primarily for those effects that occur at high levels of exposure. Human epidemiologic studies are difficult to perform because of the sensitive nature of the information required, and the problems of classifying individuals based on their exposures during the often brief periods of biologic interest. Toxicologic studies are no less difficult to interpret given the differences between the reproductive biology of humans and experimental animals. The net effect is that even though the issue of reproductive and developmental health in the workplace has emerged as serious and promi-

nent, the availability of scientific and clinical data on which to make decisions remains limited. In response to these realities, many employers, physicians, and regulators proposed that women who were pregnant or were capable of becoming pregnant should be removed from exposure to any hazard that could possibly affect fecundity, fertility, or the outcome of pregnancy. However, this approach was based on the mistaken premise that environmental agents would have their impacts only on fetuses in recognized pregnancies following exposure of the female. In fact, successful reproduction and development entails events in men and women that long precede recognized pregnancy as well as critical events in the fetus, placenta, and infant (Fig. 27.2.1). Further, many harmful agents are known to have long half-lives in the body, whereas others, at least theoretically, could cause effects on target tissues that persist long after the agent itself has disappeared. Therefore, protection of reproductive and developmental health could not scientifically be viewed as a matter of limiting exposure to pregnant women alone, but must also include the infant after birth and men. Even if such policies could scientifically limit their focus to women, there remain the social issues raised by a health policy that overtly discriminates against the job rights of fecund or pregnant women, a view underscored by the United States Supreme Court in its 1991 decision of United Auto Workers versus Johnson Controls, Inc., in which the majority found that federal law prohibits job discrimination purely on the basis of pregnancy or the ability to become pregnant. This chapter (1) defines adverse reproductive and developmental health outcomes, (2) summarizes non-environmental risk factors, (3) reviews sources of agent-specific data, (4) provides listings of agents for which reasonably good data exist, (5) outlines clinical approaches to common reproductive and developmental health concerns, and (6) discusses workforce or population-based prevention strategies.

ADVERSE REPRODUCTIVE AND DEVELOPMENTAL HEALTH OUTCOMES Because of the complexity of reproduction and development, there are multiple adverse reproductive outcomes in humans (Fig. 27.2.1). Although knowledge of the biologic bases of these events remains limited, some information exists regarding their background distribution in populations of interest (Table 27.2.1) and the factors that predispose to their occurrence.

Infertility and subfertility Because the concept of fertility often is discussed imprecisely, some definitions are presented here. Fecundity is the capability of the male, female, or couple to produce

Adverse Reproductive and Developmental Health Outcomes 619 Processes of normal reproduction and development

Adverse outcomes

Techniques to evaluate

- Impaired fecundity

- Hormone concentrations and patterns of release - Semen characteristics - Ovulatory frequency

- Impaired fertility

- Time to pregnancy - Proportion pregnant after fixed interval

Implantation and preclinical gestation

- Subclinical spontaneous abortion

- Serum or urinary hCG

Clinical pregnancy and fetal development

- Spontaneous abortion - Fetal death - Fetal growth retardation

- Fetal growth and development

Birth

- Prematurity - Congenital malformation/ chromosomal anomalities

- Gestational length - Infant structure and function

Postnatal development

- Developmental disorders - Childhood cancer

- Infant and childhood growth and development

Male fecundity

Female fecundity

Conception

Reproductive or developmental outcome

Percent

Infertility (of couples) Preimplantation pregnancy loss Spontaneous abortion (including clinically unrecognized) Clinically recognized spontaneous abortion Stillbirth (beyond 28 weeks) Premature, postmature, growth retardation Birth defects identified at birth identified over first year of life

10–15 20–30 15–50 10–15 2.5 Gy), few studies have been conducted to explore the effect.47 The reproductive effects of many members of the large class of organochlorine chemicals have been extensively studied. PCBs and estrogenic organochlorine compounds appear capable of disrupting female fecundity based on endocrine effects on the timing of HPO axis hormonal function, but infertility, per se, has not been documented among women who have been exposed. Taiwanese and Japanese women who consumed cooking oil contaminated with PCBs and furans have more menstrual abnormalities, but not higher rates of infertility, than control women.73 A recent large retrospective study of women exposed to PCBs, heavy metals, and other contaminants by consuming Great Lakes fish, found no effect on fecundity.74 Limited data have linked exposure to dioxins in monkeys and in a nude-mouse model to endometriosis, a disease frequently associated with infertility in women.75,76 As in males, animal studies have demonstrated effects of in-utero exposures to toxicants on reproductive system function in adulthood in females as well. Gestational exposure to the insecticide methoxychlor67 and to dioxin77 are the most well-studied examples.

632 Disorders of Reproduction and Development

Spontaneous abortions Female exposures Despite the methodologic problems in studying spontaneous abortions (SAB), a variety of hazards and occupations have been associated with an increased risk for this complication in human populations. Although there remain uncertainties, the associations that are presently accepted as most likely to cause SAB are summarized in Table 27.2.10. Healthcare workers have been extensively studied, and increase SAB risk has been associated with exposure to antineoplastic agents,78 anesthetic gases, and ethylene oxide. Although the preponderance of evidence supports these associations, not all studies have yielded consistent results. One interpretation is that only high levels of exposure enhance the risk for SAB, with lesser exposures showing no effects. Ethylene oxide studies support this interpretation. Women who were engaged in sterilizing instruments using ethylene oxide during pregnancy had elevated risks of SAB in two studies,79,80 whereas women who handled instruments after sterilization did not.81 Similarly, exposure to anesthetic gases was significantly associated with SAB in older studies performed prior to use of scavenging equipment of waste gas, but not in more recent studies performed after the introduction of scavenging equipment. A recent study directly compared SAB risks in dental assistants exposed to the anesthetic gas nitrous oxide in the presence or absence of scavenging systems. Women in the unscavenged group had elevated risk, whereas the risk for the women in the scavenged group was not distinguishable from the unexposed women.82 It is also possible that other exposures of these women, such as the physical and other stresses of their jobs, also may account for some differences among studies. The data for metals are surprisingly limited given their ubiquitous nature in the workplace and environment. Many older descriptive studies documented that women exposed to lead at work experienced SABs frequently,

Hazard Medical hazards Antineoplastic drugs Anesthetic gases Nitrous oxide Ethylene oxide Metals Lead Solvents Ethylene glycol ethers Aromatic solvents Mixed organic solvents Other chemicals PCBs Pesticides Physical agents Heavy labor Shift work

Setting of established risk Oncology nurses Operating room personnel Dental assistants Sterilizer operators Occupational uses Semiconductor manufacture Occupational use Occupational use Heavy food contamination Accidental poisoning

Table 27.2.10 Hazards associated with increased risk for spontaneous abortion or fetal death after female exposure

but classification of exposures was limited and confounded by other lead effects that were difficult to evaluate; studies of environmentally exposed women have not been impressive regarding the risk for spontaneous abortion. The findings regarding mercury, a reproductive toxicant in animals, have not been consistent. The study described above, that identified waste nitrous oxide as a cause of increased spontaneous abortion, did not find an elevated risk for mercury exposure in dental assistants.82 Extensive epidemiologic literature exists on solvents, with the weight of the evidence suggesting that occupational solvent exposure in general is associated with increased risk of spontaneous abortion.83,84 Identifying which specific solvents or classes of solvents are the culprits has proven more difficult, with the possible exception of the ethylene glycol ethers, which are potent reproductive toxicants in animals, and the aromatic solvents toluene and benzene. Multiple studies of the semiconductor industry have all found excesses of spontaneous abortions among women exposed to ethylene glycol ethers in computer chip production.85-88 However, these data must be interpreted cautiously because the solvent exposures in this setting are generally low and the environment is replete with other physical and chemical hazards that may confound the association of up to a doubled risk for first trimester spontaneous abortion. Studies of laboratory workers,89 electronics workers,90 and petrochemical workers91 have implicated aromatic solvents as potential causes of SAB. The solvent perchloroethylene has also been associated with increased risk of SAB in a study of dry-cleaning workers.92 The associations between spontaneous fetal loss and both PCBs and pesticides are based on studies of women who have been very heavily exposed. Increased spontaneous abortion was one of the outcomes noted among women poisoned by PCBs in the infamous rice-oil contamination episodes in Japan and Taiwan (see Chapters 44 and 45). Fetal loss after pesticide poisoning has been widely described in case reports, but usually after the mother herself is poisoned. There are few human studies that incriminate either occupational or ambient environmental levels of exposures in humans for this endpoint, although many widely used compounds are suspected based on animal effects. A recent study in Turkey found that the rate of self-reported SAB increased with increasing serum levels of the fungicide hexachlorobenzene in women who had been intoxicated with the pesticide during the 1950s and in control women.93 A relationship between occupational exposure to pesticides and SAB is also suggested by a recent case-control study which found that women who worked in agriculture or lived on farms were at increased risk for infertility.94 The data on physical activity are extensive and controversial. The evidence is strongest for shiftwork (night shifts or rotating shifts), which was associated with elevated risk of SAB in six of nine studies.95 Some, but not all, studies have shown that heavy lifting, standing 8 or more hours per day, and physically heavy work increased the risk for spontaneous abortion.96 In contrast, many studies of leisure time exercise did not find an increase in SAB rate,

Classification of Human Disorders of Reproduction and Development 633 and some even found protective effects.96 This apparent contradiction may be explained by various factors, including the much longer duration of occupational exposures compared to leisure exposures, and differences in the fitness levels and socioeconomic status of the women in the two kinds of studies. Few data exist on exposure to noise, vibration, heat, and cold and rates of SAB. Several widespread physical factors are less likely to be important causes of spontaneous abortion at typical exposures found in occupational or environmental settings, including background ionizing radiation, non-ionizing radiation, and the use of video display terminals. While high doses of radiation during the first 8 weeks of pregnancy, as experienced by victims of the atomic bomb blast in Japan, clearly cause SAB, lower occupational and environmental doses do not.97 Non-ionizing, electromagnetic radiation from occupational exposure to video display terminals98,99 and magnetic resonance imaging machines,7 and residence in houses with high current configurations (wire codes)100 have not been found to increase risk of SAB. A few exposures, such as use of electric blankets during the first trimester of pregnancy and occupational exposure to microwave radiation, have been associated with increased SAB risk,100,101 and reproductive effects of non-ionizing radiation remain under intensive investigation. Psychological stress has received increasing attention as a cause of adverse pregnancy outcomes. In the only study to look at the effect of stress on spontaneous abortion rates, major negative life events were associated with increased risk of chromosomally normal SABs compared to chromosomally abnormal SABs.102

Male exposures As noted earlier, there are some theoretical ways in which male exposure could enhance risk for spontaneous abortion. It is important to note that animal studies have demonstrated these effects, predominantly for those chemicals that produce mutations or other types of sperm and somatic cell DNA damage. Several studies have noted increased spontaneous abortions in spouses of men exposed to various suspect agents, such as lead, solvents,103 and mercury. However, none has been confirmed or further elucidated mechanistically at this time. A recent large study of male welders did not confirm previous findings that this common paternal exposure causes SAB in wives of exposed men.104

Fetal growth retardation and prematurity Female exposures Given the rising awareness of the associations between these developmental endpoints and infant and childhood morbidity and mortality, there is great interest in identifying preventable environmental causes. Unfortunately, few have been found to date. Lead is probably the best appreciated preventable exposure. Women exposed occupationally during gestation have

had documented smaller and often premature offspring.105 In addition, the same exposures are associated with various birth defects and childhood developmental abnormalities consistent with direct fetal exposure in utero and associated tissue disruption. Lead exposure in utero has also been associated with reduced postnatal growth rates.106 Children born to women exposed to PCBs during foodborne outbreaks also had smaller offspring, although the infants were not obviously premature. Other stigmata of exposure in these children included skin discoloration and other evidence of ectodermal dysplasia. Small dose-related effects on birth weight also have been documented in areas of heavier environmental exposure via contaminated fish.107 Another well-defined cause of fetal growth retardation is external (gamma or x-ray) ionizing radiation, which also is a cause of birth defects. The fetus is especially sensitive in the second part of the first trimester and early second trimester. Although the dose–response for this effect is not well established, it is a clear risk after accidental exposures in the range of 0.05–0.5 Gy (5–50 rad). What is less clear is the risk at levels just below this limit, which are most likely to be found in the workplace. Since 5 rads is the allowable dose per year for non-pregnant workers in many places, exposures in this range are not rare. Radionuclide exposures that might occur in the workplace or environmentally have not generally shown potential for causing growth retardation. Only radioactive iodide has been well established as a cause of fetal injury and then only after large (i.e., therapeutic) exposure doses. Exposures among healthcare workers have recently been associated with low birth weight and prematurity. Midwives with nitrous oxide exposure during pregnancy were found to have higher rates of low birth weight.69 Dental assistants exposed to the sterilant ethylene oxide had non-significantly elevated risk of preterm birth.80 Knowledge of the effects of alcohol on fetal growth has raised many concerns about solvent exposures, especially those occurring occupationally. Although these are biologically realistic concerns, data gathered to date suggest only an increased risk for maternal pre-eclampsia by an unknown mechanism.108 In mothers who develop this complication, growth retardation and prematurity are probably more likely. Similarly, knowledge of the risks to fetal growth of cigarette smoking in pregnancy have raised concerns about exposures to carbon monoxide and other asphyxiants. However, evidence of fetal injury, including growth retardation, has been shown only after episodes of maternal intoxication. On the other hand, the very much lower levels of ambient oxygen associated with high altitude do result in babies of smaller weight, which is of unclear consequence. Further, fetal toxicity has been documented anecdotally in women who experience pressure changes associated with deep sea diving, and increased fetal mortality was demonstrated in a sheep model for maternal-fetal bends.109 As with spontaneous abortions, there is an extensive amount of literature evaluating physical activity late in

634 Disorders of Reproduction and Development pregnancy and the risk for delivery of small or premature babies. Taken together, the weight of the evidence does not support a large effect of physical activity on birth weight.110-113 While physical activity per se does not appear to increase risk of prematurity,112,113 high energy expenditure combined with high work speed,111 prolonged standing, and long working hours110 were more consistently associated with preterm birth.

Male exposures There are no established factors that relate these outcomes to male exposures. There are suggestions that the spouses of male workers exposed to lead may have smaller offspring than others, but the possibility that these women were also exposed, either at the worksite or by indirect exposure via contaminated clothing, is difficult to exclude.

Birth defects Female exposures Possibly the most dreaded concern about environmental exposure is the likelihood of risk for birth defects. Thus far, only a few hazards have been associated definitively with birth defects. Known associations are summarized in Table 27.2.11. Most of the well-described human environmental developmental toxicants have been recognized after an environmental disaster or epidemic, with presumably high levels of exposure to the causal agent, often with associated maternal morbidity as well. Although in each case there is incontrovertible evidence of effects on humans at high doses and a parallel animal model, the dose–response relationship at low exposure levels remains uncertain, which may be important in counseling pregnant women who seek advice after an exposure has occurred. Historic reports and isolated descriptions of babies born to women Hazard Metals Lead Methyl mercury Solvents Toluene Ethanol Glycol ethers Other chemicals PCBs

Identified defects Various patterns Psychomotor retardation (seizures, paresis, mental retardation)

Reported setting Occupational, environmental Food-borne contamination

Microcephaly, craniofacial anomalies Microcephaly, craniofacial anomalies Various patterns

Glue sniffing

Ectodermal dysplasia (dermal pigmentation, dystrophic nails, mucosal dysplasia, mental retardation)

Food-borne contamination

Physical hazards Ionizing radiation Microcephaly Heat Neural tube defects

Abuse Occupational

Atomic blast

Table 27.2.11 Hazards that cause developmental toxicity after maternal exposure

occupationally exposed to lead document frequent occurrences of various defects including CNS disorders and urorectal malformations similar to patterns seen in some experimental models of lead-induced teratogenesis.114 The effects at doses that would be expected to occur in working women under current lead standards are less certain. At least one large study found an overall excess of minor anomalies of various kinds, whereas other studies have not identified such excesses. Methyl mercury developmental toxicity was first described during the 1950s after residents around Minamata Bay in Japan consumed fish contaminated with methyl mercury that formed from industrial discharge of mercury into the water. Affected offspring had microcephaly, limb deformities, cerebral palsy, mental retardation, and other defects.115 Similar outcomes have been seen following other mass poisoning episodes. During the past decade, concern has emerged that levels of methyl mercury found in large, predatory fish such as tuna may pose a developmental hazard in populations who regularly consume large quantities of fish. These studies have focused mainly on neurodevelopmental deficits that are not apparent at birth, and are discussed in the subsequent section. Similar to the situation with methyl mercury, the risk of birth defects in humans following PCB exposure was first appreciated after two mass poisoning episodes in which cooking oil was contaminated with PCBs.116 Also similar to the situation with methyl mercury, much attention has focused on possible developmental effects of lower levels of PCB exposure due to consumption of contaminated fish. These studies are also discussed below. The well-known adverse developmental effects of ethanol and toluene abuse117 have generated concern that occupational exposure to these and other solvents may also increase risk for birth defects. Thus far, occupational exposures to either of these solvents, which tend to be orders of magnitude lower than with abuse, have not been associated with birth defects. Although many other solvents are developmental toxicants in animals, few human studies have demonstrated increased rates of malformations with occupational solvent exposure. This is likely due both to the rarity of major malformations and to the relatively lower doses to which humans are exposed. Recently, a large multicenter European group reported increased risk for congenital malformations among women exposed to glycol ether solvents during the first trimester of pregnancy.118 A significantly elevated odds ratio for major malformations and solvent exposure was reported in a meta-analysis of studies on solvent exposure.83 Over the past decade, much attention has been paid to the effects of low-level environmental exposures to agents that act by disrupting endocrine function on in-utero development. Compounds that act as antiandrogens or estrogen agonists have been shown to cause malformations of the urogenital system in laboratory animals, and the hypothesis has been put forward that in-utero exposure to environmental pollutants may be responsible for the increasing incidence of abnormalities of human male urogenital development, such as cryptorchidism and

Classification of Human Disorders of Reproduction and Development 635 hypospadias.4 Only a few epidemiological studies have been completed so far. One demonstrated an increased risk of cryptorchidism, but not hypospadias, in the sons of female gardeners.119 An ecological study found high rates of surgery to correct cryptorchidism in municipalities with high rates of pesticide use.120

Male exposures Female spouses of workers exposed to lead, solvents, and anesthetic gases have all been reported to have excesses of birth defects in isolated studies. However, none of these observations has been duplicated, nor has a clear model emerged, so at present there are no established or highly suspect male-mediated teratogenic hazards. However, it should be borne in mind that this possibility has not been readily evaluable in the environmental disaster situations in which most female-mediated causes were discovered because men and women were, in general, both exposed simultaneously.

Impaired postpartum development Female exposures, including breast milk Four important developmental toxicants are associated with delays in CNS development: lead,121 PCBs,116,122 methyl mercury,115 and ionizing radiation.97,123 Controversy exists over whether environmental exposures to the general population, such as from eating fish that contain low levels of PCBs or methyl mercury, pose a measurable risk to CNS development. Carefully conducted longitudinal studies suggest that subtle changes in outcomes such as intelligence quotient, learning, and behavior may be detectable following these exposures on a population basis, though not on an individual basis. Children whose mothers consumed Lake Michigan fish during pregnancy continued to have cognitive deficits through 11 years of age that correlated with cord blood PCB concentrations.124,125 Data from studies of two populations of women that consumed ocean fish, containing background levels of methyl mercury, daily during pregnancy have produced conflicting results. Reports from the Faroe Islands study demonstrated subtle dose-related effects on motor function, language, and memory even at 7 years of age,126,127 whereas the Seychelles Islands study did not report adverse effects through 5.5 years of age.128 One caveat with these studies, as with any epidemiological study of environmental exposures, is that the measured exposure (PCB or mercury concentrations) represents one of numerous potential exposures. It is possible that the observed effects are due to an unmeasured exposure that is present in similar proportions to the measured exposure. With ionizing radiation exposure, typically delivered between weeks 7–16 in utero, the cause of persistent deficits is likely inability to compensate for early diffuse injury. With chemical causes, the pathogenesis is more complicated because the toxicants themselves persist for long periods in the fetus and child and in the target CNS tissue, suggesting that ongoing neurologic injury may

occur through early childhood. Of clear therapeutic importance in this regard is the possibility of further postpartum exposures to the causal agent. For lead, the source generally is the home environment, including drinking water, paint, and dust. For methyl mercury and PCBs, breast milk is also a source of concern, because each agent is well concentrated in milk. In fact, anecdotal evidence suggests that postpartum exposures to mercury and PCBs in breast milk alone may cause CNS deficits, but few cases have been available for study in which exposure to the mother began, de novo, postpartum. Of interest, organic solvents, except in the setting of glue sniffing, have not been shown to cause chronic neurologic or developmental impairment in children when mothers have been exposed during pregnancy or breast feeding.129 Evidence for other neurotoxicants, such as various pesticides, is too limited to judge risk.

Male exposures Presently no model for male transmission of a developmental lesion has been demonstrated involving exposure prior to birth. Of course, the possibility of contamination of children from paternal work materials, such as indirect paraoccupational exposures from clothing remains of concern for such hazards as lead and pesticides, and other toxicants.

Childhood cancer Female exposures Although in-utero exposure to carcinogenic or mutagenic agents may be substantial when pregnant women are occupationally or environmentally exposed, there is very little data documenting increased risks for childhood cancer as a result. The single situation for which transplacental carcinogenesis has been unequivocally proved is for DES. Taken medicinally by mouth, this agent confers a substantial risk of a rare vaginal adenocarcinoma in female offspring after puberty.130 No counterpart has been shown in women exposed to estrogens at work, although this is clearly a concern. Several studies also have demonstrated risk for childhood leukemia in offspring of women exposed medically to ionizing radiation. Surprisingly, however, follow-up studies of atomic blast victims have not demonstrated a high rate of leukemia in children exposed to very high doses of radiation in utero, although increased risks of several cancers have been reported among individuals exposed to high doses of radiation in utero or during childhood compared to individuals exposed as adults.152 Increased risk of childhood brain tumor has been associated with nitrosamine exposure due to maternal consumption of cured meats during pregnancy in some studies,131,132 but not in others.133 Elevated risk of brain tumors has also been observed in children whose mothers were exposed to farm animals during pregnancy.134 Of note, two recent studies have not substantiated an association between maternal exposure to electromagnetic fields during gestation and childhood brain tumor risk.135,136

636 Disorders of Reproduction and Development Beyond these examples, other evidences of childhood cancer risk from maternal exposures before or during pregnancy have not been demonstrated, although some epidemiologic studies have reported higher rates of solvent exposures during pregnancy in mothers of children with brain tumors and an association between agricultural work and childhood leukemia.137

Male exposures The availability of cancer registries that provide information on parental occupations has led to the investigation of the relationship between paternal occupation and rates of the common childhood cancers. Employment in occupations with exposures to motor vehicle exhaust and pesticides have been linked to childhood leukemia, albeit inconsistently. A highly publicized study found a similar link with paternal exposure to ionizing radiation at an English nuclear power plant; however, subsequent studies have not substantiated this association, and it is at odds with the experience of atomic blast victims.97 In several studies, brain tumors have been associated with paternal work in agriculture and aircraft industries and with painting. A previous association between preconception paternal exposure to electromagnetic fields and childhood brain tumor risk has not been substantiated.138 Somewhat more consistently, several studies have found metal work, especially auto body and auto mechanic work, and welding, to be more common among fathers of children with Wilms’ tumor and other brain tumors.138,139 Similar associations have been noted with retinoblastoma and hepatoblastoma. At present, none of these associations can yet be considered causal. Nonetheless, some consistency in pattern and a biologic basis for plausibility make this a fruitful area for future preventive investigations and underscore the importance of male factors in healthful reproduction.

CLINICAL EVALUATION OF PATIENTS WITH REPRODUCTIVE AND DEVELOPMENTAL RISKS AND DISORDERS There are four clinical contexts in which individuals or couples typically seek medical advice regarding environmental and occupational effects on reproduction or development, or in which the practitioner should consider the possibility of such effects. These are: (1) infertile couples, (2) couples or women planning pregnancy or who are newly conceiving, (3) pregnant women or their partners who have had an exposure to an environmental hazard of concern before or during pregnancy, and (4) couples or women who have experienced an adverse reproductive or developmental outcome. For each of these clinical situations, the practitioner will want to consult additional sources of information about the possible reproductive hazards that are identified. These

include books7,140-142 electronic databases,143-145 and articles, including those referenced in this chapter.

Infertile couples The standard approach to the evaluation and management of infertility involves a detailed medical and reproductive history of both parents and the use of laboratory tests, such as semen analysis and evaluation of female hormonal cycling and anatomy as appropriate to identify the physiologic basis for reproductive failure. Prior to identifying the apparent problem, it is difficult to evaluate or manipulate environmental or occupational factors meaningfully. When the male partner has been demonstrated to have a low sperm count or otherwise abnormal sperm, it is appropriate to consider the common causes of testicular dysfunction and to review the male exposure history in detail, with emphasis on exposures within the past year, especially current ones. Any possible exposure to the hazards reviewed in Table 27.2.8 should form the basis for further consideration, irrespective of the possibility of another non-environmental risk. Furthermore, if no other risk factors are identified, suspect agents of interest, such as those mentioned in the section on hazards to male fertility, should be considered. If an established or possible suspect exposure is identified, it is reasonable to attempt to further qualify, and if possible, quantify the intensity of exposure (e.g., by blood or urine testing where applicable) before establishing an intervention plan. The choice of how to proceed depends on this investigation, with the clinician establishing some estimate of the likelihood that the environmental factor is important based on the agent, the intensity of exposure, and the coexistence of other risks. Having established some reasonable likelihood of effect, the next step is an intervention trial, removing the man from further exposure. Before such a plan is undertaken, at least two preintervention semen specimens should be obtained for analysis because of intrinsic variability. If azoospermia is demonstrated twice, a trial still is reasonable but the prognosis for recovery is poor, which should be considered in further planning; assessment of the role of prior exposures is a separate task, which is discussed later. After discussing the economic and personal ramifications with the patient, the therapeutic trial of removal should be continued for 6–12 months (2–5 cycles of spermatogenesis) before any judgment is made, with repeat semen analysis performed bimonthly. The choice as to whether to continue attempting pregnancy during this interval depends on other properties of the hazard, such as the potential for other male-mediated effects. In the case of lead, at least, deferral may be the conservative choice. If there is no improvement after a year, a period of continued removal may be scientifically reasonable (depending on, for example, body burden and disposition) but, practically speaking, is far less likely to be successful. At this stage, other factors should be given greater weight and the likelihood of trial failure added to the therapeutic equation.

Clinical Evaluation of Patients with Reproductive and Developmental Risks and Disorders 637 Infertility of the female is far less readily evaluated, in part because so little is known about factors that may cause it, and in part because there is no counterpart to semen for ready analysis. Further, not only are few factors studied but the physiologic basis for possible environmental effects remains unclear, such as whether hazards act only by disrupting ovulation or may cause loss of fecundity at a later stage, such as during tubal transport or implantation. The approach for women, therefore, is based on a more global perspective of the situation. If other female or couple-specific factors are not identified, it is reasonable to consider removing the woman from any environmental exposure that has been strongly associated with adverse outcomes (Table 27.2.9); indeed, this would be appropriate advice for the fertile couple as well. For agents less strongly associated with reproductive risk, the choice is more difficult and depends on the availability of other therapeutic options and the informed choice of the woman. In every case, it is very important that the patient appreciate how little is known about occupational risks to female fecundity and, therefore, that removal from the workplace or affected environment has only a small likelihood of remedy. On the other hand, it is not unreasonable to emphasize the risks to the pregnancy or offspring if conception is successful within the exposed environment. If a therapeutic trial of removal of the individual from the workplace is undertaken, the duration and endpoint should also be considered. If cycles are anovulatory, it is reasonable to wait for at least 6–12 months, because the preovulatory maturation of the corpus luteum takes about 3 months. If the trial is based on anovulation, direct or indirect hormonal evaluation is essential to establish the basis for success or failure. For women who are infertile despite normal ovulation, the only reasonable endpoint is conception.

Couples or women planning conception or who are newly pregnant Advising prospective parents about environmental risks begins by establishing a detailed profile of non-environmental factors that may be relevant and understanding the concerns and questions of the couple or woman patient. Has there been a prior adverse outcome in the history of the patient or a coworker or colleague? Has there been a specific exposure or risk of concern? Are there hidden fears about pregnancy that may be focused on the environment or workplace? If the answer to any of these is yes, the specific issues and circumstances should be directly addressed and resolved. If there is no extraordinary basis for concern, the general principles of preventive care are appropriate. Occupational, avocational, and other ambient environmental exposures should be noted, with special attention to factors summarized in Tables 27.2.10 and 27.2.11, and suspected agents described in the sections on effects of exposures during pregnancy. If there is potential exposure to any of these hazards, opportunities for prevention should be reviewed, which may be as simple as modifying a normal or chosen activity, such

as a hobby involving pesticides, metals, or solvents, or as complex as planning a job modification. The need for developing a detailed or expansive database is often inverse to the patient’s willingness or ability to modify or discontinue exposure. For example, it may be reasonable to recommend that a pregnant woman suspend use of garden chemicals during the period of pregnancy and breast feeding without much investigation, whereas considerable information might be necessary before advising an agricultural worker or horticulturist in the same situation. For exposures whose elimination or control is not easily accomplished, a more formal investigation including risk assessment and control strategy is indicated. Each potential exposure of interest and its dose should be identified to the greatest extent possible, because high or uncontrolled occupational exposures merit steps very different from trace environmental contamination, a point that is crucial for the patient to understand as well. Also, the potential for postpartum exposure to the agent in breast milk should be evaluated based on the principles listed earlier. Next, the database for each hazard should be reviewed, starting with the discussions in the preceding sections. The clinician may need to consult more detailed sources referenced above. With that information in hand, physician and patient together can proceed to establish the third element of the equation, the options for control. In this setting, these options include not only the factors normally considered in occupational and environmental health practice, such as a job change or use of personal protective equipment, but also certain alternatives, such as considering bottle feeding rather than breast feeding, or a choice to modify exposures for a specific limited period of time. For each control strategy, the two important aspects of concern are the effectiveness of the approach to reduce exposure and the acceptability of the approach to the patient, employer, or both. Once the relevant exposures are deciphered, databases obtained, and control options ranked, the choices for preventive management of a pregnancy may be undertaken. There are no definite rules, although elimination or strict and rigid control of the known and suspected factors is highly desirable. Similarly, because maternal illness itself is a risk factor for the fetus, occupational or environmental risks of sufficient magnitude to make the mother systemically ill, such as with solvent intoxication or metal fumes, are best completely avoided or rigidly controlled. Management of physical stresses, although included in the tables, are less straightforward because we presently lack adequate dose–response information; strenuous labor, cold, noise, vibration, and shift work are best curtailed, if possible, after the second trimester. Beyond these basic principles, there is wide latitude for choice, depending on the preferences of the patient. Although it is appropriate for the clinician to be conservative and generally risk averse, strong or sweeping generalizations to avoid everything, which may be reasonable when advising about use of medications that are not important in the treatment of life-threatening disease, are not easily adopted to environmental and occupational factors. Rather, more hazard-specific and balanced information is likely to be of

638 Disorders of Reproduction and Development most use to the patient, although gathering this information is far more demanding of the time and thoughtfulness of the clinician.

Pregnant women who have been exposed to a hazard Sometimes, women or couples consider the possibility of adverse effects from the environment after there has been an exposure of concern. Unlike the preventive setting described earlier, this situation demands a very specific and goaldirected role for the clinician, centered on a single specific task of risk assessment. At the outset, it must be appreciated that there will ultimately be only three options for subsequent action: reassurance, cessation of exposure, or recommendation for therapeutic termination of pregnancy. After establishing the basis for concern, the first challenge is determining the risk of an adverse outcome based on knowledge of the exposure. As always, this begins with the determination of what actual hazard was involved, the timing of exposure with respect to the pregnancy, some estimate of dose, and a characterization of other factors that may modify risk. Almost invariably, these facts must be reconstructed from the patient’s history; occasionally, a toxicant or its metabolites can be measured directly in the parent. Once an exposure is verified and an estimate of dose and timing with respect to pregnancy is made, the possibility of a measurable effect on the pregnancy must be determined using the available databases. Given the gravity of the situation, it is advisable to obtain the most complete and current information available for that toxicant. The information in this text may serve as a guide, but it should be supplemented by the most current animal and human data available, except in a case of gross overexposure to a wellcharacterized hazard such as lead or methyl mercury. If it appears possible that the exposure at its estimated dose may have an impact on the outcome of the pregnancy, then the timing of exposure should be assessed. Similarly, exposures that occur in the second trimester and beyond are unlikely to cause fetal malformations, although growth and development of the fetus remain at risk. If after review of exposure, dose, and timing there is still basis for concern about the outcome of the pregnancy, then it becomes important to quantify the concern. For example, exposures that occur in the second trimester and beyond are unlikely to cause gross fetal malformations, although growth and development of the fetus remain at risk. Using conservative estimates from the available data, how likely is the outcome of concern to occur? For childhood cancers or fetal malformations, these are best expressed as straight probabilities of an all-or-nothing outcome. For developmental or growth effects, these possibilities may be conceptualized on a continuous scale from minimal to severe; however, inherent uncertainties often limit specific characterization. As noted, the real challenge in this situation is that, at the end, the choices for action are extremely limited. Although there may be some small room for action, such as taking

steps to hasten removal of a toxicant or reduce later exposures, the basic choice comes down to therapeutic abortion versus reassurance. Because there is almost always some degree of uncertainty regarding the risk of an adverse developmental outcome and because many people have very strong pre-existing feelings towards therapeutic abortion, the clinician in these situations must present choices in the clearest and most distinct way possible. It is best for the clinician to very clearly delineate the risks and associated uncertainties, to present pregnancy termination as an option in cases with very high risks of developmental effects, to provide reassurance for continuation of pregnancy in the majority of cases in which the risk is small, and, perhaps most importantly, to be supportive of what ultimately must be the patient’s choice.

Patients who have had adverse outcomes Evaluation of a couple who has experienced an adverse reproductive outcome in the past may be important for two reasons: (1) concern about future pregnancies; and (2) attribution for medical-legal purposes. When the medicallegal issues are involved, the approaches are similar to those used to evaluate any environmental or occupational injury or illness, using the principles outlined in Chapter 3 of this text and the data provided earlier, as supplemented by other detailed and updated sources of information. Evaluation of a prior event for the purpose of subsequent reproductive planning requires a somewhat different formulation. In general, patients who have had a prior adverse outcome, such as a developmentally abnormal child, fear a recurrence based on some underlying, immutable risk factor such as genetics. Recognition of a preventable or remediable basis for prior risk could provide a basis for reassurance and optimism about future pregnancies. However, depending on the circumstances surrounding exposure with the previous pregnancy, issues of self-blame and regret may complicate the benefit of this message. Although it is important to decipher the patient’s actual agenda in seeking consultation around an adverse event, it is best to separate the past and the future. First, one can evaluate the likelihood that a defined exposure may have been responsible for a prior adverse outcome. Second, based on this calculation and all other available information about the couple, such as current environment, persistence of prior exposures in the body, and results of genetic or other evaluations that have occurred, the prospects for the future can be estimated. The possibility of an immutable risk needs to be specifically considered in the patient’s interest, so that referral for genetic or other appropriate counseling can be undertaken. After these steps are complete, the general preventive approach for future pregnancies can be laid out as described earlier, modified by the special knowledge about the past and the unique perspective the patient will undoubtedly have because of it. Importantly, the recommendation for the future may appear to contradict the verdict on the past, a paradox the clinician must anticipate and be prepared to

Prevention of Reproductive and Developmental Risks in an Occupational Setting 639 explain. For example, a couple may have just lost a 2-yearold child to leukemia, which they might attribute to lowlevel ionizing radiation exposure to the father in the year prior to conception. Although support for this possibility in the current literature is small, advising the couple to reduce or eliminate such exposures prior to the next conception would not be unreasonably risk averse. In such a case, however, it would also be very important to be certain that an underlying heritable risk, such as Fanconi’s syndrome, is not a factor.

PREVENTION OF REPRODUCTIVE AND DEVELOPMENTAL RISKS IN AN OCCUPATIONAL SETTING One of the foremost challenges for corporations in the United States and other developed countries has been the development of reproductive and developmental health policies consistent with good preventive principles, existing exposure regulations, and the law. Although our knowledge of reproductive and developmental effects of hazards suggest possibly differential effects on each parent and fetus, United States law, subsequent to the case of Johnson Controls, requires women be given the right to work in any place where men may work, as long as they are adequately

informed about possible fetal risks should they get pregnant. In other words, gender-specific administrative controls may not be used to prevent reproductive or developmental harm.146 In this section, we outline some approaches that may accomplish the goals of preventive reproductive health while adhering to these social mandates (Fig. 27.2.3). There are four key components that we recommend for institution in all workplaces and other defined organizations. 1. Primary control of hazards to levels safe for men and nonpregnant and pregnant women. Although existing occupational standards have not, in general, required exposures be reduced to levels believed to minimize risks to reproduction and development, such levels are definable and achievable in many, if not most, workplaces by engineering controls; they may be achievable in all, with the addition of personal protective equipment and administrative controls.147 The adherence to such strict standards, which are gender neutral and as safe as possible for reproduction and development, is the cornerstone of an effective plan. 2. Hazard identification. All agents known or suspected to present a risk for reproduction or development should be identified clearly as such and well marked within the work environment, so that inadvertent exposures become far less likely. 3. Reproductive and developmental health education. Because the existing standards have not been calculated to

Alleged reproductive and/or developmental toxicant or hazard

Animal and/or human studies confirm hazard

Inadequate animal and/or human data

Animal and/or human studies adequate and negative

Define safety level if possible

Initial studies for risk assessment

Reassurance No job change

Do any jobs exceed safety level?

Interim measures pending study results

Yes Reduce exposures to lowest achievable level

No

Inform workers of possible risk

Interim measure Reduce or eliminate exposure

Worker decision regarding acceptability of risk

Not acceptable Reassurance No job change

Job change or transfer

Acceptable Continue in job

Figure 27.2.3: Proposed scheme for management of reproductive and developmental toxicants in the occupational setting. Many occupational exposures fall in the category identified as ‘inadequate animal and/or human data’. In this case, it will be necessary to gather additional information as well as put interim measures in place to both inform and protect the workers. Exposures should be reduced to the lowest level achievable using engineering controls and personal protective equipment. The workers also must be informed that the exposure may pose a hazard to reproduction or development. In this setting, the workers should be allowed to determine whether or not the potential for effect is sufficient for concern. If not, the worker may reasonably decide to continue on the job. If there is concern, the worker should be provided with a choice of alternative jobs, without professional disincentives, that carry a minimal reproductive or development risk.

640 Disorders of Reproduction and Development provide a margin of safety for reproductive and developmental health, and because the basis for personal decisions must be made with adequate information, broad-based and unbiased education of all potentially exposed workers of both sexes is crucial. 4. A flexible work-modification plan. Despite the institution of controls to minimize risk in a gender neutral fashion, some workers are likely to choose a greater level of risk aversion when planning pregnancy, after conception, or while breast feeding. There are many excellent medical reasons and other understandable personal reasons for this choice. Ideally, these choices should be available to individual employees of either sex without undue company scrutiny, interference, or disincentives, such as loss of wage, job security, or career path. Although the experience in most companies is that few employees actually use these options, their availability is widely appreciated by employees as a benefit. More important, such flexibility facilitates the practice of the highest quality preventive medicine.148

CONCLUSIONS Physicians and other healthcare professionals are increasingly called on to counsel patients, unions, corporations, and public health agencies concerning the impact of occupational exposures on reproductive and developmental health. It is important to recognize certain fundamental principles in responding to these concerns: (1) the available data on human reproductive and developmental hazards are deficient; (2) because human data are lacking, it is essential to extrapolate information from animal data (this is also a central component of the protection of public reproductive and developmental health); (3) the criteria or stringency of data analysis for assigning causation are substantially different from those required for the protection of the public health; (4) our knowledge of reproductive and developmental toxicity is developing rapidly and is expected to change substantially in the future; and (5) finally, men and women must be treated equally so that reproductive and developmental health are protected.

References 1.

2.

3.

4.

5.

6.

7.

Colborn T, vom Saal FS, Soto AM. Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ Health Perspect 1993; 101:378–84. Colborn T, Dumanoski D, Myers JP. Our stolen future: are we threatening our fertility, intelligence, and survival? A scientific detective story. New York: Dutton, 1996 Sharpe RM, Skakkebaek NE. Are oestrogens involved in falling sperm counts and disorders of the male reproductive tract? Lancet 1993; 341:1392–5. Toppari J, Larsen JC, Christiansen P, et al. Male reproductive health and environmental xenoestrogens. Environ Health Perspect 1996; 104(Suppl 4):741–803. Safe DH. Environmental and dietary estrogens and human health: is there a problem? Environ Health Perspect 1994; 103:346–51. Daston GP, Gooch JW, Breslin WJ, et al. Environmental estrogens and reproductive health: a discussion of the human and environmental data. Reprod Toxicol 1997; 11:465–81. Frazier LM, Hage ML, eds. Reproductive hazards of the workplace, 1st edn. New York: John Wiley and Sons, 1998.

8.

9.

10.

11.

12.

13. 14.

15. 16. 17.

18.

19.

20. 21.

22.

23.

24. 25. 26.

27.

28.

29. 30. 31.

Abma J, Chandra A, Mosher W, Peterson L, Piccinino L. Fertility, family planning and women’s health: new data from the 1995 National Survey of Family Growth. National Center for Health Statistics. Vital Health Stats 1997; 23. Ford WC, North K, Taylor H, Farrow A, Hull MG, Golding J. Increasing paternal age is associated with delayed conception in a large population of fertile couples: evidence for declining fecundity in older men. The ALSPAC Study Team (Avon Longitudinal Study of Pregnancy and Childhood). Hum Reprod 2000; 15:1703–8. Jensen TK, Henriksen TB, Hjollund NH, et al. Adult and prenatal exposures to tobacco smoke as risk indicators of fertility among 430 Danish couples. Am J Epidemiol 1998; 148:992–7. Alderete E, Eskenazi B, Sholtz R. Effect of cigarette smoking and coffee drinking on time to conception. Epidemiology 1995; 6:403–8. Jensen TK, Hjollund NH, Henriksen TB, et al. Does moderate alcohol consumption affect fertility? Follow up study among couples planning first pregnancy. Br Med J 1998; 317:505–10. Hakim RB, Gray RH, Zacur H. Alcohol and caffeine consumption and decreased fertility. Fertil Steril 1998; 70:632–7. Jensen TK, Henriksen TB, Hjollund NH, et al. Caffeine intake and fecundability: a follow-up study among 430 Danish couples planning their first pregnancy. Reprod Toxicol 1998; 12:289–95. Wilcox AJ, Weinberg CR, O’Connor JF, et al. Incidence of early loss of pregnancy. N Engl J Med 1988; 319:189–94. Fan XG, Zheng ZQ. A study of early pregnancy factor activity in preimplantation. Am J Reprod Immunol 1997; 37:359–64. Mesrogli M, Dieterle S. Embryonic losses after in vitro fertilization and embryo transfer. Acta Obstet Gynecol Scand 1993; 72:36–8. Ness RB, Grisso JA, Hirschinger N, et al. Cocaine and tobacco use and the risk of spontaneous abortion. N Engl J Med 1999; 340:333–9. Chatenoud L, Parazzini F, di Cintio E, et al. Paternal and maternal smoking habits before conception and during the first frimester: relation to spontaneous abortion. Ann Epidemiol 1998; 8:520–6. Abel EL. Maternal alcohol consumption and spontaneous abortion. Alcohol Alcoholism 1997; 32:211–9. Windham GC, Von Behren J, Fenster L, Schaefer C, Swan SH. Moderate maternal alcohol consumption and risk of spontaneous abortion. Epidemiology 1997; 8:509–14. Saling E. Basic aspects of prematurity prevention and results achieved by a suitable, simple program. J Perinatal Med 1998; 26:466–8. MMWR. Use of folic acid-containing supplements among women of childbearing age –United States, 1997. MMWR 1997; 47:131–4. Koren G, Pastuszak A, Ito S. Drugs in pregnancy. N Engl J Med 1998; 338:1128–37. Evenson D, Jost L. Sperm chromatin structure assay: DNA denaturability. Methods Cell Biol 1994; 42(B):159–76. Bonde JPE, Ernst E, Jensen TK, et al. Relation between semen quality and fertility: a population-based study of 430 firstpregnancy planners. Lancet 1998; 352:1172–7. Rutledge JC, Generoso WM. Malformations in pregastrulation developmental toxicology. In: Korach KS, ed. Reproductive and developmental toxicology. New York: Marcel Dekker, 1998;73–86. Welsch F, Stedman DB. Inhibition of intercellular communication between normal human embryonal palatal mesenchyme cells by teratogenic glycol ethers. Environ Health Perspect 1984; 57:125–33. Wolff MS. Occupationally derived chemicals in breast milk. Am J Ind Med 1983; 4:259–81. Mattison DR, Wohlleb J, To T, et al. Pesticide concentrations in Arkansas breast milk. J Ark Med Soc 1992; 88:553–7. Jensen AA, Slovak SA. Chemical contaminants in human milk. Boca Raton: CRC Press, 1991.

Conclusions 641 32. Jelovsek FR, Mattison DR, Chen J. Prediction of risk for human development toxicity: how important are animal studies? Obstet Gynecol 1989; 74:624–36. 33. Mattison DR, Working PK, Blazak WF, et al. Criteria for identifying and listing substances known to cause reproductive toxicity under California’s Proposition 65. Reprod Toxicol 1990; 4:163–75. 34. Mattison DR, Hanson J, Kochhar DM, Rao KS. Criteria for identifying and listing substances known to cause developmental toxicity under California’s Proposition 65. Reprod Toxicol 1989; 3:3. 35. Gray LEJ. Tiered screening and testing strategy for xenoestrogens and antiandrogens. Toxicol Lett 1998; 102103:677–80. 36. Carlsen E, Giwercman A, Keiding N, Skakkebaek NE. Evidence for decreasing quality of semen during past 50 years. Br Med J 1992; 305:609–13. 37. Bromwich P, Cohen J, Stewart I, Walker A. Decline in sperm counts: an artefact of changed reference range of ‘normal’? Br Med J 1994; 309:19–22. 38. Apostoli P, Kiss P, Porru S, Bonde JP, Vanhoorne M. Male reproductive toxicity of lead in animals and humans. ASCLEPIOS Study Group. Occup Environ Med 1998; 55:364–74. 39. Whorton D, Krauss RM, Marshall S, Milby TH. Infertility in male pesticide workers. Lancet 1977; 2:1259–61. 40. Whorton MD, Foliart DE. Mutagenicity, carcinogenicity, and reproductive effects of dibromochloropropane (DBCP). Mutat Res 1983; 123:13–30. 41. Potashnik G, Porath A. Dibromochloropropane (DBCP): a 17year reassessment of testicular function and reproductive performance. J Occup Environ Med 1995: 37:1287–92. 42. Kim Y, Jung K, Hwang T, et al. Hematopoietic and reproductive hazards of Korean electronic workers exposed to solvents containing 2-bromopropane. Scand J Work Environ Health 1996; 22:387–91. 43. Welch LS, Schrader SM, Turner TW, Cullen MR. Effects of exposure to ethylene glycol ethers on shipyard painters: II. Male reproduction. Am J Ind Med 1988; 14:509–26. 44. Gray TJB, Moss EJ, Creasy DM, Gangolli SD. Studies on the toxicity of some glycol ethers and alkoxyacetic acids in primary testicular cell cultures. Toxicol Appl Pharmacol 1985; 79:490–501. 45. Mieusset R, Bujan L. Testicular heating and its possible contributions to male infertility: a review. Int J Androl 1995; 18:169–84. 46. Jordan S, Hasegawa CM, Keehn RJ. Testicular changes in atomic bomb survivors. Arch Pathol 1966; 82:542–54. 47. Ash P. The influence of radiation on fertility in man. Br J Radiol 1980; 53:271. 48. Harrington JM, Stein GF, Rivera RO, de Morales AV. The occupational hazards of formulating oral contraceptives: a survey of plant employees. Arch Environ Health 1978; 33:12–5. 49. Schrader SM, Turner TW, Ratcliffe JM. The effects of ethylene dibromide on semen quality: a comparison of short-term and chronic exposure. Reprod Toxicol 1988; 2:191–8. 50. Ratcliffe JM, Schraeder SM, Steenland K, Clapp DE, Turner T, Hornung RW. Semen quality in papaya workers with longterm exposure to ethylene dibromide. Br J Ind Med 1987; 44:317–26. 51. De Celis R, Feria-Velasco A, González-Unzaga M, Torres-Calleja J, Pedrón-Nuevo N. Semen quality of workers occupationally exposed to hydrocarbons. Fertil Steril 2000; 73:221–8. 52. Svensson B-G, Nise G, Erfurth E-M, Nilsson A, Skerfving S. Hormone status in occupational toluene exposure. Am J Ind Med 1992; 22:99–107. 53. Svensson B-G, Nise G, Erfurth EM, Olsson, H. Neuroendocrine effects in printing workers exposed to toluene. Br J Ind Med 1992; 49:402–8. 54. Luderer U, Morgan MS, Brodkin CA, Kalman DA, Faustman EM. Reproductive endocrine effects of acute exposure to

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

toluene in men and women. Occup Environ Med 1999; 56:657–66. Plenge-Bönig A, Karmaus W. Exposure to toluene in the printing industry is associated with subfecundity in women but not in men. Occup Environ Med 1998; 56:443–8. Kolstad HA, Bisanti L, Roeleveld N, Baldi R, Bonde JP, Joffe M. Time to pregnancy among male workers of the reinforced plastics industry in Denmark, Italy and The Netherlands. Scand J Work Environ Health 2000; 26:353–8. Padungtod C, Lasley BT, Christiani DC, Ryan LM, Xu X. Reproductive hormone profiles among pesticide factory workers. J Occup Environ Med 1998; 40:1038–47. Padungtod C, Savitz DA, Overstreet JW, Christiani DC, Ryan LM, Xu X. Occupational pesticide exposure and semen quality among Chinese workers. J Occup Environ Med 2000; 42:982–92. Boekelheide K. 2,5-Hexanedione alters microtubule assembly: I. Testicular atrophy, but not nervous system toxicity correlates with enhanced tubulin polymerization. Toxicol Appl Pharmacol 1987; 88:370–82. Mori K, Kaido M, Fujishiro K, et al. Dose dependent effects of inhaled ethylene oxide on spermatogenesis in rats. Br J Ind Med 1991; 48:270–4. Hjollund NHI, Bonde JPE, Jensen TK, et al. A follow-up study of male exposure to welding and time to pregnancy. Reprod Toxicol 1998; 12:29–37. Klinefelter GR, Laskey JW, Perreault SD, et al. The ethane dimethylsulfonate-induced decrease in the fertilizing ability of cauda epididymal sperm is independent of the testis. J Androl 1994; 15:318–27. Klinefelter GR, Hess RA. Toxicology of male excurrent ducts and accessory sex glands. In: Korach KS, ed. Reproductive and developmental toxicology. New York: Marcel Dekker, 1998;553–91. Mylchreest E, Sar M, Cattley RC, Foster PM. Disruption of androgen-regulated male reproductive development by di(n-butyl) phthalate during late gestation in rats is different from flutamide. Toxicol Appl Pharmacol 1999; 156:81–95. Gray LE, Kelce WR, Monosson E, Ostby JS, Birnbaum LS. 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 1995; 131:108–18. Kelce WR, Lambright CR, Gray LE Jr, Roberts KP. Vinclozolin and p,p′-DDE alter androgen-dependent gene expression: in vivo confirmation of an androgen receptor-mediated mechanism. Toxicol Appl Pharmacol 1997; 142:192–200. Gray LE Jr, Ostby J, Ferrell J, et al. A dose–response analysis of methoxychlor-induced alterations of reproductive development and function in the rat. Fundamental Appl Toxicol 1989; 12:92–108. Rowland AS, Baird DD, Weinberg CR, Shore DL, Shy CM, Wilcox AJ. Reduced fertility among women employed as dental assistants exposed to high levels of nitrous oxide. N Engl J Med 1992; 327:993–7. Bodin L, Axelsson G, Ahlborg GJ. The association of shift work and nitrous oxide exposure in pregnancy with birth weight and gestational age. Epidemiology 1999; 10:429–36. Taskinen HK, Kyyrönen P, Sallmén M, et al. Reduced fertility among female wood workers exposed to formaldehyde. Am J Ind Med 1999; 36:206–12. Zielhuis G, Peelen SJM, Florack EIM, Roeleveld N. Hospital work and fecundability. Scand J Work Environ Health 1999; 25(Suppl 1):47–8. Lim CH, Maeng SH, Lee JY, et al. Effects of 2-bromopropane on the female reproductive function in Sprague-Dawley rats. Ind Health 1997; 35:278–84. Yu M-L, Guo YL, Hsu C-C, Rogan WJ. Menstruation and reproduction in women with polychlorinated biphenyl (PCB) poisoning: long-term follow-up interviews of the women

642 Disorders of Reproduction and Development

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90. 91.

92.

93.

from the Taiwan Yucheng Cohort. Int J Epidemiol 2000; 29:672–7. Buck GM, Sever LE, Mendola P, Zielezny M, Vena JE. Consumption of contaminated sport fish from Lake Ontario and time-to-pregnancy. Am J Epidemiol 1997; 146:949–54. Rier SE, Martin DC, Bowman RE, Becker JL. Immunoresponsiveness in endometriosis: implications of estrogenic toxicants. Environ Health Perspect 1995; 103(Suppl 7):151–6. Bruner-Tran KL, Rier SE, Eisenberg E, Osteen KG. The potential role of environmental toxins in the pathophysiology of endometriosis. Gynecol Obstet Invest 1999; 48(Suppl 1):45–56. Gray LE, Ostby JS. In utero 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD) alters reproductive morphology and function in female rat offspring. Toxicol Appl Pharmacol 1995; 133:285–94. Selvan SG, Lindbohm M-L, Hornung RW, Hemminki K. A study of occupational exposure to antineoplastic drugs and fetal loss in nurses. N Engl J Med 1985; 313:1173–8. Hemminki K, Mutanen P, Saloniemi I, Niemi ML, Vaino H. Spontaneous abortions in hospital staff engaged in sterilising instruments with chemical agents. Br Med J 1982; 285:1461–3. Rowland AS, Baird DD, Shore DL, Darden B, Wilcox AJ. Ethylene oxide exposure may increase the risk of spontaneous abortion, preterm birth, and postterm birth. Epidemiology 1996; 7:363–8. Hemminki K, Kyyronen P, Lindbohm M. Spontaneous abortions and malformations in the offspring of nurses exposed to anesthetic gases, cytostatic drugs, and other potential hazards in hospitals, based on registered information of outcome. J Epidemiol Commun Health 1985; 39:141–7. Rowland AS, Baird DD, Shore DL, Weinberg CR, Savitz DA, Wilcox AJ. Nitrous oxide and spontaneous abortion in female dental assistants. Am J Epidemiol 1995; 141:531–8. McMartin KI, Chu M, Kopecky E, Einarson TR, Koren G. Pregnancy outcome following maternal organic solvent exposure: a meta-analysis of epidemiologic studies. Am J Ind Med 1998; 34:288–92. Gold EB, Tomich E. Occupational hazards to fertility and pregnancy outcome. Occup Med State Art Rev 1994; 9:435–69. Pastides H, Calabrese EJ, Hosmer DW Jr, Harris DR Jr. Spontaneous abortion and general illness symptoms among semiconductor manufacturers. J Occup Med 1988; 30:543–51. Huel G, Mergler D, Bowler R. Evidence for adverse reproductive outcomes among women microelectronics assembly workers. Br J Ind Med 1990; 47:400–4. Lipscomb JA, Fenster L, Wrensch M, Shusterman D, Swan S. Pregnancy outcomes in women potentially exposed to occupational solvents and women working in the electronics industry. J Occup Med 1991; 33:597–604. Schenker MB, Gold EB, Beaumont JJ, et al. Association of spontaneous abortion and other reproductive effects with work in the semiconductor industry. Am J Ind Med 1995; 28:639–59. Taskinen H, Kyyronen P, Hemminki K, Hoikkala M, Lajunen K, Lindbohm M-L. Laboratory work and pregnancy outcome. J Occup Med 1994; 36:311–9. Ng TP, Foo SC, Yoong T. Risk of spontaneous abortion in workers exposed to toluene. Br J Ind Med 1992; 49:804–8. Xu X, Cho S-I, Sammel M, et al. Association of petrochemical exposure with spontaneous abortion. Occup Environ Med 1998; 55:31–6. Doyle P, Roman E, Beral V, Brookes M. Spontaneous abortion in dry cleaning workers potentially exposed to perchloroethylene. Occup Environ Med 1997; 54:848–53. Jarrell J, Gocmen A, Foster W, Brant R, Chan S, Sevcik M. Evaluation of reproductive outcomes in women inadvertently exposed to hexachlorobenzene in Southeastern Turkey in the 1950s. Reprod Toxicol 1998; 12:469–76.

94. Fuortes L, Clark MK, Kirchner HL, Smith EM. Association between female infertility and agricultural history. Am J Ind Med 1997; 31:445–51. 95. Nurminen T. Shift work and reproductive health. Scand J Work Environ Health 1998; 24(Suppl 3):28–34. 96. Nesbitt T. Ergonomic exposures. In: Frazier LM, Hage, ML, eds. Reproductive hazards of the workplace. New York: John Wiley and Sons, 1998;431–62. 97. Suruda AJ. Radiation. In: Frazier LM, Hage, ML, eds. Reproductive hazards of the workplace. New York: John Wiley and Sons, 1998;367–90. 98. Schnorr TM, Grajewski BA, Thun MJ, et al. Video display terminals and the risk of spontaneous abortion. N Engl J Med 1991; 324:727–33. 99. Delpizzo V. Epidemiological studies of work with video display terminals and adverse pregnancy outcomes (1984–1992). Am J Ind Med 1994; 26:465–80. 100. Belanger K, Leaderer B, Hellenbrand K, et al. Spontaneous abortion and exposure to electric blankets and heated water beds. Epidemiology 1998; 9:36–42. 101. Ouellet-Hellstrom R, Stewart WF. Miscarriages among female physical therapists who report using radio- and microwavefrequency electromagnetic radiation. Am J Epidemiol 1993; 138:775–86. 102. Neugebauer R, Kline J, Stein Z, Shrout P, Warburton D, Susser M. Association of stressful life events with chromosomally normal spontaneous abortion. Am J Epidemiol 1996; 143:588–96. 103. Taskinen H, Anttila A, Lindbohm M-L, Salimen M, Himminki K. Spontaneous abortions and congenital malformations among the wives of men occupationally exposed to organic solvents. Scand J Work Environ Health 1989; 15:354–62. 104. Hjollund NHI, Bonde JPE, Hansen KS. Male-mediated risk of spontaneous abortion with reference to stainless steel welding. Scand J Work Environ Health 1995; 21:272–6. 105. Sokas RK. Metals. In: Frazier LM, Hage, ML, eds. Reproductive hazards of the workplace. New York: John Wiley and Sons, 1998;123–61. 106. Shukla R, Bornschein R, Dietrich K, et al. Fetal and infant lead exposure: effects on growth in stature. Pediatrics 1989; 84:604–12. 107. Jacobson JL, Jacobson SW, Humphrey HE. Effects of exposure to PCBs and related compounds on growth and activity in children. Neurotoxicol Teratol 1990; 12:319–26. 108. Eskenazi B, Bracken MB, Holford TR, Grady J. Exposure to organic solvents and hypertensive disorders of pregnancy. Am J Ind Med 1988; 14: 177–88. 109. Mitchell LV, DeHart RL. Temperature, hypoxia, and atmospheric pressure. In: Frazier LM, Hage, ML, eds. Reproductive hazards of the workplace. New York: John Wiley and Sons, 1998;415–28. 110. Berkowitz GS. Employment-related physical activity and pregnancy outcome. J Am Med Women Assoc 1995; 50:16–9. 111. Florack EIM, Pellegrino AEMC, Zielhuis GA, Rolland R. Influence of occupational physical activity on pregnancy duration and birthweight. Scand J Work Environ Health 1995; 21:199–207. 112. Alderman BW, Zhao H, Holt VL, Watts DH, Beresford SAA. Maternal physical activity in pregnancy and infant size for gestational age. Ann Epidemiol 1998; 8:513–9. 113. Magann EF, Evans SF, Newnham JP. Employment, exertion, and pregnancy outcome: assessment by kilocalories expended each day. Am J Obstet Gynecol 1996; 175:182–7. 114. Gerber G, Leonard A, Jacquet P. Toxicity, mutagenicity and teratogenicity of lead. Mutat Res 1980; 76:115–41. 115. Burbacher T, Rodier R, Weiss B. Methyl mercury developmental neurotoxicity: a comparison of effects in humans and animals. Neurotoxicol Teratol 1990; 12:191–202. 116. Chen YCJ, Guo YO, Hsu CC, Rogan WJ. Cognitive development of Uu-Chen (oil disease) children prenatally exposed to heat degraded PCBs. JAMA 1992; 268:3213–8.

Conclusions 643 117. Wilkins-Haug L, Gabow P. Toluene abuse during pregnancy; obstetric complications and perinatal outcomes. Obstet Gynecol 1991; 4:504–9. 118. Cordier S, Bergeret A, Goujard J, et al. Congenital malformations and maternal occupational exposure to glycol ethers. Epidemiology 1997; 8:355–63. 119. Weidner ID, Moller H, Jensen TK, Skakkebaek NE. Cryptorchidism and hypospadias in sons of gardeners and farmers. Environ Health Perspect 1998; 106:793–6. 120. García-Rodríguez J, García-Martín M, Nogueras-Ocaña M, et al. Exposure to pesticides and cryptorchidism: geographical evidence of a possible association. Environ Health Perspect 1996; 104:1090–5. 121. Bellinger D, Needleman H. Prenatal and early postnatal exposure to lead: developmental effects, correlates, and implications. Int J Mental Health 1985; 14:78–111. 122. Yu M-L, Hsu C-C, Gladen BC, Rogan WJ. In utero PCB/PCDF exposure: relation of developmental delay to dysmorphology and dose. Neurotoxicol Teratol 1991; 13:195–202. 123. Wood JW, Johnson KG, Omri Y. In utero exposure to Hiroshima atomic bomb. An evaluation of head size and mental retardation: twenty years later. Pediatrics 1967; 39:385–92. 124. Jacobson JL, Jacobson SW, Humphrey HE. Effects of in utero exposure to polychlorinated biphenyls and related contaminants on cognitive functioning in young children. J Pediatrics 1990; 116:38–45. 125. Jacobson JL, Jacobson SW. Evidence for PCBs as neurodevelopmental toxicants in humans. Neurotoxicology 1997; 18:415–24. 126. Grandjean P, Weihe P, White R, Debes F. Cognitive performance of children prenatally exposed to ‘safe’ levels of methylmercury. Environ Res 1998; 77:165–72. 127. Grandjean P, Budtz-Jorgensen E, White RF, et al. Methylmercury exposure biomarkers as indicators of neurotoxicity in children aged 7 years. Am J Epidemiol 1999; 150:301–5. 128. Davidson PW, Myers GJ, Cox C, et al. Effects of prenatal and postnatal methylmercury exposure from fish consumption on neurodevelopment: outcomes at 66 months of age in the Seychelles child development study. JAMA 1998; 280:701–7. 129. Eskenazi B, Gaylord L, Bracken MB, Brown D. In utero exposure to organic solvents and human neurodevelopment. Develop Med Child Neurol 1988; 30:492–501. 130. Herbst AL, Ulfelder H, Poskanzer DC. Adenocarcinoma of the vagina: association of maternal stilbestrol therapy with tumor appearance in young women. N Engl J Med 1971; 284:878–81. 131. Bunin GR, Kuijten RR, Boesel CP, Buckley JD, Meadows AT. Maternal diet and risk of astrocytic glioma in children: a report from the childrens cancer group (United States and Canada). Cancer Causes Control 1994; 5:177–87. 132. Preston-Martin S, Pogoda JM, Mueller BA, Holly EA, Lijinsky W, Davis RL. Maternal consumption of cured meats and vitamins in relation to pediatric brain tumors. Cancer Epidemiol Biomark Prev 1996; 5:599–605.

133. Lubin F, Farbstein H, Chetrit A, et al. The role of nutritional habits during gestation and child life in pediatric brain tumor etiology. Int J Cancer 2000; 86:139–43. 134. Holly EA, Bracci PM, Mueller BA, Preston-Martin S. Farm and animal exposures and pediatric brain tumors: results from the United States West Coast Childhood Brain Tumor Study. Cancer Epidemiol Biomark Prev 1998; 7:797–802. 135. Preston-Martin S, Gurney JG, Pogoda JM, Holly EA, Mueller BA. Brain tumor risk in children in relation to use of electric blankets and water bed heaters. Results from the United States West Coast Childhood Brain Tumor Study. Am J Epidemiol 1996; 143:1116–22. 136. Gurney JG, Mueller BA, Davis S, Schwartz SM, Stevens RG, Kopecky KJ. Childhood brain tumor occurrence in relation to residential power line configurations, electric heating sources, and electric appliance use. Am J Epidemiol 1996; 143:120–28. 137. Savitz, DA, Chen J. Prenatal occupation and childhood cancer: review of epidemiologic studies. Environ Health Perspect 1990; 88:325–37. 138. Wilkins JR, Wellage LC. Brain tumor risk in offspring of men occupationally exposed to electric and magnetic fields. Scand J Work Environ Health 1996; 22:339–45. 139. Olshan AF, Breslow NE, Daling JR, et al. Wilms’ tumor and paternal occupation. Cancer Res 1990; 50:3212–7. 140. Paul M, ed. Occupational and environmental reproductive hazards. Baltimore: Williams & Wilkins, 1993. 141. Korach KS, ed. Reproductive and developmental toxicology. New York: Marcel Dekker, 1998. 142. Schettler T, Solomon G. Generations at risk: reproductive health and the environment. Cambridge: MIT Press, 1999. 143. REPRORISK. Micromedex Healthcare Series Databases. Micromedex Inc., Englewood, CO. (12/00). 144. TOXNET. Toxicology Data Network. Toxicology and environmental health information program, National Library of Medicine, 2000. http://toxnet.nlm.nih.gov//. 145. MEDLINE. PUBMED. National Library of Medicine, 2000. http://www.ncbi.nlm.nih.gov/PubMed//. 146. Mattison DR. Exclusion of fertile women from the workplace: bad medicine, worse law. J Ark Med Soc 1990; 86:491–2. 147. Stijkel A, van Dijk FJH. Developments in reproductive risk management. Occup Environ Med 1995; 52:294–303. 148. Brooks L, Merkel SF, Glowatz MJ, Comstock ML, Shoner LG. A comprehensive reproductive health program in the workplace. Am Ind Hyg Assoc J 1994; 55:352–7. 149. Mattison DR. An overview of biologic markers in reproductive and developmental toxicology: concepts, definitions and use in risk assessment. Biomed Environ Sci 1991; 4:8–34. 150. Brent RL, Beckman DA. Principles of teratology. In: Evans MI, ed. Reproductive risks and prenatal diagnosis. Norwalk: Appleton and Lange, 1992. 151. Rayburn WF, Zuspan FZ, eds. Drug therapy in obstetrics and gynecology, 3rd edn. St Louis: Mosby Year Book, 1992. 152. UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation) 2000. Sources and effects of ionizing radiation. Report to the General Assembly, with scientific annexes. New York: United Nations, 2000.

Chapter 28 Neurologic and Psychiatric Disorders 28.1 Central Nervous System Diseases Jordan A Firestone, William T Longstreth Jr Chemical agents capable of damaging the central nervous system (CNS) are ubiquitous in the environment. Industrial processes are notorious sources for some of the most well known of these neurotoxins, which contaminate both the worksite and the surrounding environment. The United States Environmental Protection Agency lists over 65,000 chemicals currently used in the US for commercial purposes, adding 2000–3000 new ones each year.1 Reflecting the growing public concern about chemical exposures, the Environmental Protection Agency2 and the Food and Drug Administration3 have issued revised guidelines for risk assessment, including specific test batteries for neurotoxicity and developmental neurotoxicity. Unfortunately, only a few chemicals have been thoroughly tested, and the guidelines have been criticized for being ineffective at identifying potential neurotoxins.4 Although neurotoxicity is apparent when intense highlevel exposures result in acute illness, there is increasing evidence linking chronic low-level exposures with neurodegenerative diseases.5 Given the frequency with which potentially neurotoxic exposures occur, clinicians may be missing these causal factors in many patients presenting with central neurologic symptoms. The effects of toxins on the nervous system are protean. The nervous system is uniquely susceptible to intoxication for several reasons (Table 28.1.1). When the peripheral nervous system is affected, characteristic symptoms and signs develop (see Chapter 28.2). In these cases, much can be done to specifiy the degree and type of dysfunction, including a histologic examination of a peripheral nerve biopsy specimen. On the other hand, the effects of toxins on the CNS – the brain and spinal cord – are more complex than on the peripheral nervous system. Clinical disorders of the CNS are variable in their presentation and difficult to classify, often involving a host of non-specific symptoms. Typically, the brain and spinal cord cannot undergo biopsy for histologic examination. Thus, clinicians often focus on describing neurologic function and excluding other neurologic conditions, rather than on diagnosing a specific neurotoxic disease. This chapter will review the challenges in diagnosis, organize an approach to patients with CNS dysfunction as the result of a neurotoxic exposure, and present a scheme for classifying diseases related to neurotoxins.

CLINICAL EVALUATION In order to facilitate an accurate diagnosis, a clinical evaluation must include a detailed account of the patient’s

initial presentation. The most important elements of this evaluation are a comprehensive history and detailed neurologic examination, which together yield the most relevant diagnostic information. Because neurotoxic diseases may be evanescent, documentation of a thorough baseline neurologic examination is crucial, and close follow-up is important. The differential diagnosis can be further refined by review of relevant exposure documents, such as Material Safety Data Sheets (MSDS) and duty logs, as well as literature searches for reported associations. Consultation with a neurologist may be needed to define subtle features of the neurologic examination and help decide what ancillary testing is appropriate to clarify the disease process and exclude other causes for the patient’s problems.

History The first challenge facing the clinician is to determine if the patient’s condition is related to a neurotoxic exposure; the cornerstone for this evaluation is a detailed history. The manner in which a patient with a potential neurotoxic exposure presents for medical evaluation is related in part to the level of the exposure and potency of the toxin. The most straightforward presentation is when the patient is aware of a specific chemical exposure. The person may have learned that exposure to some substance is potentially hazardous and is worried about possible injury. The concern is augmented when the organ targeted for injury is the brain. In this situation, determining the level of exposure represents the challenge. At one extreme, the exposure may be extremely small, and the risk of detectable disease equally small; for example, the patient who is otherwise healthy and is concerned because of an exposure to aluminum while using aluminum cookware and soft-drink cans. For such cases, a history and physical examination are generally adequate to convince the clinician whether a serious problem exists. At the other extreme, a patient may present with obvious dysfunction of the CNS related to a massive neurotoxic exposure. Many times, such large exposures are accidental, but other times, they are intentional, as occurs with homicide, suicide, and substance abuse. Examples would include the accidental exposure of a worker to carbon monoxide, the use of cyanide for homicide, and the sniffing of glue in search of euphoria. When the patient has symptoms but does not immediately associate them with some exposure, the clinician must question the patient about environmental exposures,

646 Central Nervous System Diseases • Neurons and their processes have a high surface area, increasing their effective exposure to chemicals • The high lipid content of neuronal structures results in accumulation and retention of lipophilic chemicals • Metabolic demands are high, so neurons are strongly affected by energy or nutrient depletion • High blood flow, for high metabolic demands, increases the effective exposure to circulating chemicals • Chemical toxins interfere with normal neurotransmission by mimicking the structures of endogenous molecules • Following chemical injury, recovery of the normal, complex interneuronal connections is imperfect • Neurons cannot regenerate once killed by chemical exposures Table 28.1.1 Factors contributing to the susceptibility of the nervous system to chemical injury

particularly those occurring at work. Questions to identify neurotoxic exposures should include a detailed occupational and environmental health history (see Chapter 3). An exposure may have resulted in a disease that has forced the patient to retire; in this case, the key will lie in questioning the patient or the patient’s family about exposures that have occurred in the past. It may also be necessary to explore past exposures in cases of chronic neurodegenerative diseases, for which the clinician may not have a ready explanation and may therefore erroneously attribute to aging alone. If a worker has neurologic symptoms that he or she believes are related to a particular neurotoxin, then the clinician needs to document the symptoms and determine whether they are consistent with the known effects of the neurotoxin at that level of exposure. For example, a worker spills an insecticide, such as malathion, and develops transient confusion, agitation, flushing, blurred vision, dry skin, and dry mouth. Here, the neurotoxic effects of an organophosphate resulting in cholinergic excess are readily apparent (see Chapter 48). Unfortunately, most clinical situations are not as straightforward. If after consulting with appropriate sources there is no apparent link between the symptoms and exposure, then the condition is either not related to the exposure or represents a newly recognized manifestation of a neurotoxin. In such cases, principles of causation must be applied (see Chapter 1). The pattern of associated symptoms is useful in recognizing and identifying a particular neurotoxic syndrome, and certain neurologic complaints should alert a clinician to explore the occupational health history thoroughly (Table 28.1.2). Higher cortical functions can be extremely sensitive to various brain insults, including neurotoxins. Complaints of changes in cognition, behavior, and mood should always prompt questions about exposure to neurotoxins. In addition, many parts of the nervous system

• • • • •

Demonstrable exposure Temporal connection Cognitive or behavioral changes Incoordination Non-focal dysfunction

Table 28.1.2 Pattern recognition for neurotoxicity

must be operating normally for coordination to be intact, so complaints about incoordination should also raise suspicion about neurotoxins. Such complaints can take many forms, including dizziness, unsteadiness of gait, or difficulty with fine finger movements. The patient is usually aware of problems with coordination, although they may be difficult to characterize. The patient may not necessarily be as aware of problems with cognition. These complaints may come from friends and family members instead of the patient. A key element in recognizing a pattern consistent with neurotoxicity is the distinction between focal and more diffuse involvement of the CNS. Typically, evidence of focal involvement weighs against the problem being related to a neurotoxin. In this context, focal changes need to be distinguished from system-specific involvement. A focal lesion generally refers to a specific site of injury, such as a tumor or stroke, which is anatomically localized but has a broader functional impact. Thus, the patient with temporary hemiparesis and aphasia may have had an ischemic injury in the middle cerebral artery distribution due to vascular disease, rather than a neurotoxic cause. In contrast, system-specific injury often reflects that system’s inherent metabolic or physiologic susceptibility to intoxication rather than its anatomic localization. A recent example that has revolutionized thinking about neurotoxins involves the effects of a meperidine derivative, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), that selectively damages the dopaminergic motor system of the brain. The resulting clinical picture is indistinguishable from Parkinson’s disease. This system-specific neurotoxin will be discussed in more detail later in this chapter. Another element to be considered is the temporal sequence of events. Key distinctions include acute versus insidious onset, constant versus intermittent dysfunction, and progression versus stabilization or improvement. A consistent temporal association between symptoms and exposures is an important aspect in assessing a likely toxic etiology. Little doubt exists when acute symptoms are associated with accidental overexposures, as in organophosphate poisoning. Intermittent complaints may be associated with non-continuous exposures, such as with a rotating duty roster. Perhaps the most challenging time course is one of insidious onset with progressive deterioration. In this case, other clues may be useful, such as anatomic or functional localization. For example, a progressive hemiparesis may suggest a focal neoplasm, whereas progressive gait instability may reflect a system-specific cerebellar effect of chronic solvent toxicity. By combining facts about the exposure scenario with the symptom complex and temporal sequence, a pattern of CNS dysfunction consistent with neurotoxicity may emerge. Recognition of this pattern is essential to diagnosis. The information gleaned from the initial history further guides the clinician in the subsequent physical examination and later selection of ancillary tests.

Clinical Evaluation 647 Test areas Orientation What is the (year) (season) (date) (day) (month)? Where are we: (state) (county) (town) (hospital) (floor)? Registration Name three objects: 1 second to say each. Then ask the patient all three after you have said them. Give 1 point for each correct answer. Then repeat them until he or she learns all three. Attention and calculation Serial 7s. 1 point for each correct. Stop after five answers. Alternatively spell ‘world’ backward. Recall Ask for the three objects repeated above. Give 1 point for each correct response. Language Name a pencil and watch. Repeat the following ‘No ifs, ands, or buts.’ Follow a three-stage command: ‘Take a paper in your right hand, fold it in half, and put it on the floor.’ Read and obey the following: Close your eyes. Score 1 point only if he or she actually closes his or her eyes. Write a sentence. Do not dictate a sentence; it is to be written spontaneously. It must contain a subject and verb and be sensible. Correct grammar and punctuation are not necessary. Copy design. On a clean piece of paper, draw intersecting pentagons, each side about 1 inch, and ask the patient to copy it exactly as it is. All 10 angles must be present and 2 must intersect to score 1 point. Tremor and rotation are ignored.

Maximum score

Score

5 5

________ ________

3

________

5

________

5

________

2 1 3

________ ________ ________

1

________

1

________

1

________

* Modified from Folstein MF, Folstein SE, McHugh PR. “Mini-Mental State.” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12:189–198. Copyright 1975, with kind permission from Pergamon Press Ltd, Headington Hill Hall, Oxford OX3 OBW, UK; and Bleecker ML, Bolla-Wilson K, Kawas G, Agnew J. Age-specific norms for the Mini-Mental State Exam. Neurology 1988; 38:1565–8.

Table 28.1.3 Mini Mental State Examination*

Physical examination By the time the history is completed, the clinician should have strong suspicions about what the neurologic examination will show. The examination thus provides an opportunity to test hypotheses generated from the history by either confirming expected findings or eliminating alternative explanations. Although the general medical examination may supply clues that the patient has been exposed to a neurotoxin – for instance, gingivitis can occur with chronic mercury poisoning and transverse white lines in the fingernails (Mees’ lines) can be seen with arsenic poisoning – such pathognomonic findings are often not present. Therefore, the neurologic examination is essential to document the localization, the type, and the degree of neurologic impairment. There are six broad areas that should be screened in a routine neurologic examination: • mental status; • cranial nerves; • motor system; • reflexes; • sensation; and • coordination and gait. A structured approach to these six areas helps ensure completeness and reproducibility.6 Testing the mental status and higher cortical functions begins with an assessment of the form and content of communication with the patient during the clinical interview. More formal, structured neuropsychologic screens provide advantages of reproducibility and reliability.7 One

commonly used screen that can be easily performed in the clinic is the Mini Mental State Examination (MMSE; Table 28.1.3). Age-adjusted norms have been established, with 29 or 30 being a normal score for persons under age 65 years.8 Although a normal score cannot rule out subtle cognitive problems, an abnormal score is concerning. Bedside neuropsychologic testing is insensitive to subtle abnormalities that the patient, family members, or friends may readily recognize. In these situations, more detailed neuropsychologic testing is indicated. Certain abnormalities of higher brain function, such as aphasias or apraxias, may suggest focal cortical involvement. The finding of such focal signs on detailed testing makes it unlikely that neurotoxicity is the sole explanation for the patient’s complaints. The 12 cranial nerves should be screened to search for evidence as much to refute as to support the hypothesis that a patient’s complaints are related to a neurotoxin. For instance, the finding of a bilateral intranuclear ophthalmoplegia would push the clinician strongly toward a diagnosis of multiple sclerosis. Multiple sclerosis is frequently a consideration in patients suspected of having a neurotoxin-related disease, given the various symptoms and signs with which multiple sclerosis can present and that it affects young men and women who may try to relate their symptoms to something in the workplace. The first cranial nerve serves olfaction. While not easy to test in a reliable fashion, impairment of olfaction as an early finding is noteworthy. An area of controversy has been the hypothesis proposed by some investigators

648 Central Nervous System Diseases that certain neurotoxins gain access to the brain via the olfactory system, thence spreading to the neighboring hippocampal systems, affecting memory, or the basal ganglia, affecting motor function. Olfactory impairment is a common finding in patients with neurodegenerative diseases,9,10 as well as chronic rhinosinusitis. Various substances can be used to test olfaction, and kits are available for formal assessment.11 The second cranial nerve, serving vision, is susceptible to neurotoxic injury. Bilateral findings can suggest a diffuse neurotoxic process, and bilateral optic neuropathy has been described with exposures to heavy metals, solvents, and insecticides. However, unilateral dysfunction suggests other disease processes. Physical findings include diminished visual acuity, a relative afferent pupillary defect to a swinging light, and a pale optic disc. Testing of the visual field is an important screen for focal brain disease, and therefore most visual field defects suggest an alternative cause for the patient’s complaints. Bilateral field defects can rarely result from toxins. Blindness with unreactive pupils points to a problem anterior to the optic chiasm, as can occur with methanol intoxication. On the other hand, preservation of the pupillary light response indicates that the problem is in the brain, not in the eyes, termed cortical blindness. Poisoning with organic mercury can produce this syndrome, with constricted visual fields and eventual loss of vision. Neuropathologic examination in such patients confirms damage to the cerebral cortex serving vision. Several aspects of the second cranial nerve can be tested. Visual acuity, visual fields, pupillary light response, and fundoscopy should be checked (see Chapter 20.1). Tests of color vision can be easily administered in the clinical setting and serve as sensitive screens of optic nerve function. For this reason, they have also been advocated for inclusion as a standard element in epidemiologic field studies.12 Examination of extraocular movements – cranial nerves III, IV, and VI – can give evidence to support a neurotoxic exposure. Certain types of gaze-evoked nystagmus can be seen with acute intoxications affecting the cerebellar system. Diplopia resulting from extraocular muscle paresis can occur with a number of neurotoxins, but this typically reflects dysfunction of the peripheral, not the central nervous system. Many degenerative diseases of the brain, including those related to toxins, can be associated with slowing of the saccadic, or normally lightning-like, eye movements. The remaining cranial nerves are less frequently affected by neurotoxins acting on the brain. The fifth and seventh cranial nerves serve facial sensation and strength. The eighth cranial nerve serves hearing and balance. The ninth and tenth cranial nerves serve the function of swallowing. These cranial nerves are more frequently affected in their peripheral portion by neurotoxins; CNS involvement typically suggests a vascular, neoplastic, inflammatory, or demyelinating etiology. The cranial nerves serving motor function, including V, VII, XI, and XII, can be involved in processes such as

Atrophy Tone Reflexes Plantar response Fasciculations

Upper motor neuron

Lower motor neuron

Moderate Increased Increased Extensor Absent

Marked Decreased Decreased Flexor Present

Table 28.1.4 Localizing weakness based on examination

amyotrophic lateral sclerosis, that affect alpha motor neurons in the anterior horn of the spinal cord. Some associations between motor neuron disease and environmental exposures, including exposures to specific solvents and heavy metals, have been observed.13,14 The tongue, served by cranial nerve XII, can be a good place to detect the resulting weakness and atrophy. Fasciculations can also be seen, though they can be difficult to distinguish from other movements. The task is made easier by examining the tongue while it is resting in the mouth, rather than protruded. For unclear reasons, muscles that serve eye movements are typically spared. Examination of the motor system should include assessments of the strength, bulk, and tone of the muscles. All three elements must be considered, because although weakness suggests a problem in the motor system, it may reflect dysfunction of the muscle itself, the neuromuscular junction, the peripheral nerve, the anterior horn cell, or the so-called upper motor neuron. Localization within this pathway strongly influences diagnostic considerations and the ordering of ancillary tests. Although the lower (alpha) motor neuron is influenced by many higher centers and descending systems, the simplified model distinguishing between upper and lower motor neuron lesions is useful clinically. Findings on the examination often suggest localization to the upper or lower motor neuron (Table 28.1.4). Both categories involve weakness. Diminished bulk, or atrophy, can also occur with either type of lesion, although it is usually more marked with a lower motor neuron lesion. In addition to reflex abnormalities (discussed later), the most useful feature in separating the two categories is a change in muscle tone. Tone is best assessed with passive movements of the limb, usually the upper extremity. It is important for the patient to be fully co-operative to avoid a superimposed component of volitional tone, which can confound interpretation. Lower motor neuron lesions, involving the motor system from the anterior horn cell to the muscle proper, produce decreased muscle tone, or hypotonia. Upper motor neuron lesions, involving higher centers and descending systems that modulate activity of the anterior horn cell, produce increased tone, or hypertonia. Two exceptions are those related to cerebellar dysfunction and to certain lesions of the basal ganglia, which can produce hypotonia. The increased tone seen with involvement of the descending motor tracts is called spasticity. It is characterized by a resistance to passive movement, which is variable throughout the range of movement and depends on the speed of movement. Unlike spasticity, rigidity is

Clinical Evaluation 649 characterized by increased tone that does not vary throughout the range of movement or with the speed of movement. Rigidity usually results from basal ganglion dysfunction. Spasticity may reflect a toxin-related dysfunction of the descending tracts that directly influences the lower motor neuron, producing a clinical myelopathy. These tracts contain some of the longest axons in the nervous system, and are thus vulnerable to some of the same agents that cause length-dependent peripheral axonopathy, such as tri-ortho-cresyl-phosphate.15,16 Nitrous oxide is another agent that can damage central tracts, giving a clinical picture similar to that seen with vitamin B12 deficiency,17 though this typically occurs only after repeated exposures, such as with substance abuse. Early after an exposure to such toxins, the signs and symptoms related to the peripheral nervous system injury may predominate, while only later does the injury to the CNS become evident. Rigidity, reflecting dysfunction in central motor systems that do not directly influence the lower motor neuron, can also be related to neurotoxins. Rigidity is one of the cardinal manifestations of parkinsonism, reflecting dysfunction of the basal ganglion. The other features of this syndrome are bradykinesia with a paucity and slowness of movements, resting tremor with a typical frequency of 3–7 Hz, and postural instability with the potential for frequent falls. The diagnosis of parkinsonism is based on clinical assessment; ancillary tests are not helpful. Several neurotoxins can damage the neural inputs or outputs of the basal ganglion and result in this syndrome. Such selective neurotoxins include carbon monoxide, carbon disulfide, manganese, rotenone, and MPTP. The marked similarities between toxin-induced movement disorders and idiopathic Parkinson’s disease have raised concerns that the two conditions may be related,5,18 as discussed in more detail later. Rather than manifesting with rigidity and bradykinesia, injury to certain regions of the basal ganglion can manifest with the opposite picture: reduced tone and hyperkinesias. The hyperkinesias can manifest as a variety of involuntary movements involving the face, trunk, and limbs. Careful observation during the history and physical examination is usually enough to suspect such a movement disorder. The causes of such disorders are numerous, including genetic diseases such as Huntington’s or Wilson’s disease. Neurotoxic exposures can also produce this picture; for example, as an acute effect of a drug such as cocaine, or as a delayed effect of a drug such as haloperidol.19–21 Tendon reflexes provide a quick screen, reflecting balance between the function of the upper and lower motor neuron. Increased tendon reflexes, especially in the setting of other findings of upper motor neuron dysfunction, strongly suggest a problem in the CNS. Initially reduced or absent tendon reflexes cannot rule out CNS involvement, because the injury to the peripheral nervous system that produces hyporeflexia may initially mask the manifestations of the concomitant CNS injury, as in the case of tri-ortho-cresyl-phosphate poisoning. Examination of the tendon reflexes can also suggest an injury to the

cerebellum. In these situations, pendular reflexes are present: the limb keeps swinging longer than usual following elicitation of the reflex. A reflex that gives strong evidence in favor of an upper motor neuron lesion is the extensor plantar response or the so-called Babinski sign. Examination of sensation can help localize dysfunction to the peripheral as opposed to the central nervous system. Screening involves checking the patient’s sensitivity to several modalities in several locations, including vibration in the distal lower extremities. Most patients with a normal sensory system can appreciate vibration in the great toe from a vigorously thumped 128-Hz frequency tuning fork for more than 10 seconds. Typically, findings on the sensory examination that point to a problem in the CNS do not suggest a neurotoxic lesion. Thus, the dissociated sensory loss of a spinal cord syrinx or the crossed sensory changes of a brainstem stroke exclude an etiology related to a neurotoxin. In patients in whom the sensory examination shows a stocking–glove type of reduced sensation, a diffuse peripheral polyneuropathy should be suspected, which can be metabolic or toxin induced. The presence of incoordination on the physical examination is non-specific. For coordination to be normal, many of the aspects of the motor and sensory function already discussed must be intact. Incoordination could indicate a problem in the CNS, the peripheral nervous system, or both. It could indicate a problem in the motor systems, sensory systems, or both. Incoordination can result from either acute or chronic intoxication. Despite its lack of specificity, incoordination is a sensitive screen for evidence of neurotoxic effects on the nervous system, because coordination testing reflects broadly the function of the nervous system. Evaluation of coordination should include both the upper and lower extremities. Tasks should be carefully observed for speed, smoothness, and agility. For the upper extremities, the patient should move quickly from one target to another, e.g., finger to nose. For the lower extremities, the heel should be moved up and down the shin of the opposite leg. Rapid alternating movements, such as finger and toe tapping, or clapping first the front and then the back of one hand on another, should be evaluated. Gait should be observed first with the patient walking naturally and then with increasing difficulty by having the patient walk on heels and toes, tandem walk forward and in reverse, and in a semi-squat while holding a tandem position. Though computerized posturography has progressed as a tool for clinical assessment,22,23 incoordination is not easily quantified, and the decision about its presence or absence ultimately relies on clinical judgment.

Ancillary testing Ancillary tests of CNS function can serve several purposes in the evaluation of patients suspected of having a neurotoxin-related disease. Most importantly, these tests can exclude other diseases. For example, imaging studies may

650 Central Nervous System Diseases show focal abnormalities such as neoplasm or stroke that account for a patient’s complaints. Tests can be used to help clarify uncertainties in the clinical presentation by demonstrating objective abnormalities underlying the patient’s subjective descriptions. For example, a substantially elevated carboxyhemoglobin level may explain vague complaints of headache and decreased concentration resulting from occult carbon monoxide exposure. Ancillary tests can also be used as more sensitive screens for the effects of neurotoxins on the CNS than the routine history and physical examination. This benefit is particularly apparent in the evaluation of cognitive function, which can be rigorously assessed with neuropsychologic testing, to follow up screening findings on the MMSE. Finally, ancillary testing can be used to quantify neurologic deficits. Such measurements may be used to gauge the degree of impairment for compensation purposes, to assist with vocational counseling by providing a guide for what a patient may be able to accomplish with such deficits, and to provide a baseline from which to judge the clinical course of the illness or response to treatment. Ancillary tests will be considered in a few broad groups as follows: • anatomic studies, such as computed tomographic (CT) scanning and magnetic resonance imaging (MRI) of the head; • physiologic tests, such as elicitation and averaging of sensory-evoked potentials; • functional assessments, such as neuropsychologic testing; • bioassays, such as cholinesterase levels; and • miscellaneous tests, such as positron emission tomography (PET).

Anatomic studies Most neurotoxins do not produce injuries that can be detected by imaging studies that detail the anatomy of the brain and spinal cord. An exception is the group of toxins that can injure the basal ganglion and may result in bilateral abnormalities in the globus pallidus. In most circumstances, the clinician will use imaging studies to exclude other, non-neurotoxic diseases. CT scanning is an excellent screen for mass lesions, such as brain tumors, which will usually lead to symptoms or signs indicating a focal dysfunction of the brain. If the history and physical examination lack such findings, then the yield from CT scanning is low. As a screening test, MRI has proven to be more sensitive than CT scanning for most conditions.24 MRI supplies more anatomic detail than does CT scanning: contrast agents can increase the sensitivity of both techniques. One situation in which MRI is particularly good is in detecting abnormalities of the white matter, so-called leukoencephalopathy.25 Multiple sclerosis has many clinical presentations, and although most suggest focal or multifocal involvement of the brain or spinal cord, some could be confused with neurotoxin-related disease. Confusion is most apt to occur in distinguishing the chronic progressive forms of multiple sclerosis from a disease caused by a neurotoxin. In such cases, the presence of characteristic periventricular white matter hyperintensities may support

the diagnosis of multiple sclerosis. It is imperative that the medical history correlates with the imaging results, because in many cases the MRI may also be non-specific, showing abnormalities that are of uncertain significance. For instance, the finding of unidentified bright objects (UBOs) is common in otherwise normal people. Toxins are often suspected in the differential diagnosis of cognitive decline or dementia. Imaging may be used to evaluate such patients, though it most commonly does not yield a specific diagnosis and cannot help distinguish whether or not dementia is related to a neurotoxin. Cerebral cortical atrophy, often demonstrated in such cases, does not correlate well with cognitive performance.26 Atrophy is a non-specific finding that may look the same in Alzheimer’s disease, Huntington’s disease, or a solvent-related dementia. One finding on imaging studies that suggests a neurotoxic etiology is bilateral and symmetric damage to the basal ganglion, most commonly involving the globus pallidus. These findings are of particular interest when the patient has clinical evidence of parkinsonism. Carbon monoxide, carbon disulfide, and manganese are examples of agents that can produce these lesions. The case of a 27-year-old man found unconscious in a parked recreational vehicle is illustrative. He had been using a small charcoal burner to keep warm and had suffered severe, acute carbon monoxide poisoning. The patient eventually regained consciousness but had parkinsonism, with bradykinesia, rigidity, and resting tremor. MRI showed bilateral symmetric hyperintensities in the globus pallidus, typical of what can be seen in such cases (Fig. 28.1.1). Although CT scanning can demonstrate the bilateral abnormalities of the globus pallidus, MRI is the preferred method in these circumstances. However, there are some caveats to interpretation. First, the finding of bilateral basal ganglia abnormalities on imaging is not specific to a neurotoxic etiology: Wilson’s disease and certain inherited mitochondrial disorders can have a similar appearance. Next, if the neurotoxin causes parkinsonism by selective damage to the substantia nigra, as occurs with MPTP, in contrast to widespread damage to the basal ganglia, such as with carbon monoxide, then the imaging may be unremarkable. Finally, although MRI abnormalities may be seen early in the course of disease, the findings may normalize over time, even when symptoms persist. To summarize, the clinician suspecting that a patient whose history or physical examination findings suggest neurotoxic involvement of the brain or spinal cord should have low expectations about the imaging results. Nonetheless, given the broad differential diagnosis, which includes diseases like multiple sclerosis whose presentations can be non-specific, imaging is often necessary. Screening should be done with the most sensitive test, which is MRI. The clinician must be cautious not to overinterpret non-specific findings.

Physiologic tests Physiologic dysfunction of the CNS may not be manifest by imaging tests. Tests that reflect physiologic rather than

Clinical Evaluation 651

A

B

Figure 28.1.1: Magnetic resonance imaging scan of a 27-year-old man with parkinsonism related to carbon monoxide poisoning. Axial (A) and coronal (B) spin density images of the brain, including basal ganglia, are shown. Hyperintensities are present in the lentiform, or lens-shaped, nucleus of the basal ganglion. The lentiform nucleus has two parts: (1) the more medial globus pallidus and (2) the more lateral putamen. Note involvement of the globus pallidus, with relative sparing of the putamen.

anatomic changes are useful in the evaluation of many diseases related to a neurotoxic exposure. The electroencephalogram (EEG) is used to record the electrical activity of the brain. Many things can affect the EEG tracing, so changes rarely point to a specific cause. The EEG is also difficult to quantify, and subtle intoxication can lead to EEG changes that can be difficult to interpret. Spectral analysis techniques have been developed,27 but although they have been used in epidemiologic studies, there remain considerable barriers to their effective application in most clinical situations.28 Many toxins that affect brain function, such as organic solvents and heavy metals, will affect the appearance of the EEG diffusely. The changes that are most obvious on the EEG, such as diffuse slowing, are often associated with evidence of toxic encephalopathy on the neurologic examination, with alteration of cognition, coordination, or both. These diffuse findings can be contrasted with EEG evidence of focal brain dysfunction, which is typically not related to neurotoxins. A common application of EEG testing is in the evaluation of seizures. Epilepsy can be diagnosed when the EEG tracing shows characteristic epileptiform discharges in the appropriate clinical setting. Generalized convulsions can be a manifestation of a neurotoxic exposure, such as to an organochlorine insecticide. However, the convulsions themselves do not necessarily indicate a diagnosis of epilepsy, which is defined by the recurrence of unprovoked

seizures. Unless the EEG is done around the time of a toxin-induced seizure, the tracing should be free of epileptiform activity. Whereas quantitative information is difficult to obtain from a routine EEG, it is more readily obtained from evoked potentials. In these tests, a sensory stimulus is used to trigger a neural response. By averaging the response to many discrete stimuli, a time-linked sensory-evoked response can be detected. Although evoked potentials can provide quantitative information, many variables can confound interpretation and results are not specific. As always, the entire clinical scenario must be considered. Nonetheless, event-related potentials have shown great promise as sensitive indicators of brain dysfunction in population studies.22 Functional information about the visual pathways can be obtained from visual-evoked potentials. Providing retinal function is normal, the results can provide an indication of dysfunction in the optic nerves. Although such optic nerve dysfunction could result from a neurotoxin, it may also result from a vitamin deficiency, a tumor, or a demyelinating disease. As in other parts of the nervous system, asymmetric involvement tends to favor etiologies other than those related to neurotoxins. The brainstem auditory-evoked response (BAER) screens auditory pathways that are both peripheral and central, from the cochlea of the inner ear to the medial geniculate nucleus of the diencephalon. The most common

652 Central Nervous System Diseases abnormalities related to toxic exposures involve injuries to the peripheral portions of this system. Somatosensory-evoked potentials can be elicited by stimulation of peripheral nerves in the upper and lower extremities. Like the BAER, this test indicates the function of not only the peripheral, but also the centrally projecting sensory pathways. If the peripheral system is capable of carrying a signal, then the coexistence of central and peripheral involvement, as with tri-ortho-cresyl-phosphate poisoning, can be demonstrated with this test.

Functional assessments The ultimate impact of neurologic disease is in terms of functional limitations. The neurologic examination itself is purposely organized into functional components. These basic maneuvers have been extended through a variety of functional assessments developed to recognize, characterize, and quantify functional deficits. In certain cases, tests of sensory threshold and posturography may be useful, though a detailed examination is generally adequate to detect clinically significant functional limitations. By far the most important for evaluating global cerebral changes within the CNS are assessments of neuropsychologic functioning. Neuropsychologic testing is the best method to assess subtle abnormalities of cognitive function. The spectrum of neuropsychologic tests is great, and selection must be tailored to the particular situation. For instance, the tests administered to an individual with complaints of chronic memory problems may differ substantially from the tests used to screen a large number of workers in an epidemiologic study or from that used to monitor acutely exposed workers. When screening large numbers of persons, the test battery must be short. Some tests are hand administered, such as the Neurobehavioral Core Test Battery of the World Health Organization;29 others are administered via a microcomputer, such as the Neurobehavioral Evaluation System.30 These screens can typically be administered in 40 to 90 minutes. Although such screens may perform well for assessing and comparing groups of exposed persons, more extensive testing (typically lasting many hours) is performed for clinical evaluation of an individual patient. Extensive neuropsychologic testing provides the most sensitive assessment of brain dysfunction. When dysfunction is identified, the testing can also indicate the degree of impairment, and where rehabilitation efforts should be focused. A baseline is also useful for comparison in subsequent testing: neurotoxic injury should stabilize following cessation of an exposure without further deterioration and with possible variable improvement, whereas problems related to neurodegenerative disorders, such as Alzheimer’s disease, may be expected to worsen. Thus, repeated testing over time may be indicated depending on the disease course. Many tests are available to evaluate the areas of memory, intelligence, executive function, personality, mood, and coordination. Two of the most widely used comprehensive sets of tests are the Halstead–Reitan Battery

and the Luria Nebraska Neuropsychological Battery. Although the latter has some advantages when dealing with patients who have minimal education or severe brain damage, the former may be a more sensitive screen for the subtle effects of a neurotoxin. Tests evaluating verbal memory, attention, visual tracking, and visual memory are particularly important for assessing neurotoxic injury. Many screens of personality and mood are available, but the most commonly used is the Minnesota Multiphasic Personality Inventory (MMPI). Several tests of coordination are available that assess motor speed, reaction time, dexterity, and balance. Some of the available tests are old and have well-established norms. Others, more recently developed, have the advantage of evaluating different functions in different ways, but are limited by a lack of established normative values. Neuropsychologic testing should be performed by or under the supervision of a qualified neuropsychologist. The interpretation of the tests must be in the context of the patient’s past education. Verbal performance is thought to be relatively resistant to the effect of neurotoxins on brain function, and it is often used as an estimate of pre-exposure function. Therefore, with an important neurotoxic exposure, evaluation of memory could elicit scores much lower than would be expected based on verbal performance. Psychologic factors can also affect test results, as can alcohol and drug abuse. Concurrent depression or anxiety can adversely affect test performance, and should be considered in the assessment of cognitive impairment. Such a situation could still be related to a workplace exposure as a secondary effect, if the exposure is the proximate cause of the psychologic condition.7,31

Bioassays Many clinical laboratory tests can be performed to evaluate the possibility that a patient has an illness related to a neurotoxin. The body’s burden of some neurotoxins, such as lead or other metals, can be estimated based on levels in the blood or urine. For other neurotoxins, indirect measures are available, such as red blood cell cholinesterase activity in cases of organophosphate exposure or carboxyhemoglobin level in cases of active carbon monoxide exposure. For most neurotoxins, however, direct or indirect monitoring is not readily available. Most other tests of the blood, urine, or spinal fluid are aimed at evaluating other diagnostic possibilities. Multiple sclerosis and spirochetal diseases, such as syphilis and Lyme disease, can mimic almost any disease of the CNS, including those related to neurotoxins. Thus, developing and refining bioassays is an area of ongoing research interest.32 Genetic screening represents the latest challenge in the application and interpretation of new bioassays. With the technologic innovations that have accompanied the human genome project have come an explosion of data regarding human genetic variability. Genetic polymorphisms are now thought to hold one of the keys to understanding differences between individuals in susceptibility to toxin-related diseases.33 Candidate genes include the

Classification of Diseases of the Central Nervous System Related to Neurotoxins 653 enzyme systems responsible for bioactivation and detoxification of xenobiotics and the receptor and transporter systems involved in neurotransmission.5 The sheer quantity of data and its rapid rate of acquisition have produced significant pragmatic challenges for data storage and manipulation as well as statistical assessment, spurring development of the related field of bioinformatics. One early conclusion from this work is that the vast majority of diseases are multifactorial, involving the combined effects of several processes.

Miscellaneous tests New technology is constantly evolving. Some applications, such as transcranial Doppler (TCD), will probably never be of much help in diagnosing neurotoxic diseases. Others, such as PET, single-photon emission computed tomography (SPECT), and functional magnetic resonance imaging (fMRI), may find a use in allowing direct assessment of the effects of specific neurotoxins on brain function. Still others, such as magnetic resonance spectroscopy34 and magnetic resonance microscopy,35 may help define the pharmacokinetic and pathologic effects of toxic exposures. Whether any of these new tests will have enough specificity to be useful for clinical diagnosis remains open to question.36 Currently these tests are primarily useful as research tools.

CLASSIFICATION OF DISEASES OF THE CENTRAL NERVOUS SYSTEM RELATED TO NEUROTOXINS Diffuse toxic encephalopathy • Acute • Chronic Selective toxic encephalopathy • Cell bodies • Ion channels • Neurotransmitter systems The classification scheme presented here primarily emphasizes the clinical manifestations of toxic exposures to the CNS. While it would be most desirable to develop a concise scheme based entirely on each individual toxin’s mechanism of action,37 the multiplicity of known mechanisms for selected toxins, coupled with a limited database for most others, makes such an approach clinically impractical.38 Although some selective toxins produce characteristic clinical syndromes, it remains more useful for diagnostic purposes to rely on a classification based on two features of the toxin’s effects: anatomic distribution and time course. The clinical manifestations of a toxin are related to which anatomic areas of the brain or spinal cord are compromised. Toxic encephalopathy is the phrase used to indicate a dysfunction of the brain caused by a toxic exposure. Toxic myelopathy is the phrase used to indicate toxin-related dysfunction of the spinal cord. For encephalopathies, the effects of toxins that are diffuse and

lead to non-specific clinical syndromes are distinguished from those that are selective and lead to specific clinical syndromes. The time course of neurotoxic injury often impacts on the reversibility of the effect. The nervous system is noted for its plasticity, or ability to compensate for subtle alterations of function. However, when tolerances are exceeded, more sustained or irreversible damage to the CNS may occur. This is very important, because under most circumstances, the brain and spinal cord lack the capacity to regenerate neurons. While it would be possible to draw distinctions between the acute and chronic forms of both diffuse and selective encephalopathies, these distinctions will only be discussed in detail in the context of diffuse encephalopathy. The selective encephalopathies are instead categorized by toxic effects involving clearly established cellular mechanisms.

Diffuse toxic encephalopathy–acute Acute diffuse toxic encephalopathy reflects a global cerebral dysfunction of rapid onset (typically days or weeks). It is a condition that many have experienced when they have felt the effects of an alcoholic beverage. There is nothing specific about these encephalopathies; virtually any organic solvent has the potential to produce such effects (see Chapter 40). The effects typically resolve completely, but if overwhelming can result in a non-specific chronic encephalopathy, as discussed in a later section.

Pathogenesis Toxins producing an acute encephalopathy interfere with basic cell functions in the brain, although the exact mechanisms remain obscure for some agents. Typically, such toxins gain easy access to the brain, and thus have the ability to alter brain function rapidly. Although active transport systems exist to convey many compounds into the CNS, most of these agents gain entry because they are highly lipid soluble and can readily diffuse across membranes. Besides organic solvents, which can alter cellular membrane function, several gases can also affect brain function diffusely, including the gas anesthetics, carbon monoxide, hydrogen sulfide, and cyanide. Heavy metals can also cause acute encephalopathies, but this is more commonly associated with the organic metals (e.g., methyl mercury or tetraethyl lead) than with the inorganic ones (e.g., lead, arsenic, mercury, or thallium).

Clinical and diagnostic approach The clinical manifestations of an acute diffuse toxic encephalopathy depend on the intensity of exposure. The symptoms and signs can range from mild euphoria with a normal examination, to stupor, coma, and death. The earliest manifestations are behavioral, with some alteration in mood typically reflecting disinhibition. Headache and seizures may also be relatively early manifestations. In general, the greater the exposure, the more severe the impairment of cerebral function, and suppression of level of consciousness. With moderate exposures, the physical examination shows evidence of the intoxication, such as

654 Central Nervous System Diseases lateral gaze-evoked nystagmus and incoordination, with sloppy fine finger movements and gait ataxia. The cerebral cortex is more sensitive to these toxins than the brainstem; typically, even when consciousness is lost, brainstem function remains intact. Thus, the pupillary light responses, corneal reflexes, oculocephalic reflex, gag reflex, and spontaneous respiration are preserved until late in the clinical course, unless some complication occurs. In the absence of asymmetric signs suggesting a focal brain injury, the combination of unconsciousness with intact brainstem function suggests a toxic or metabolic etiology. Aspiration with respiratory arrest or cardiotoxicity with hypotension may lead to further loss of function and death. Diagnosis does not generally present a challenge for acute syndromes, because the exposure and clinical manifestations are closely linked in time. An accurate history, if it can be obtained, will usually indicate the diagnosis. However, because toxic and metabolic encephalopathies can be indistinguishable, blood should be sent for electrolyte and glucose levels, renal function tests, liver function tests, and, when appropriate, endocrinologic screens. Arterial blood gases may also be indicated, both to evaluate acid–base disturbances and to assess hypoxia or hypercarbia in order to guide decisions regarding ventilatory support. If the patient presents at the time of maximal symptomatology, the toxin is most likely to be present systematically, with the greatest likelihood of detection. Because a detailed history is not always initially available, a high index of clinical suspicion should guide the selection of ancillary tests. Blood and urine toxicologic screens will detect most substances of abuse. Screens are more difficult for the types of exposures that would occur in the workplace. Most heavy metals can be detected, but most solvents cannot be detected using readily available tests. The presence of a solvent may be suggested by an elevated serum osmolality that is not explained by abnormal sodium, glucose, urea or ethanol levels, or by measuring metabolic levels. A carboxyhemoglobin level may be useful in detecting occult carbon monoxide, if the specimen is obtained at, or very soon following, the period of active exposure. Acute and convalescent cholinesterase activity levels are useful in documenting organophosphate exposure. While such confirmatory testing is desirable, a complete history is often adequate to establish a causal link between a toxic exposure and an acute clinical syndrome severe enough to require medical attention. The postictal confusional state, following a generalized seizure or ongoing partial complex status epilepticus, can present with symptoms such as delirium that might be confused with toxic encephalopathy. In such situations, an EEG may be useful in establishing the proper diagnosis. The most difficult situation to interpret is when a neurotoxin lowers the seizure threshold in an individual with an underlying susceptibility for seizures. Thus, for some individuals, provoked seizures occur only in the setting of exposure to a neurotoxin, though the toxin may not be the only causal factor. With a suggestive history and a physical examination that does not reveal focal brain dysfunction, further ancil-

lary testing is typically not needed. However, if questions exist about the history or physical examination, then imaging may be necessary. Other conditions can mimic a toxicmetabolic encephalopathy, including psychosis, meningitis, subarachnoid hemorrhage, head trauma, early central herniation syndrome, and multiple small cerebral emboli. For any of these conditions, brain imaging followed by sampling of the cerebrospinal fluid may be necessary.

Treatment and prognosis As with most overdoses, treatment of an acute diffuse toxic encephalopathy is primarily supportive. The patient should be physically removed from the source of exposure, including removal of contaminated clothing and thorough cleansing of exposed skin. Vital functions, such as respiration and circulation, are supported until the toxin is cleared. Because most toxins causing these conditions act diffusely on the brain, specific antidotes are typically not available. Specific treatments are indicated for certain exposures, such as ethanol for methanol intoxication, induced methemoglobinemia for cyanide or hydrogen sulfide exposure, chelation for heavy-metal toxicity and high-dose oxygen (including possible hyperbaric therapy) for acute carbon monoxide poisoning. For most of the toxins that act diffusely on the brain, recovery from a single or limited number of exposures is complete. The process of recovery is a reversal of the sequence described previously, with the level of consciousness gradually returning to normal. Withdrawal syndromes involving severe agitation and seizures are uncommon without repeated exposures. In severe cases the brain is damaged from a single exposure, when a chronic encephalopathy ensues. More often, the prognosis is a function of the cardiac and pulmonary decompensation that can complicate an acute diffuse toxic encephalopathy.

Specific agents The syndrome of acute diffuse toxic encephalopathy is not specific. It may be caused by any neurotoxin that can gain ready access to the CNS. Such agents include solvents, metals, and gases (see Table 28.1.5). As the acute Neurotoxin

Target

Clinical syndrome

Tetrodotoxin Organochlorines Organophosphate MPTP* Carbon monoxide Carbon disulfide Manganese 3-Nitropropionic acid Domoic acid BOAA† Organic mercury Nitrous oxide Fluoroethyl acetate Rotenone

Ion channel Ion channel Acetylcholinesterase Dopaminergic cells Globus pallidus Globus pallidus Globus pallidus Basal ganglion Glutaminergic system Glutaminergic system Occipital cortex Spinal cord Cerebellum Mitochondrion

Paralysis Paralysis Cholinergic excess Parkinsonism Parkinsonism Parkinsonism Parkinsonism Dystonia Amnesia Lathyrism Cortical blindness Spastic paraparesis Ataxia Parkinsonism

*1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine. †Beta-N-oxalylamino-L-alanine.

Table 28.1.5 Examples of selective neurotoxins

Classification of Diseases of the Central Nervous System Related to Neurotoxins 655 encephalopathy clears, evidence may emerge of a residual chronic encephalopathy, which is discussed in the next section.

Diffuse toxic encephalopathy–chronic Chronic diffuse toxic encephalopathy represents persistent injury to the brain, as a result of cumulative or multiple repeated exposures (often over a period of months or years) or, rarely, a single massive exposure to a toxic that causes severe acute diffuse encephalopathy. Evidence exists to implicate numerous toxins causing acute toxin encephalopathy, such as solvents and metals, in producing a chronic diffuse toxic encephalopathy. With repeated episodes of acute encephalopathy, often over many months or years, persistent impairment can occur.

Pathogenesis Although the mechanisms are not completely understood, continuous or repeated exposures can lead to an encephalopathic condition that can no longer be completely reversed, even though the toxin itself has cleared. This process is thought to reflect the fact that typical protective mechanisms have been damaged or overwhelmed. The resulting irreversible brain damage produces persistent neurologic dysfunction. It is likely that metabolic pathways for detoxifying and excreting exogenous toxins are involved.39 These systems are known to be in a dynamic balance, responding to external factors through complex interactions that may result in relative excesses of toxic metabolites. Extensive evidence has accumulated regarding the mechanisms of oxidative stress and apoptosis in neurodegeneration.40 These mechanisms may provide common final pathways for injury when activated by a variety of toxic exposures.

Clinical and diagnostic approach The clinical manifestations of chronic diffuse toxic encephalopathy usually involve varying degrees of cognitive impairment. The earliest problems are often subtle, and include changes in behavior and mood. Alternatively, a toxin may produce cognitive dysfunction to which the patient responds emotionally. The clinician may recognize the emotional change, rather than cognitive difficulties, and consider psychiatric referral. In either case, the organic nature of the complaints must be recognized. Some agents have been considered to alter mood more commonly. For example, inorganic mercury (historically used in hat manufacture) has been associated with mania, typified by the Mad Hatter in Alice in Wonderland, while carbon disulfide has been associated with depression. Generalizations regarding mood effects are probably not justified; psychosocial factors unique to the exposed individual are likely more important in determining that individual’s response. With progressive dysfunction, cognitive impairment becomes more readily apparent in the history and physical examination. Complaints and examples of declining skills include trouble balancing a checkbook, difficulty

remembering appointments, and getting lost in conversations or while traveling. Unfortunately, the condition can remain unrecognized or attributed to other causes until it has advanced to the stage of frank dementia. Much of the laboratory evaluation is devoted to excluding the rare, treatable causes of dementia. Exactly what constitutes an appropriate panel of screening tests has been debated.41 In the appropriate exposure scenario, a broader battery of testing is indicated. For example, evidence of exposure to heavy metals can be assessed directly by measuring levels in the blood or urine. Because these levels are more reflective of recent exposures, they may not adequately assess relevant past exposures or the body’s accumulation of the metal (see Chapter 39). When the index of suspicion is high, a chelation challenge can be performed to exclude more completely the possibility of previous exposures, such as with lead. Unfortunately, such measurements are variable, making it crucial that clinical correlation guide appropriate treatment.42 The differential diagnosis of presenile or senile dementia of the Alzheimer’s type should also be considered.

Treatment and prognosis Specific therapies for chronic diffuse toxic encephalopathies are limited, as would be predicted with diffuse destruction of brain tissue that generally lacks the capacity to regenerate. The patient should be separated as soon as possible from the neurotoxic exposure, and, in the case of certain metals, chelation therapy can be considered in an attempt to reduce the body’s accumulated burden of the toxin. Other agents that can further injure the brain should be strictly avoided, as it does not make sense to avoid one neurotoxin but to continue being exposed to others – e.g., alcohol use. Medications prescribed by physicians should be reviewed carefully to exclude those drugs whose effects might adversely impact cognitive function. The prognosis for a return of cognitive function is poor. Even in the absence of further exposure, cognitive function may continue to decline with aging. On the other hand, changes in behavior and mood may be more amenable to treatment; consultation with a psychiatrist may be helpful, in this regard. Anxiety or depression can aggravate cognitive impairment, and treatment of psychologic factors may result in significant improvement. Cognitive therapy and strategies to improve organizational skills may be appropriate, and should be supervised by a trained neuropsychologist.

Specific agents Similar to the acute diffuse toxic encephalopathy, clinical symptoms and signs of the chronic diffuse toxic encephalopathy are non-specific and are not associated exclusively with a particular exposure. Thus, the clinical appearance may be the same whether the exposure has been a solvent, such as toluene, or a metal, such as inorganic lead. Of note, such non-specific encephalopathies can also follow exposures to agents that have the potential to yield selective injuries with specific syndromes. For example, in some patients, exposure to carbon monoxide

656 Central Nervous System Diseases results in a chronic diffuse toxic encephalopathy instead of the selective injury to the basal ganglia, discussed in the next section.

Selective toxic encephalopathy Why some neurotoxins exert selective effects, injuring a particular part of the nervous system, remains uncertain in most cases. For instance, the globus pallidus is particularly sensitive to carbon monoxide, and the occipital cortex to mercury. For some of these toxins, selective effects reflect pathophysiologic actions through specific cellular mechanisms.38 The resulting clinical syndromes can reflect either reduced function within the injured neurotransmitter system or a relative imbalance between the functions of remaining neurotransmitter systems. Although it is likely that selective encephalopathies reflect both nervous system-related and exposure-related factors, the reasons for selective vulnerability of specific brain regions are largely unknown. One factor that may explain some of the variation in susceptibility is the timing of exposures relative to critical periods of nervous system development. The nervous system’s many discrete neuronal populations and interacting systems continue to develop at variable rates throughout the first three decades of life. Toxic exposures may exert their most profound effects when the organism is in a particularly vulnerable stage, leading to problems that would not occur with exposures at other stages of life.43 The most prominent example of this phenomenon is the susceptibility of infants to lead encephalopathy, which underscores the importance of thorough neurodevelopmental toxicology testing.4 Most of the toxins capable of producing an acute selective toxic encephalopathy act on neurons. While it is possible for any constituent of the CNS to be the target of a selective toxin, clinical examples of damage to the CNS’s glial cells, myelin, or vasculature are uncommon and will not be discussed here. Rather, the discussion will concentrate on the toxins directed at the neuron itself, considering those acting on the neuron cell body, ion channels, neurotransmitter systems, or structural integrity. As opposed to the toxins leading to the non-specific encephalopathies described in the previous section, many of these toxins produce characteristic syndromes. Examples of some selective neurotoxins (Table 28.1.5) are discussed in the following section.

Cell bodies Some neurologic diseases are characterized by pathologic findings involving loss of a particular population of neuronal cell bodies. For example, pigmented cells of the substantia nigra are lost in Parkinson’s disease, anterior horn cells in amyotrophic lateral sclerosis, and pyramidal cells of the hippocampal formation in Alzheimer’s disease. Although the initiating factors may not necessarily involve attack directly at the cell body, the loss of cell bodies is what distinguishes these diseases from others in which the pathophysiology primarily involves cellular dysfunction

rather than cellular death. The mechanisms of cell death activated in these diseases remain uncertain, but evidence implicating the related mechanisms of mitochondrial dysfunction, energy depletion, oxidative stress, and apoptosis continues to accumulate.40,44 Perhaps the most prominent example of a selective neurotoxin is MPTP. Research on the effects of MPTP has revitalized interest in the basic mechanisms activated by selective neurotoxins in producing neurologic disease. These findings have raised substantial concerns regarding the contribution of toxins to degenerative diseases of the CNS. The distinct possibility exists that devastating illnesses such as Parkinson’s disease, amyotrophic lateral sclerosis, Alzheimer’s disease, and many others, whose causes have been considered idiopathic, are actually related to environmental toxins. The strength of these arguments is largely related to what has been learned from the MPTP model of Parkinson’s disease. MPTP became infamous as a meperidine derivative that some intravenous drug abusers mistakenly injected in the hope of achieving narcotic effects.45 These individuals rapidly developed a condition virtually indistinguishable from idiopathic Parkinson’s disease. MPTP was later shown to produce similar effects when given to certain animal species, and using those models, the mechanism of action of MPTP has been identified.46 A key step involves its conversion by glial monoamine oxidase B to 1-methyl-4phenylpyridium ion (MPP+). Support for the concept of MPTP being a toxic precursor, or protoxin, is that inhibition of monoamine oxidase B can prevent the effects of MPTP. MPP+ has avid affinity for the dopamine transporter located at the terminal axons of dopaminergic neurons. It is thereby concentrated selectively in dopaminergic cells of the substantia nigra, activating a lethal biochemical cascade. Most intriguing from an occupational and environmental perspective is that MPP+ is a substance called cyperquat, used in the past as an herbicide and chemically related to paraquat, an herbicide that remains in common use. Even more compelling is a recent report describing the effects of rotenone, a common pesticide and mitochondrial complex I inhibitor.47 With chronic, systemic exposures to rotenone, rats developed a neurodegenerative disease with pathologic findings comparable to human Parkinson’s disease patients.48 Elucidation of the mechanism for neurotoxic injury related to MPTP raises the possibility that other neurotoxins could affect selective neuronal populations, resulting in clinical manifestations of neurodegenerative diseases. An apparent inconsistency is that these diseases usually manifest as progressive disorders in later life, whereas the most intense chemical exposures proposed to cause them probably occur over a limited period, at an earlier age. This inconsistency may be explained through the combined effects of aging and exposure to a selective toxin (Fig. 28.1.2). Later in life, neurons begin to die, presumably as a part of natural aging. Normally, the excess of neurons in the brain protects against the development of symptoms. However, the loss of neurons may be more important if a proportion of the available cells have been eliminated by an earlier exposure. In

Classification of Diseases of the Central Nervous System Related to Neurotoxins 657

Neurons surviving (%)

Exposure to neurotoxin

Seizures can also occur, especially with organochlorine pesticides. With severe poisoning, paralysis and respiratory arrest may intervene, resulting in death. Treatment is usually supportive, and, assuming no complication such as cardiopulmonary arrest has occurred, recovery may be complete.

Normal loss of cells with aging

100

1

2

Neurotransmitter systems

3

0 Age of patient

Threshold for symptomatic disease

Figure 28.1.2: Hypothetical loss of neurons with exposure to a selective neurotoxin. The loss may be so great that the threshold for symptomatic disease is crossed at the time of an acute exposure (option 1). Alternatively, not enough neurons may be destroyed to produce symptomatic disease acutely. But with further loss due to increased vulnerability (option 2) or to aging alone (option 3), the threshold for symptomatic disease may be crossed. (Modified from Langston JW. Predicting Parkinson’s disease. Neurology 1990; 40(Suppl 3):70-74.)

this case, the subsequent age-related loss of cells may lead to a point where a threshold is passed. Beyond this threshold, the loss of cells leads to clinically apparent disease that slowly progresses as the aging process continues. Alternatively, the toxin may have made the cells more vulnerable to the effects of aging, resulting in accelerated senescence. Such hypotheses have generated interest in many idiopathic degenerative diseases of the brain and spinal cord, as conditions that are potentially related to environmental toxins, and prompted a search for a multifactorial model to explain why some individuals are more susceptible to neurotoxic effects.

Ion channels Normal nervous system function depends intimately on membrane-associated ion channels. Consequently, brain function can be severely disrupted when ion channels are affected by a neurotoxin. Some of the most potent biologic toxins have ion channels as their target.49–52 Some of these channel blockers include tetrodotoxin from puffer fish and saxitoxin from shellfish contaminated by a red tide. Other toxins potentiate channel function by increasing either the frequency or duration of channel openings, thereby increasing ionic permeability. Examples include ciguatoxin from certain marine fish, and some insecticides, including organochlorines and pyrethrins. Biologic toxins are typically found in specific geographic areas, and although the clinical syndrome may suggest the diagnosis, these conditions are so rare that a high index of suspicion and thorough history are the keys to diagnosis. A greater threat is probably associated with insecticides, because of their widespread use. With any toxin that affects membrane channels, sensory symptoms around the mouth and face can be the earliest manifestations. With higher exposures, sensory symptoms become more widespread and motor signs of intoxication develop.

The number and diversity of known neurotransmitter systems have steadily increased. These systems provide a direct means by which a toxin can produce selective dysfunction of the CNS. Regulation of synaptic neurotransmission is complex, and any aspect can become the target of a toxin. Thus, a toxin can act to block or stimulate a postsynaptic receptor, block or be taken up by a synaptic transport system, or block transmitter degradation enzymes. Many of these effects occur with pharmaceutical agents commonly used in clinical practice. More relevant to the current discussion are the effects of naturally occurring biologic toxins and synthetic chemicals encountered in occupational settings and as environmental contaminants.

Cholinergic system

One of the most well-characterized neurotransmitter systems is the cholinergic system. Acute cholinergic dysfunction produces a set of distinct clinical syndromes, characterized by relative excess or deficiency of cholinergic activity. A number of toxins can produce excess activation of the cholinergic system. The hyperactivity can occur directly with toxins acting as receptor agonists, such as those found in certain species of mushrooms (e.g. Amanita muscaria, for which a whole class of cholinergic receptors is named). Alternatively, activation may occur indirectly with toxins that inhibit acetylcholinesterase and thereby increase the availability of acetylcholine at the synapse. This is the mechanism of the organophosphate pesticides and organophosphorus chemical warfare agents (see Chapter 48). Still other toxins can inhibit the cholinergic system by acting as receptor blockers, such as in the plant species used in mystical rituals (e.g. Datura stramonium). Regardless of whether the toxin gains access to the CNS, effects on the autonomic nervous system and neuromuscular junction are common and can be used to define the severity of the poisoning. The syndrome of cholinergic excess variably includes nausea, vomiting, diaphoresis, sialorrhea, lacrimation, abdominal cramps, diarrhea, bradycardia, miosis, and muscle fasciculations with weakness. The specific constellation of symptoms can vary between individuals. If the toxin gains access to the CNS, symptoms may include confusion, agitation, ataxia, tremor, and seizures. In some cases, the relative preponderance of peripheral symptoms can mask injury to the CNS. For example, the early course of poisoning with tri-ortho-cresyl-phosphate may be dominated by signs of lower motor neuron hypotonic weakness, later replaced by signs of upper motor neuron spasticity and incoordination. Given the characteristic symptoms and signs, the diagnosis of cholinergic excess should be suspected and

658 Central Nervous System Diseases empiric treatment initiated even in the absence of a detailed history. Atropine will block the muscarinic effects of excess acetylcholine but not the nicotinic effects. Certain oxime compounds, such as pralidoxime, can facilitate acetylcholinesterase reactivation, and should be used for poisoning involving prominent nicotinic symptoms. Although these treatments may reverse the acute manifestations of cholinergic excess, with severe poisoning some cognitive and behavioral symptoms may persist and new problems may arise at the neuromuscular junction. With toxins that produce cholinergic inhibition, hallucinatory effects are common, and for some agents are the motivation to consume the toxin. Many of the other CNS manifestations are similar to cholinergic excess, including confusion, agitation, and seizures. On the other hand, systemic symptoms are in many ways opposite to those described earlier, and include atropine-like symptoms of flushing, dry skin, tachycardia, and mydriasis. Cholinergic drugs can be administered, but usually the altered state will clear with supportive therapy alone.

Catecholaminergic systems and parkinsonism

The catecholaminergic neurotransmitter systems are common targets for selective agents. This makes them some of the most important systems for pharmaceutical drug actions. The systems can become hypoactive, as with dopamine receptor blockade by antipsychotic agents, or hyperactive, as when release of endogenous stores of norepinephrine is potentiated by amphetamines. Catecholamine transporters may also be targeted, as with serotonin selective reuptake inhibitors, providing another mechanism for increasing synaptic transmission. Although many other examples exist relevant to clinical pharmaceutics and drug dependence, or occupational diseases, the most pertinent effects relate to diminished dopaminergic activity in the basal ganglia resulting in a parkinsonian syndrome. The best understood toxin that produces parkinsonism is MPTP. As described earlier, its metabolite, MPP+, accesses neurons through a selective dopamine transporter. However, its ultimate effect is to decrease dopaminergic activity by killing cells in the substantia nigra. Other toxins that can lead to parkinsonism are carbon monoxide, carbon disulfide, and manganese.53 Although this occurs more commonly following chronic exposures, the syndrome has been reported following a single severe exposure. Often, a diffuse encephalopathy occurs initially, with the more selective injury to the basal ganglia recognized later. For example, this has been described in patients from China with acute poisoning from ingesting mildewed sugar cane, in which the toxic agent has been identified as 3-nitropropionic acid.54 While the severe poisoning initially produced a diffuse toxic encephalopathy, a fourth of survivors later demonstrated persistent dystonia, and MRI of these patients revealed hypodensities in the basal ganglia. In contrast to MPTP, the mechanism by which these agents induce parkinsonism involves a selective insult to the output cells of the basal ganglia located in the globus pallidus. Despite mechanistic differences, the clinical syndrome of parkinsonism is similar.

Regardless of the cause, a trial of dopaminergic drugs for bradykinesia and rigidity, or anticholinergic drugs for tremors, may be efficacious.

Glutamatergic system

Another neurotransmitter system that has been studied extensively in recent years is the excitatory glutamatergic system. Excessive activation of glutamate receptors results in toxic levels of calcium influx, thereby mediating so-called excitatory neurotoxicity.55 The complex intracellular events leading to cell death are not yet entirely clear, but the process may be involved in a variety of toxic, ischemic, and traumatic injuries to the CNS. This has motivated extensive research in the hope that modulation of such processes could provide therapeutic benefit. So far, though, clinical applications remain elusive. Several neurotoxins are now recognized to act through hyperactivity of the glutamatergic system. One clinical syndrome was first reported in a group of patients from Canada who became acutely ill following ingestion of contaminated mussels.56 The acute illness was characterized by headaches, seizures, hemiparesis, ophthalmoplegia, and alterations in the level of consciousness. In patients who recovered from the acute illness, chronic memory problems and peripheral polyneuropathy persisted. Neuropathologic examination in four patients who died demonstrated necrosis in the hippocampus, similar to that demonstrated in experimental animals given excitatory amino acid neurotransmitters. The offending agent has been identified as domoic acid, a potent excitotoxin. Lathyrism is another disease related to a neurotoxin acting through the glutamatergic system. In times of famine, excessive consumption of the grass pea, Lathyrus sativus and related species, has led to a neurologic condition characterized by spastic paraparesis. The offending agent in the pea is thought to be beta-N-oxalylaminoL-alanine (BOAA). Like domoic acid, BOAA is proposed to mimic an excitatory neurotransmitter,57 producing damage to the brain and spinal cord. Amyotrophic lateral sclerosis shares some clinical features of lathyrism, suggesting that it might result from exposure to a substance that could kill or injure anterior horn cells. The increased incidence of motor neuron disease that occurs on Guam has been a mystery for years, and consumption of the cycad nut has been proposed as a cause of the condition. The cycad nut has a BOAA-like constituent, beta-N-methylamino-L-alanine. Although many uncertainties remain, the idea that an environmental excitotoxin contributes to this and other neurodegenerative diseases has gained wide acceptance.58

Other systems Many other neurotransmitter systems exist, including those served by small molecules, such as gamma-aminobutyric acid (GABA) and adenosine; peptides, such as the endogenous opiates; and the recently recognized atypical transmitters, such as nitric oxide and Dlysine.59 The roles for these systems in normal neurologic function are incompletely understood. Thus, while tempting, it seems premature to speculate about the potential

Summary 659 role for these transmitters in disease processes. Nonetheless, many drugs and toxins are known to affect these systems, and many more undoubtedly exist that are not yet recognized. History suggests that as basic information about these systems continues to accumulate, evidence for their involvement in neurotoxic processes will be forthcoming.

10. 11. 12. 13.

SUMMARY Diseases affecting the CNS are challenging to diagnose and to treat, and those related to neurotoxins are no exception. The history and physical examination should allow the clinician to form and test hypotheses about the existence of neurotoxin-related disease. Problems of cognition and coordination should always prompt the clinician to question the patient about exposure to potential neurotoxins. Diagnostic hypotheses can be refined with judicious use of ancillary tests; however, these are frequently limited by lack of either sensitivity or specificity. Consultation with a neurologist may be useful in ruling out specific diagnoses. A classification scheme that may be useful to a clinician involves a distinction between diffuse and selective effects, characterized further by time course and specific mechanism of toxic effect, where known. This scheme will need to be revised as more is learned about the detailed mechanism of action of neurotoxins and the interactions of the toxin with host factors, including age and genetic susceptibility. As this knowledge increases, it is likely that more and more diseases will be linked to neurotoxic exposures. Some may be previously unrecognized diseases, but more promising are the well-recognized degenerative diseases for which the cause is currently unknown and for which environmental exposures may play a role. Identification of such exposures may allow the opportunity for prevention of these often disabling and sometimes fatal degenerative diseases.

References 1. US Government, EPA. Toxic Substances Control Act (TSCA). Chemical substances inventory – revised inventory synonym and preferred name file. Washington, DC: Office of Pollution, Prevention, and Toxics; 2000. 2. US Government, EPA. Health Effects Test Guidelines. OPPTS 870 Series. Washington, DC: Office of Prevention, Pesticides, and Toxic Substances (OPPTS); 1998. 3. US Government, FaDA. Neurotoxicity studies. In: Redbook 2000 – Toxicological Principles for the Safety of Food Ingredients. Washington, DC: Center for Food Safety & Applied Nutrition, Office of Premarket Approval; 2000. 4. Claudio L, Kwa WC, Russell AL, Wallinga D. Testing methods for developmental neurotoxicity of environmental chemicals. Toxicol Appl Pharmacol 2000; 164:1-14. 5. Feldman RG, Ratner MH. The pathogenesis of neurodegenerative disease: neurotoxic mechanisms of action and genetics. Curr Opin Neurol 1999; 12:725-31. 6. Swanson PD, ed. Signs and symptoms in neurology. Philadelphia, PA: JB Lippincott; 1984. 7. Feldman RG. Occupational and environmental neurotoxicology. Philadelphia, PA: Lippincott-Raven; 1999. 8. Bleecker ML, Bolla-Wilson K, Kawas C, Agnew J. Age-specific norms for the Mini-Mental State Exam. Neurology 1988; 38:1565-8. 9. Mesholam RI, Moberg PJ, Mahr RN, Doty RL. Olfaction in neurodegenerative disease: a meta-analysis of olfactory

14.

15. 16. 17.

18.

19. 20.

21.

22.

23.

24. 25. 26.

27. 28.

29.

30.

31. 32.

33. 34.

35.

36.

functioning in Alzheimer’s and Parkinson’s diseases. Arch Neurol 1998; 55:84-90. Wszolek ZK, Markopoulou K. Olfactory dysfunction in Parkinson’s disease. Clin Neurosci 1998; 5:94-101. Smith DV. Assessment of patients with taste and smell disorders. Acta Otolaryngol Suppl 1988; 458:129-33. Kulig BM. Comprehensive neurotoxicity assessment. Environ Health Perspect 1996; 104(Suppl 2):317-22. Olanow CW, Arendash GW. Metals and free radicals in neurodegeneration. Curr Opin Neurol 1994; 7:548-58. McGuire V, Longstreth WT Jr, Nelson LM, et al. Occupational exposures and amyotrophic lateral sclerosis. A populationbased case-control study. Am J Epidemiol 1997; 145:1076-88. Woolf AD. Ginger Jake and the blues: a tragic song of poisoning. Vet Hum Toxicol 1995; 37:252-4. Glynn P. Neuropathy target esterase. Biochem J 1999; 344(Pt 3):625-31. Louis-Ferdinand RT. Myelotoxic, neurotoxic and reproductive adverse effects of nitrous oxide. Adverse Drug React Toxicol Rev 1994; 13:193-206. Riess O, Kruger R. Parkinson’s disease – a multifactorial neurodegenerative disorder. J Neural Transm Suppl 1999; 56:113-25. Cardoso F, Jankovic J. Movement disorders. Neurol Clin 1993; 11:625-38. Montastruc JL, Llau ME, Rascol O, Senard JM. Drug-induced parkinsonism: a review. Fundam Clin Pharmacol 1994; 8:293-306. Howard JS III. Cocaine, neuroleptics, and tardive dyskinesia as paleocortical escape. Integr Physiol Behav Sci 1996; 31:306-14. Araki S, Yokoyama K, Murata K. Neurophysiological methods in occupational and environmental health: methodology and recent findings. Environ Res 1997; 73(1-2):42-51. Allum JH, Shepard NT. An overview of the clinical use of dynamic posturography in differential diagnosis of balance disorders. J Vestib Res 1999; 9:223-52. Forsting M. MR imaging of the brain: metabolic and toxic white matter diseases. Eur Radiol 1999; 9:1061-5. Filley CM. Toxic leukoencephalopathy. Clin Neuropharmacol 1999; 22:249-60. Triebig G, Lang C. Brain imaging techniques applied to chronically solvent-exposed workers: current results and clinical evaluation. Environ Res 1993; 61:239-50. Lin Z, Chen JD. Advances in time-frequency analysis of biochemical signals. Crit Rev Biomed Eng 1996; 24:1-72. Duffy FH, Hughes JR, Miranda F, Bernad P, Cook P. Status of quantitative EEG (QEEG) in clinical practice, 1994. Clin Electroencephalogr 1994; 25:VI-XXII. Anger WK, Cassitto MG. Individual-administered human behavioral test batteries to identify neurotoxic chemicals. Environ Res 1993; 61:93-106. Anger WK, Otto DA, Letz R. Symposium on computerized behavioral testing of humans in neurotoxicology research: overview of the proceedings. Neurotoxicol Teratol 1996; 18:347-50. Bleecker ML, Hansen JA. Occupational neurology and clinical neurotoxicology. Baltimore, MD: Williams & Wilkins; 1994. Manzo L, Artigas F, Martinez E, et al. Biochemical markers of neurotoxicity. A review of mechanistic studies and applications. Hum Exp Toxicol 1996; 15(Suppl 1):S20-35. Brookes AJ. The essence of SNPs. Gene 1999; 234:177-86. Van Zijl PC, Barker PB. Magnetic resonance spectroscopy and spectroscopic imaging for the study of brain metabolism. Ann N Y Acad Sci 1997; 820:75-96. Lester DS, Pine DS, Delnomdedieu M, Johannessen JN, Johnson GA. Virtual neuropathology: three-dimensional visualization of lesions due to toxic insult. Toxicol Pathol 2000; 28:100-4. Imaging Brain Structure and Function: Emerging Technologies in the Neurosciences. Proceedings of a conference. Bethesda, MD, March 28-29, 1996. Ann N Y Acad Sci 1997; 820:1-315.

660 Central Nervous System Diseases 37. Philbert MA, Billingsley ML, Reuhl KR. Mechanisms of injury in the central nervous system. Toxicol Pathol 2000; 28:43-53. 38. Spencer PS, Schaumburg HH, Ludolph AC. Experimental and clinical neurotoxicology, 2nd edn. New York: Oxford University Press; 2000. 39. Costa LG. The emerging field of ecogenetics. Neurotoxicology 2000; 21:85-9. 40. Cassarino DS, Bennett JP Jr. An evaluation of the role of mitochondria in neurodegenerative diseases: mitochondrial mutations and oxidative pathology, protective nuclear responses, and cell death in neurodegeneration. Brain Res Brain Res Rev 1999; 29:1-25. 41. Fleming KC, Adams, AC, Petersen RC. Dementia: diagnosis and evaluation. Mayo Clin Proc 1995; 70:1093-107. 42. Porru S, Alessio L. The use of chelating agents in occupational lead poisoning. Occup Med (Lond) 1996; 46:41-8. 43. Weiss B, Landrigan PJ. The developing brain and the environment: an introduction. Environ Health Perspect 2000; 108(Suppl 3):373-4. 44. Beal MF. Mitochondrial dysfunction in neurodegenerative diseases. Biochim Biophys Acta 1998; 1366:211-23. 45. Langston JW, Ballard P, Tetrud JW, Irwin I. Chronic parkinsonism in humans due to a product of meperidineanalog synthesis. Science 1983; 219:979-80. 46. Kopin IJ. Parkinson’s disease: past, present, and future. Neuropsychopharmacology 1993; 9:1-12. 47. Degli Esposti M. Inhibitors of NADH-ubiquinone reductase: an overview. Biochim Biophys Acta 1998; 1364:222-35. 48. Betarbet R, Sherer TB, MacKenzie G, Garcia-Osuna M, Panov AV, Greenamyre JT. Chronic systemic pesticide exposure

49. 50. 51.

52. 53. 54. 55.

56.

57. 58.

59.

reproduces features of Parkinson’s disease. Nat Neurosci 2000; 3:1301-6. Watters MR. Organic neurotoxins in seafoods. Clin Neurol Neurosurg 1995; 97:119-24. Grishin E. Polypeptide neurotoxins from spider venoms. Eur J Biochem 1999; 264:276-80. Possani LD, Becerril B, Delepierre M, Tytgat J. Scorpion toxins specific for Na+-channels. Eur J Biochem 1999; 264:287-300. Jones RM, Bulaj G. Conotoxins – new vistas for peptide therapeutics. Curr Pharm Des 2000; 6:1249-85. Albin RL. Basal ganglia neurotoxins. Neurol Clin 2000; 18:665-80. Ming L. Moldy sugarcane poisoning – a case report with a brief review. J Toxicol Clin Toxicol 1995; 33:363-7. Dawson R Jr, Beal MF, Bondy SC, Di Monte DA, Isom GE. Excitotoxins, aging, and environmental neurotoxins: implications for understanding human neurodegenerative diseases. Toxicol Appl Pharmacol 1995; 134:1-17. Teitelbaum JS, Zatorre RJ, Carpenter S et al. Neurologic sequelae of domoic acid intoxication due to the ingestion of contaminated mussels. N Engl J Med 1990; 322:1781-7. Vecsei L, Dibo G, Kiss C. Neurotoxins and neurodegenerative disorders. Neurotoxicology 1998; 19:511-4. Spencer PS, Ludolph AC, Kisby GE. Are human neurodegenerative disorders linked to environmental chemicals with excitotoxic properties? Ann N Y Acad Sci 1992; 648:154-60. Snyder SH, Ferris CD. Novel neurotransmitters and their neuropsychiatric relevance. Am J Psychiatry 2000; 157:1738-51.

28.2 Disorders of the Peripheral Nervous System Michael Pulley, Alan R Berger The peripheral nervous system (PNS) is vulnerable to toxic and occupational injuries that may result in generalized polyneuropathies, focal compressive mononeuropathies, impaired neuromuscular transmission, or myopathy. Neurologic dysfunction may occur in isolation or in concert with other organ system involvement. Occasionally, the latter type of dysfunction suggests a toxic etiology (e.g., gastrointestinal symptoms in acute lead poisoning or alopecia and hyperkeratosis with subacute thallium poisoning). Most often, nervous system dysfunction occurs in isolation, its clinical manifestations being indistinguishable from naturally occurring disorders. A detailed social and occupational history, and determination of the disease’s course relative to the toxin exposure, is vital in establishing a link between the nervous system damage and an occupational or toxic insult. For these reasons, it is imperative that physicians caring for patients with potential occupational and environmental diseases have extensive knowledge of the way in which naturally occurring PNS diseases present, progress, and differ from those of occupational and toxic origin. The peripheral neuropathies are a heterogeneous group of disorders; although much is known about the frequency of the different varieties of diabetic and hereditary neuropathies, there is little data concerning the overall frequency of PNS disease in the general population. A Centers for Disease Control survey of 5000 veterans, utilizing electrodiagnostic and strict clinical criteria, observed a 5% incidence of peripheral nerve dysfunction. A recent survey of 200 workers in a petrochemical facility, utilizing similar criteria, discovered 18% to have evidence of clinical or subclinical neuropathy; the overwhelming majority were traumatic-compressive. Others represented diabetic, alcoholic, and hereditary neuropathy seen in the general population. While symptomatic occupationally induced toxic neuropathy is relatively rare in North America, incidence has not been well characterized; asymptomatic toxic neuropathy may be more common, especially among exposed groups, e.g., exterminators and grouters.

CLINICAL EVALUATION History A focused neurologic, toxicologic, and occupational history is the cornerstone of the clinical evaluation. Because the clinical signs are usually insufficient to establish the etiology, specific elements of the history may be the only clue that suggests toxic or occupational causes. The examiner should inquire about both positive and negative symptoms relevant to motor, sensory, and autonomic function. A complete list of medications and family illnesses should be obtained. Attention should be paid to the following symptoms: (1) motor symptoms and signs, e.g., cramps, fasciculations, myokymia (undulating muscle movements), weakness, easy fatigability, and muscle wast-

ing; (2) sensory symptoms and signs, e.g., paresthesias (tingling or burning), dysesthesias (distorted sensations elicited by tactile stimuli), pain, numbness, anesthesia, and ataxia; and (3) autonomic symptoms, e.g., gustatory sweating, postural hypotension, gastrointestinal and genitourinary dysfunction and anhidrosis. In patients suspected of having a peripheral neuropathy, the medical history may suggest not only the presence of neuropathy but also a pre-existing medical condition that could predispose the patient to either focal nerve entrapments or a generalized peripheral neuropathy (e.g. diabetes mellitus), that may suggest the underlying physiologic nature of the neuropathy. The clinical history may also indicate whether the process is generalized (as might be caused by toxins) or focal (as seen with occupationally related nerve lesions). The history may also disclose specific items that definitively identify a specific toxin or occupational insult. Myopathic and neuromuscular junction dysfunction are manifested by weakness without sensory loss. Weakness is most prominent in the shoulder and pelvic girdle muscles and may be reported as difficulty climbing stairs or arising from a low chair. Getting out of a car is often difficult for patients with proximal weakness. Fatigability may result from impaired neuromuscular transmission. The historian should specifically inquire about the distribution of weakness (e.g., proximal versus distal and pelvic girdle versus shoulder), the presence or absence of pain and tenderness, sensory loss, and involvement of the extraocular muscles (manifested as ptosis or diplopia). The temporal profile should be determined, specifically the acuteness of onset, chronicity, and rapidity of progression. Almost all toxic neuropathies have their onset temporally related to exposure. It is a general rule in neurotoxicology that nerve dysfunction should begin around the time of ongoing or recent exposure and should eventually stabilize after exposure is terminated. In some cases, prolonged low-level exposure to toxins may damage the PNS so insidiously that the patient either does not recognize the dysfunction or is not able to establish a temporal relationship between exposure and disease. Therefore, the history must carefully explore all possible exposures or inciting conditions, both by the compound’s formal and common name and source (e.g., mercury and its lay term, ‘thermometer tubes’, or nitrous oxide and its common reference, ‘sniffing the blue tank’). The occupational history should focus on habits that potentially predispose the patient to occupationally related nervous system disease (e.g., does the patient wear protective devices and change clothing before coming home; whether there is eating in the workplace). The health of coworkers, family members, and household pets should be inquired about; a similar illness might indicate a common exposure. Improvement of symptoms during time away from potential exposure (e.g., weekends or holidays away from the job) and the condition of the workplace (ventilation and drainage) are important items to be determined. In

662 Disorders of the Peripheral Nervous System addition, a complete list of occupational exposures should be obtained because, occasionally, a combination of toxins may be responsible. In some instances, a visit to the workplace is crucial in identifying a specific toxic or responsible environmental condition. In suspected iatrogenic or domestic toxicants, family members and friends should be questioned. A visit to the home may also be helpful. Specific questions regarding hobbies, food and water sources and recent pesticide applications or other chemical often yield crucial information.

Physical examination The scope and depth of the physical examination are shaped by the information and differential diagnosis generated by the history. As an example, a history of occupational trauma suggests that a focal compressive neuropathy is likely and focuses the examination on potentially vulnerable nerves. Alternatively, a history indicating generalized nerve dysfunction directs the clinician towards probable polyneuropathy. Motor symptoms, without sensory loss, suggest muscle or neuromuscular junction dysfunction and mandate specific attention to proximal muscle strength and examination of extraocular muscles. The neurologic examination is performed in the conventional manner; a determination should be made regarding the predominant modalities affected (motor, sensory, autonomic, or mixed) and the overall pattern, distribution, and severity of the deficits. Specific topographic patterns of peripheral nerve dysfunction include focal or multifocal, bilaterally symmetric, proximal versus distal, and upper versus lower extremity. A brief bedside examination of mental status usually suffices when assessing patients suspected of having PNS disease. Questions should be included that examine orientation, short- and long-term memory, abstract thinking, calculation, and attention. An overall sense of the patient’s mental status can often be obtained by the manner and attentiveness with which the patient relates the neurologic history. More subtle cognitive abnormalities are best identified by formal neuropsychologic testing (see Chapter 28.1). The cranial nerve examination should especially focus on facial sensation and the strength of facial musculature. Because most neuropathic toxins produce a lengthdependent distal axonopathy, facial sensation and muscle strength are spared until late in the disease course. Although facial numbness is a feature of trichlorethylene (TCE) intoxication, its presence may be related to a naturally occurring condition, such as Sjogren’s syndrome or scleroderma, rather than a toxic neuropathy. The motor examination should determine the topographic distribution of the deficit (e.g., multiple distal nerve segments, radicular or segmental, or limited to a few specific nerves). Most toxic insults produce generalized distal axonopathies; as such, weakness initially involves the distal legs with subsequent progression to the proximal leg muscles and hands. Findings that substantially differ from this, such as early proximal muscle weakness or hand involvement occurring prior to distal leg weakness, are less

consistent with most toxic neuropathies and suggest another diagnosis. Weakness and/or atrophy in a single nerve distribution usually indicates a focal peripheral nerve lesion. The examination should determine whether all muscles of that nerve are affected or only those distal to potential entrapment sites. Aside from lead and TCE, most toxins do not cause focal nerve damage; such findings most often result from focal compression, which may or may not be caused by occupationally related trauma. Muscle atrophy suggests a chronic problem. Fasciculations are clinically obvious muscle twitchings, which, aside from suggesting a lower motor neuron process, may occur with lesions anywhere from the anterior horn cell to the distal nerve terminal. The sensory examination should attempt to identify the predominant modality (and fiber size) affected. Large-fiber dysfunction involves loss of joint position sense, vibration, touch-pressure sensitivity, areflexia, sensory ataxia, and pseudoathetosis. Small-fiber dysfunction is characterized by a loss of pain and temperature sense, retained reflexes, and occasionally, autonomic dysfunction. In general, sensory abnormalities are most reliable when they correspond to appropriate patient complaints. Unfortunately, the sensory examination findings may be inconsistent or variable. Too much diagnostic emphasis should not be placed on inconsistent or neurologically inappropriate sensory findings. The presence of hyperesthesia (exaggerated response to sensory stimuli) or dysesthesia (unpleasant response to a normal non-noxious stimulus) should be noted. Mild sensory deficits are best appreciated when sensory stimuli are initially applied to areas of lesser sensitivity, and then the results are compared with those in corresponding normal areas. Patterns of sensory deficits include a distal stocking– glove distribution (which would be most common in generalized neuropathies), segmental or dermatomal sensory loss (as occurs with radicular disease), or sensory loss in the distribution of one or more peripheral nerves. Most toxic polyneuropathies are of a mixed nature with all sensory modalities affected. The sensory examination should be correlated with the motor findings to determine whether a generalized or focal problem exists and to localize it topographically (e.g., anterior horn cell, ventral root, dorsal root ganglia, or peripheral nerve). Tendon reflex testing should include the biceps (C5 and C6), brachioradialis (C5 and C6), triceps (C7), patella (L3 and L4), and Achilles (S1) reflexes. Reflex loss usually localizes the process to the PNS; in contrast, hyper-reflexia usually results from upper motor neuron lesions. In generalized axonal neuropathies, the initial tendon reflex to be diminished is the Achilles, followed by the patella and upper limb reflexes. Focal or asymmetric reflex loss suggests radicular or focal peripheral nerve disease. The presence or absence of a reflex, and its intensity, are poor indicators of disease severity, progression, or recovery. Patients often substantially recover sensory and motor function, yet have a poor return of tendon reflexes. Examination of gait and stance is of critical importance in neurologic diagnosis. Patients with weakness of distal leg muscles may walk with a steppage gait as a result of bilateral

Laboratory Evaluation 663 weakness of the ankle dorsiflexors. A waddling gait occurs in patients with proximal weakness. Severe large-fiber sensory loss may be initially manifested as gait instability, resulting in a wide-based stance and instability on standing with the eyes closed but not with eyes open (positive Romberg’s test). Painful neuropathies may preclude the patient from walking; such individuals often walk on their toes to avoid tactile stimulation to their soles or have a slow, hesitant gait because of painful movements (antalgic gait). Tests of coordination are generally normal in patients with peripheral nerve disease. Severe sensory loss resulting in limb and gait ataxia should not be confused with cerebellar disease.

LABORATORY EVALUATION Electrodiagnostic studies The most critical laboratory investigation in evaluating peripheral neuropathies is the electrodiagnostic examination. Commonly referred to as nerve conduction velocity (NCV) and electromyography (EMG), the study consists of a nerve conduction portion, in which electrical stimulation is externally applied to the nerves to determine their conductive properties, and a needle EMG portion, which utilizes a sterile needle electrode, inserted directly into the muscle. Needle EMG assesses the electrical activity of a muscle and provides information regarding muscle function and motor unit integrity. Taken in concert, nerve conduction and needle EMG studies provide objective measurements of muscle, neuromuscular junction, and peripheral nerve function and are invaluable in diagnosing diseases of the PNS.

Techniques employed Sensory conduction studies. The most sensitive electrophysiologic technique in evaluating peripheral neuropathies is the determination of sensory and mixed nerve conduction velocities and sensory potential amplitudes. Sensory conduction studies are performed by electrically stimulating either the pure sensory nerve or the mixed sensorimotor nerve. The recordings are made over the respective nerves, either distal (antidromic recording) or proximal (orthodromic recording) to the site of stimulation. Stimulation must always be supramaximal to ensure activation of all available sensory axons, and strict attention must be paid to limb temperature and proper positioning of recording and stimulating electrodes. Sensory potentials may be difficult to obtain in elderly patients or those with peripheral nerve disease and usually require computer averaging to increase the signal-to-noise ratio. Information obtained from conventional nerve conduction studies predominantly reflects conduction within large-diameter sensory fibers. Degeneration of sensory axons is directly reflected by a diminution of sensory potential amplitudes. As such, amplitude abnormalities, rather than conduction velocities, tend to be the most sensitive indicator of an axonal peripheral neuropathy. Segmental demyelination results in prolongation of distal latencies and slowed conduction velocity. Sensory

potentials are often temporally dispersed, resulting in complex polyphasic (serrated) potentials. Unlike neuropathies, preganglionic lesions (at or proximal to the root level) do not affect the distal sensory potential. Sensory conduction studies are therefore useful in distinguishing radicular disease from peripheral neuropathy.

Motor conduction studies. Motor conduction studies are performed by supramaximally stimulating mixed or motor nerves and recording the compound muscle action potential (CMAP) over the muscle’s endplate region. Stimulation is at a proximal and distal site, and conduction velocity is determined for the intervening nerve segment. The amplitude of the CMAP represents the surface-recorded summation of multiple muscle fiber potentials, and as such, it parallels but does not directly correlate with the number of available motor axons. Because CMAPs are in the millivolt rather than the microvolt range (like sensory potentials), they are usually easier to obtain and do not require averaging techniques. Motor conduction velocities can be determined for any accessible nerve by varying the site of stimulation. Similar to sensory conduction, motor conduction studies generally reflect large-fiber function; the distal latency and conduction velocity reflect conduction in the fastest fibers. All neuropathic lesions, including peripheral neuropathies, which result in axonal degeneration, will diminish CMAP amplitude. In most axonal sensorimotor neuropathies, changes in motor potential amplitudes lag behind those of sensory potentials. In axonal neuropathies, distal latencies and motor conduction velocities tend to remain unchanged until the loss of large-diameter, fast-conducting fibers necessitates conduction through only small-diameter fibers. In contrast, demyelinating neuropathies result in distal motor latency prolongation, slowed motor conduction velocities, and CMAP temporal dispersion; this is especially evident with proximal stimulation. Low-amplitude CMAPs with intact corresponding sensory potentials strongly suggest pathologic changes at or proximal to the root level. When motor or sensory conduction is focally slowed across a potential entrapment site, a compressive lesion should be suspected.

Needle EMG.

Needle electrode recordings from normal and injured muscle provides crucial information regarding motor unit function. Resting normal muscle has little or no electrical activity aside from brief insertional discharges. Separation of the muscle fiber from its innervating nerve fiber (denervation) results in abnormal electrical activity, termed spontaneous activity (fibrillation or positive sharp wave potentials that occur with the needle at rest). The degree of spontaneous activity is only a rough guide to the extent of motor unit degeneration. Spontaneous activity is usually indicative of a neuropathic process, although it can also be seen in certain myopathies and, occasionally, in normal muscles. Fasciculation potentials may be evident clinically and by needle EMG. They represent the involuntary and random firing of single motor units. They have little localizing value because they may occur with anterior horn cell

664 Disorders of the Peripheral Nervous System disease, radiculopathies, peripheral neuropathies, and a number of benign conditions. Quantitative analysis of voluntary motor units (amplitude, duration, and degree of polyphasia) provides information regarding the chronicity of the neuropathic lesion or may suggest the presence of a myopathic process. Chronic motor unit reinnervation results in motor unit potentials (MUPs) that are increased in amplitude and prolonged in duration and may be polyphasic compared with those of normal units. This is in contrast to the small-amplitude, short-duration MUPs evident in many myopathies. Needle EMG also provides information regarding voluntary motor unit recruitment. Reduced recruitment is characteristic of most neuropathic conditions; myopathies tend to have normal motor unit recruitment despite clinical weakness. Because needle EMG analysis can be performed on multiple proximal and distal muscles, it is useful in determining the distribution and severity of a neuropathic process.

Quantitative sensory testing.

Quantitative sensory testing (QST) is a non-invasive, painless technique to quantify vibration, temperature, and pain appreciation. It uses precisely measured and repeatable sensory stimuli to determine the absolute threshold of sensory appreciation. Several commercially available QST devices exist for each sensory modality to be quantified. QST is simple and can be administered by technicians. Accurate, age-controlled mean and standard deviation values are available. QST is especially recommended for rapid screening of large populations (e.g., workers at risk for toxic neuropathy) and/ or longitudinal evaluations of patients at risk for subtle sensory dysfunction (e.g., cumulative trauma disorders). QST abnormalities may predate nerve conduction abnormalities in generalized peripheral neuropathies.

Nerve biopsy.

The utility of nerve biopsy has been exaggerated. It has limited use in patients with generalized axonal toxic neuropathies and should be reserved for centers that have the expertise to examine and quantify the specimen fully. Nerve biopsy is usually performed on the sural nerve at the ankle and calf level. It is most helpful in identifying the etiology of multifocal neuropathies, such as amyloidosis, sarcoidosis, leprosy, and vasculitis. In general, nerve biopsy has little or no role in the evaluation of the patient with suspected toxic neuropathy because most of these entities result in axonal degeneration without specific diagnostic findings distinct from endocrine, metabolic, or nutritional neuropathies. Specific exceptions, such as suspected exposures resulting in giant axonal neuropathy, are discussed in the section on specific toxic neuropathy below.

PATHOPHYSIOLOGY OF PNS DISORDERS Neurotoxic and occupational insults to the PNS usually produce syndromes that clinically mimic naturally occurring disorders. The PNS is relatively limited in the ways it

manifests injury. Certain insults tend to produce stereotypical disease manifestations; their recognition helps establish the nature of the neuropathic lesion. Classification involves identifying the main site of neuropathic dysfunction (e.g., muscle, neuromuscular junction, or peripheral nerve, with differentiation between cell body, axon, and myelin) and the distribution of such lesions (e.g., proximal, distal symmetric, focal, multifocal, or segmental). The following is a brief summary of the generic types of peripheral nerve dysfunction, each of which may be caused by a number of metabolic, nutritional, infectious, and ischemic causes, as well as by toxins and trauma.

Symmetric generalized neuropathies Distal axonopathies (central-peripheral axonopathy and dying back neuropathy) The most common peripheral neuropathy that results from PNS insults, particularly toxin-induced injuries, is a symmetric distal axonopathy (Fig. 28.2.1). In many cases, the biochemical and pathophysiologic mechanisms are poorly understood. These neuropathies probably reflect failure of axonal transport, with resultant degeneration of vulnerable distal nerve segments, predominantly affecting large-diameter axons. Degeneration subsequently proceeds proximally toward the nerve cell body, both in the PNS and in central projections within the spinal cord. Most distal symmetric axonopathies have a subacute onset with gradual progression. Neuropathies resulting from low-level toxin exposure may be relatively asymptomatic, with deficits apparent only to the physician on careful neurologic examination. Because the longest, largest diameter fibers are usually the most clinically affected, motor and sensory findings initially appear in the feet, only later moving proximally (length-dependent relationship). Sensory loss is initially in a stocking, and later glove, distribution. As the neuropathy worsens, the distal ends of intercostal nerves are affected, producing a cuirass, or shield, over the midthorax and abdomen. With extreme progression, the vertex of the head is affected. There is usually an early and symmetric loss of ankle reflexes; the more proximal reflexes may be spared until late in the disease. In most toxic neuropathies, sensory symptoms and signs initially predominate over motor deficits. Muscle wasting may occur in chronic cases, and trophic changes may be present, including loss of hair over the distal leg, skin ulceration, and loss of sweating in the feet. Because recovery depends on axonal regrowth, complete recovery is often prolonged and slow. Axonal regeneration occurs at a rate of, on average, 2–3 mm/day. Even after removal from exposure, recovery may take months to years as the recovering nerves regenerate to their muscle end-organs through intact Schwann cell tubes, or uninjured motor axons supply collateral sprouts that innervate denervated muscle fibers. Function is restored in the reverse order to that lost; proximal muscles recover before distal muscles, and sensory loss recedes from proximal to distal levels. The clinical manifestations of concur-

Pathophysiology of PNS Disorders 665

Astrocyte proliferation

CNS

PNS

Axon regeneration

Normal

Early

Late

Recovering

Figure 28.2.1: The cardinal pathologic features of toxic distal axonopathy. The jagged lines (lightning bolts) indicate that the toxin is acting at multiple sites along motor and sensory axons in the peripheral nervous system (PNS) and central nervous system (CNS). Axon degeneration has moved proximally (dying-back) by the late stage. Recovery in the CNS is impeded by astroglial proliferation. (Adapted from Schaumburg HH, Spencer PS, Thomas PK, eds. Disorders of peripheral nerves. Philadelphia: FA Davis, 1983.)

rent degeneration of central axons may be initially masked by the lower motor neuron dysfunction but become clinically evident as peripheral nerve function recovers. The sequential manifestation of early PNS dysfunction, followed by central nervous system (CNS) symptoms as PNS function improves, is especially characteristic of some toxic neuropathies (e.g., organophosphates). Symptoms of CNS dysfunction include hyper-reflexia, Babinski’s signs, and spastic tone.

Demyelinating neuropathies Acquired demyelinating neuropathies (myelinopathies) are conditions in which the predominant lesions occur in the myelin sheath or Schwann cells (Fig. 28.2.2). Various degrees of associated axonal degeneration may accompany them. The most common example of an acquired demyelinating neuropathy is the Guillain–Barré syndrome (acute inflammatory demyelinating neuropathy, AIDP). Almost all toxins result in axonal rather than demyelinating neuropathies, with the notable exceptions of buckthorn toxin, diphtheria toxin, and perhexiline (Pexid). Acquired demyelinating neuropathies usually have a subacute onset. Although initial clinical deficits usually involve the distal limbs, similar to axonopathies, the demyelinating neuropathies may be patchy, resulting in early proximal weakness, arm involvement before legs, and facial numbness. Early and diffuse areflexia is charac-

teristic. Unlike axonopathies, in which sensory symptoms predominate, myelinopathies may be predominantly motor. Sensory symptoms may be transient, mild, or inapparent. Large- rather than small-fiber modalities tend to be most severely affected. Occasionally, this large-fiber sensory loss results in limb and gait ataxia. Muscle wasting is usually minimal unless substantial axonal degeneration has occurred. Recovery not only begins earlier than with axonopathies but is usually more complete, owing to the greater number of intact axons. Signs of CNS dysfunction are only rarely present, resulting from concurrent demyelination within the CNS.

Neuronopathies Injury to the cell body is termed neuronopathy; the clinical manifestations reflect dysfunction restricted to the segments innervated by the affected cell bodies. Neuronopathies are rarely caused by toxic insults. Motor, sensory, and autonomic neurons may be affected. Toxic neuronopathies, such as those from mercury, may affect the CNS or PNS, and neurons (Fig. 28.2.3); other causes of toxic neuronopathies include pyridoxine, megavitaminosis and doxorubicin (Adriamycin). The pathologic basis is heterogeneous and, in most cases, poorly understood. Toxic neuronopathies, such as that from doxorubicin, probably result from disruption of sensory neuron nucleic acid metabolism with subsequent

666 Disorders of the Peripheral Nervous System

CNS

PNS

Normal

Attack by inflammatory cells

Segmental demyelination

Remyelinated fibers

Figure 28.2.2: The cardinal pathologic features of an inflammatory PNS myelinopathy. Axons are spared, as is CNS myelin. (Adapted from Schaumburg HH, Spencer PS, Thomas PK, eds. Disorders of peripheral nerves. Philadelphia: FA Davis, 1983.)

sensory axon degeneration. Dorsal root ganglia may be vulnerable to high molecular weight toxins because of a poorly formed blood–nerve barrier. The presence of fenestrated blood vessels in dorsal root ganglia increases vascular permeability and therefore toxin exposure. Most neuronopathies are characterized by rapid or subacute onset of motor or sensory deficits that do not obey the length-dependent relationships seen with distal axonopathies. Initial sensory loss can occur anywhere; facial numbness occurring concurrently with sensory loss in the limbs is characteristic and reflects simultaneous involvement of cranial nerve and spinal dorsal root ganglia. Sensory loss is usually widespread, but strength is preserved. Although all sensory modalities are affected, vibration and position sense are the most severely impaired. Generalized areflexia is common and reflects loss of large-fiber function. Recovery is variable and often incomplete because of sensory neuron degeneration. Ganglion cells that are damaged but not killed may recover. Clinical recovery depends on the return of function within surviving neurons and collateral sprouting from intact sensory axons. Although CNS signs are not invariable, they may be present.

Focal (mononeuropathy) and multifocal neuropathies Focal neuropathies may be caused by traumatic compression, chronic entrapment in fibro-osseous tunnels, traction

injuries, or ischemic injury, usually caused by small-vessel angiopathy (such as diabetes or necrotizing arteritis). Focal neuropathies are rare in toxic disease (exceptions are leadrelated neuropathy, initially producing wrist drop, and TCE, producing trigeminal neuropathy). The most common focal neuropathies include: 1. Median nerve entrapment at the wrist or elbow. 2. Ulnar nerve entrapment at the elbow or wrist. 3. Radial nerve compression in the upper arm or forearm. Focal nerve lesions can be classified by the degree of myelin and axonal injury. Various nomenclature systems have been proposed; regardless of which is used, each specifies the nature of the nerve injury as follows. 1. Mild injury in which myelin is damaged but the axon cylinders remain unaffected (Class I, neurapraxia). 2. More severe injury in which axonal continuity is lost but the nerve’s connective tissue framework is maintained (Class II, axonotmesis). 3. Injury in which both nerve fibers and the connective tissue framework with varying degrees of damage (Class III, neurotmesis). Aside from severe industrial accidents with open wounds, most occupationally related focal neuropathies are Class I and II. Some entrapments may be predisposed by an underlying, baseline condition. As an example, some patients with carpal tunnel syndrome (CTS) have congenitally small diameters of the carpal tunnel, a condition predisposing to median nerve entrapment. Focal neuropathies are charac-

Toxic and Occupational PNS Disorders 667

Astrocyte proliferation

CNS

PNS

Normal

Toxic attack on DRG

1 week later

6 months later

Figure 28.2.3: The cardinal features of a rapidly involving toxic sensory neuronopathy. The jagged lines (lightning bolts) indicate that the toxin is directed at neurons in the dorsal root ganglion (DRG). Degeneration of these cells is accompanied by fragmentation and phagocytosis of their peripheral-central processes. The Schwann cells remain; there is no axonal regeneration. (Adapted from Schaumburg HH, Spencer PS, Thomas PK, eds. Disorders of peripheral nerves. Philadelphia: FA Davis, 1983.)

terized by sensory and motor deficits within the distribution of the involved nerve or nerves. Many entrapments, such as CTS, are associated with pain, which may bring the patient to medical attention. In some instances, the mononeuropathy may be symptomatic only during, or after, repeated movement or work-related activities. Other entrapments, such as ulnar nerve lesions at the elbow or wrist, may be painless and present with weakness more than with sensory loss. Such cases may initially display advanced degrees of muscle atrophy because of the long delay in seeking medical attention. In focal neuropathies, only the segmental reflexes subserved by the involved nerve are affected. The degree of recovery generally depends on the underlying etiology (overuse, ganglion, fibrous band, hypertrophied muscle, or inflammation), the ability to avoid aggravating activities, and the degree of motor axon degeneration.

TOXIC AND OCCUPATIONAL PNS DISORDERS Toxic peripheral neuropathies Toxic polyneuropathies (TxPN) are relatively infrequent in North America. Most toxic polyneuropathies encountered in routine clinical practice are due to iatrogenic pharmaceutical intoxications, with occupational exposures less

common. The majority, and unfortunately the most difficult, cases of TxPN are individual intoxications due to small-scale occupational exposures, or intentional and homicidal ingestions. The identification that a sporadic peripheral neuropathy results from toxin exposure in the occupational setting is often made difficult by an unclear exposure history. TxPN are usually distal axonopathies and thus clinically and electrophysiologically resemble neuropathies from metabolic abnormalities, nutritional deficiencies, or systemic illness. Clinically relevant and reliable toxicologic tests are often unavailable or unhelpful, either because the necessary laboratory tests are not available, or the substance is undetectable because of the delay between exposure and examination. When a naturally occurring medical cause is not readily apparent, diagnosis of a TxPN must be made with caution based on the principles discussed below (see following section). The underlying pathology of many TxPN is a centralperipheral axonopathy.1,2 Our limited knowledge of the biochemical and pathophysiologic mechanisms of most neurotoxins has led to an overly simplified classification system according to compound class (e.g., solvents, metals). Such a classification is of limited clinical utility. A compound should not be presumed to be neurotoxic because of a superficial resemblance to a related known toxin of similar class; all compounds within the same class

668 Disorders of the Peripheral Nervous System are not necessarily neurotoxic (e.g., acrylamide monomer is capable of producing a devastating peripheral neuropathy, while the polymer does not). Structure–toxicity relationships are clear for some classes of substances, such as organophosphates and hydrocarbons.

Cardinal tenets of neurotoxic illness affecting the peripheral nervous system The identification of a neurotoxic illness should satisfy, or at least not be inconsistent with, the following basic principles of neurotoxic disease.3 The key to correctly recognizing the presence of a TxPN does not depend on remembering the specific characteristics of the many potential neurotoxins, as much as understanding and applying these basic tenets.

Strong dose–response relationship. Most neurotoxins produce a consistent pattern of disease, commensurate with the dose and duration of exposure. Neurotoxins rarely cause focal or asymmetric deficits. Since most neurotoxins cause diffuse myelin and/or neuronal dysfunction, their related symptoms and signs are usually widespread and symmetric. In the case of TxPN, this usually means a relatively symmetric distal axonopathy with initial symptoms in the feet and proximal progression, with continued exposure. Less frequently does an occasional toxin cause strikingly asymmetric or focal dysfunction (e.g., lead and trichloroethylene). Consistency of response. Although the same toxin may produce strikingly different clinical syndromes if the exposure dose or duration is different, a similar and consistent illness typically results in patients with similar exposures (though individual variation may occur). Neurotoxicity should be suspected when similar clinical manifestations occur in a group of individuals with a common chemical exposure. Proximity of symptoms to exposure. Neurotoxic illness usually occurs concurrent with exposure or following a short latency. Neurologic symptoms do not generally begin months to years after exposure. The two most common exceptions are the 2–6-week delay following exposure to organophosphates and the occasional 2-month latency between cisplatin intoxication and onset of neuropathic symptoms. In addition, the extent and severity of neuropathy are usually commensurate with the degree of toxin exposure. Thus, it is unlikely that a single, brief, low-level exposure will result in a devastating peripheral neuropathy. Some lipid stored agents (e.g., chlorinated hydrocarbons) are detectable in fat biopsies years following exposure. Although this provides a valuable marker of previous exposure, there is no evidence that this state is associated with risk for future neurotoxicity, and attempts at removal or mobilizing the body burden are unnecessary.

Improvement usually follows cessation of exposure. Toxic polyneuropathies generally plateau and then gradually improve after removal of the neurotoxic agent. Some

degree of recovery is the rule, except in the most severely affected cases. A neuropathy that shows no improvement or continues to deteriorate, despite the cessation of exposure to a suspected neurotoxin, is less likely to be neurotoxic in nature. The clinical picture may become somewhat murky, however, in certain toxic axonopathies in which cessation of exposure may be followed by worsening of symptoms (coasting) for several weeks before recovery commences.

Confusing aspects of neurotoxic illness Multiple clinical syndromes may result from different levels of exposure to a single toxin. Different exposure levels to the same substance may produce dramatically different syndromes. Most confusing is the bizarre constellation of symptoms that may arise from intoxication with intermediate levels of a neurotoxin. Examples include the different clinical syndromes produced by acute high-level and intermediate-level exposure, and prolonged low-level acrylamide intoxication. Exposures to high-level acrylamide causes early CNS dysfunction with drowsiness, disorientation, hallucinations, seizures, and severe truncal ataxia, followed by neuropathy of variable severity. In contrast, prolonged, lower-level exposure causes minimal CNS dysfunction but a marked peripheral neuropathy. Exposure to intermediate levels of acrylamide causes hallucinations, mental confusion and cognitive dysfunction, followed by sensory complaints affecting the distal limbs. Another example is organophosphate poisoning in which there may be early, severe cholinergic symptoms resulting from excessive muscarinic receptor stimulation. Within 2 weeks generalized paralysis may occur with respiratory distress owing to nicotinic receptor blockade. After a few weeks, a distal axonopathy may be evident. In some instances, a single compound may produce similar clinical symptoms at both high-level and low-level exposure, although different anatomic structures are affected. High-dose pyridoxine intoxication produces widespread sensory loss due to dorsal root ganglion dysfunction; lowlevel exposure produces similar symptoms but due to a distal axonopathy.

Asymptomatic disease. Prolonged, low-level exposure may occasionally produce widespread subclinical dysfunction. Clinical deficits may go unnoticed by the patient unless they perform a skilled job that requires fine-motor control or intact sensibility. Insidiously developing subclinical TxPN may occur in individuals who deny any disability. Enhancement by chemical interaction.

An agent without known neurotoxic activity may enhance the toxicity of a known neurotoxin that is present at a generally subtoxic level, a phenomenon sometimes termed the ‘bystander effect’. This phenomenon has raised the general public’s concern that the combined effects of multiple chemicals in hazardous waste disposal sites may be more toxic than their separate effects. Such sites may contain low, levels of

Toxic and Occupational PNS Disorders 669 neurotoxic solvents, metals or pesticides, whose neurotoxicity may be potentiated by one or other of the chemicals present. Neurotoxic potentiation is illustrated by the epidemic of peripheral neuropathy which occurred in German youths who abused paint thinner containing nhexane. Initially there were no instances of neuropathy, but when the paint thinner was reformulated by lowering the concentration of n-hexane and adding methyl ethyl ketone (MEK), there resulted an epidemic of severe distal axonopathy. Experimental evidence subsequently showed that while MEK by itself was not a significant neurotoxin, the compound dramatically potentiated the neurotoxic effects of n-hexane.

Chemical formula may not predict toxicity.

The neurotoxic potential of a compound cannot be accurately predicted by its chemical formula. This is especially important to consider when evaluating cases of potential occupational exposure to chemicals that superficially resemble a known neurotoxin. An example is workers exposed to acrylamide polymer, a substance without associated neurotoxicity, who have been alarmed by healthcare providers familiar only with the effects of acrylamide monomer, a potent neurotoxin. Unpredictability exists because the underlying biochemical mechanisms and active metabolites of most neurotoxins are unknown.

Determination of body burdens Several factors potentially limit the interpretation of screening levels for heavy metals. The toxin exposure may be too remote, allowing time for the offending agent to be cleared from the blood or other biologic specimen. In cases of prolonged exposure, the neurotoxin may be sequestered in various tissues and therefore not available to laboratory identification. With some chemicals, reliable reference ranges have not been established. In some cases, identification of the toxic form of a metal is required. Caution must thus be exercised when interpreting body burden results. An example is arsenic levels, which may be raised by recent shell food ingestion due to non-toxic organic forms. In such cases special testing may be required to assess the level of toxic inorganic arsenic. In some cases prolonged exposure (e.g., lead), or elevated levels in urine or serum after chelation, may increase the sensitivity of testing.

Electrodiagnostic assessment Electrophysiologic findings should be consistent with a distal axonopathy or mixed axonal, demyelinating neuropathy. Only a few rare neuropathies, such as n-hexane, perihexiline, amiodarone, and early arsenic poisoning, have predominant slowing of conduction velocities.

Quantitative sensory testing Identification of toxic peripheral neuropathy The presence of a toxic peripheral neuropathy is suggested by the following: 1. clinical suspicion raised by history and reinforced by compatible findings on physical exam; 2. lack of naturally occurring alternative explanations; 3. consistency with basic principles of neurotoxic disease (see above); 4. compatible laboratory findings (e.g., electrodiagnostic studies); 5. demonstration of elevated body burdens for a neurotoxic agent, (usually available only for active or recent exposures) or resolution of condition with removal from exposure. The initial step is a suspicion raised by a thorough occupational history. Unfortunately, most toxic polyneuropathies are insidious in onset, and many patents are unable to discern a relationship between their symptoms and chronic, low-level, toxin exposure. Inquiry should focus on potential occupational, environmental, and iatrogenic exposures. The nature of the suspected toxin should focus the physical examination towards relevant deficits. Thus a suspicion of mercury poisoning should prompt a careful examination for tremor and mild cerebellar dysfunction. The role of the neurologic exam is to demonstrate that neurologic deficits are in a pattern and of a severity that is consistent with neurotoxic illness. Since the clinical deficits resulting from a TxPN are generally symmetric in distribution, the presence of multifocal deficits should suggest a diagnosis other than neurotoxic disease. In addition, since most TxPN affect mixed nerve function, finding a purely small-fiber neuropathy makes neurotoxic disease less likely.

Quantitative thresholds for thermal and vibration appreciation have proven useful in documenting objective evidence of sensory impairment and monitoring the course of recovery or deterioration. These procedures are non-invasive and reproducible, and can be performed by a trained technician.

Systemic features suggestive of neurotoxic disease The neuropathies resulting from most neurotoxins are remarkably similar in both their clinical and electrophysiologic characteristics. Occasionally, there may be systemic complaints or signs which suggest the nature of the neurotoxic insult. Usually these symptoms/signs are apparent with either acute high-level, or chronic low-level intoxication. The following clinical characteristics may be the identifying feature that suggests a TxPN. • Acrylamide: dermal contact associated with contact dermatitis and/or excessive sweating of hands and feet. • Carbon disulfide: chronic low-level exposure associated with a variety of behavioral and psychiatric abnormalities, along with peripheral neuropathy. • Ethylene oxide: cognitive impairment and neuropathy with prolonged low-level exposure. • Hexacarbons: acute, high-level exposure may mimic AIDP with prominent autonomic dysfunction. • Lead: Mee’s lines, blood abnormalities (basophilic stippling, anemia), gastrointestinal abnormalities, and predominantly a motor neuropathy. • Mercury: tremor and ataxia with a predominantly sensory neuropathy. • Ethyl bromide: corticospinal and cerebellar dysfunction along with an axonal neuropathy.

670 Disorders of the Peripheral Nervous System • Organophosphate intoxication: early cholinergic symptoms, with possible intermediate syndrome preceding neuropathy and late emergence of corticospinal tract dysfunction as the peripheral neuropathy resolves. • Polychlorinated biphenyls: symmetric sensory neuropathy associated with brown acneiform skin eruptions (chloracne) and brown pigmented nails. • Thallium: prominent GI distress with high-level exposure, alopecia, Mee’s lines, hyperkeratosis with prolonged exposure, and sensory greater than motor neuropathy.

Specific toxic neuropathies Metals Excessive exposure to specific inorganic and organic metal compounds may cause peripheral nerve disease. Heavy metals are commonplace in industrial and agricultural settings. Exposure may be through inadvertent contamination or through voluntary ingestion, e.g., suicide attempts with arsenic or thallium. Two properties of metals are important in regard to the peripheral neuropathies they produce. The first is that metal compounds tend to be stored in a number of body organs (e.g., lead in bones). Delayed and gradual release of the toxin back into the circulation from these tissues even after cessation of exposure, may delay the time to recovery. The second property of metals is that, when present in sufficient quantity, they rarely affect the PNS in isolation. As such, systemic symptoms and signs related to hematopoietic, renal, and gastrointestinal dysfunction may accompany the peripheral neuropathy.

manifests within about a week. Painful paresthesias and numbness are the predominant early symptoms beginning in the feet and then affecting the hands. Weakness soon follows and is also expressed in a length-dependent pattern, starting with the feet and later involving the hands.6 With high-dose exposure or inadequate treatment, the weakness may become severe and involve the respiratory muscles, mimicking Guillain–Barré syndrome. The deep tendon reflexes are depressed or absent early in the process. Chronic low-level arsenic exposure results in dermatologic manifestations prior to overt clinical neuropathy. Patients may complain of non-specific symptoms such as anorexia, malaise, generalized weakness and vomiting. The dermatologic changes that follow include white transverse lines in the nails (Mee’s lines), hyperkeratosis, hyperpigmentation of the skin, and irritation of the mucous membranes. Although the neuropathy is asymptomatic at this point, careful examination or electrophysiologic testing may reveal its presence. Continued exposure leads to development of symptomatic neuropathy. As with that caused by acute exposure, this is characterized by prominent sensory burning and numbness of the feet and later the hands. Small and large sensory fibers are affected with resultant difficulties with proprioception in addition to the dysesthesias. Weakness tends to be mild and limited to the most distal muscles. Hematologic disturbances including symptoms of anemia and pancytopenia may also result from chronic arsenic exposure. Recovery is variable, often with mild persistent neuropathy. The neuropathy may continue to worsen for a period of weeks after removal from exposure (coasting) in both the acute and chronic neuropathies, and may result in significant residua.

Acute arsenic poisoning classically occurred as a homicide or suicide, where a massive exposure was the usual case. However, chronic low-grade exposure, as well as occasional acute exposures, in the occupational setting also occur. Arsenic toxicity has been described in the smelting of lead and copper ore, mining, and the manufacture of integrated circuits or microchips. Non-occupational exposure may occur through contaminated well water, tainted illicit drugs, and the use of treated lumber (e.g., CCA). Arsenic gains entry to the body by inhalation, GI absorption and via dermal contact. The toxicity of arsenic may be related to its affinity for thiol groups. This affinity leads to binding with lipoic acid, which interferes with the conversion of pyruvate to acetyl CoA, and thus energy metabolism.4

Diagnostic considerations. Arsenic levels may be measured in the urine, and less effectively in the hair and nails. Urine arsenic levels may remain elevated for weeks after exposure. Levels greater than 25 μg per 24 hour urine specimen are abnormal unless there was recent seafood ingestion (a source of organic arsenic). Low-level long-term exposure, or exposure that has since ceased, may only be detected by measuring levels in the hair and nails, with limited reliability. Arsenic accumulates in these tissues due to binding with keratin. Blood arsenic levels are usually not helpful unless there is active exposure. Nerve conduction studies reveal low amplitude or absent sensory responses. There is mild slowing of motor conduction velocities indicating loss of large, fast-conducting myelinated axons. EMG reveals active and chronic denervation in distal muscles.7

Clinical considerations. The manifestations of arsenic toxicity depend on the level of exposure. Acute high-level exposure results in rapid onset of severe abdominal pain, vomiting and diarrhea. Cardiovascular effects include tachycardia, hypotension and vasomotor collapse with possible death. Poisoning may also cause CNS dysfunction that may be transient (psychosis, somnolence or stupor) or prolonged (behavioral and cognitive problems).5 If an individual survives acute high-level exposure, a neuropathy

Treatment. Acute arsenic toxicity may be life threatening and treatment in an intensive care setting with aggressive fluid and electrolyte resuscitation is indicated with cardiovascular compromise. The primary consideration in chronic arsenic toxicity is terminating exposure. Removal of arsenic from the body is facilitated by chelation therapy. The agents used are British antilewisite (BAL or dimercaprol) and dimercaptosuccinic acid (DMSA). BAL is a dithiol and allows excretion of arsenic by formation of a

Arsenic

Toxic and Occupational PNS Disorders 671 non-toxic ring. The treatment should be started as soon as possible after poisoning and continued for several months. Unfortunately, the fully developed neuropathy is unlikely to respond to therapy.8

Lead Lead toxicity has been a common problem in the past. Although elimination of lead-based paints and other environmental sources of contamination has reduced the frequency of lead intoxication, exposure may still occur, especially in the industrial setting. Occupational exposure has been reported in battery manufacturing, smelting plants, demolition, automobile radiator repair and working in indoor gun firing ranges. Paint ingestion is still a source of lead toxicity as is drinking ‘moonshine’ whiskey and burning batteries for heat. Both organic and inorganic lead causes toxicity. Lead gains access to the tissues via ingestion, inhalation or dermal contact. Lead interacts with carboxyl, sulfhydryl, amino and phosphate groups.9 This leads to disruption of the heme biosynthetic pathway and deficient activity of cytochromes, which are important for detoxification of harmful free radicals. Inorganic lead displaces calcium ions, disrupts ion transport through calcium channels, inhibits calcium adenosine triphosphatasae activity, and results in accumulation of intracellular calcium. These factors likely play a role in the neurologic toxicity of lead.

Clinical considerations. Children most commonly display CNS dysfunction from lead intoxication. This may present as chronic cognitive dysfunction (developmental delay or loss of milestones) or an acute encephalopathy (see Chapter 28.1). Encephalopathy is also seen with acute, highlevel exposure in adults; in both age groups dysfunction can progress to seizures, coma, or death.10 Although there may be some cognitive and behavioral dysfunction in adults with chronic, low-level, lead exposure, this is uncommon. Peripheral neuropathy resulting from chronic lead exposure is most commonly seen in adults and rarely occurs in children. Lead neuropathy develops insidiously with chronic exposure. The manifestations are unusual in that motor dysfunction predominates and there may be few, if any, sensory symptoms or signs. The pattern is distal, symmetric weakness with atrophy and loss of deep tendon reflexes and occasionally fasciculations. The arms are involved preferentially in many cases and in generalized cases become affected relatively early. Although the older literature contained reports of focal neurologic deficits such as wrist drop, manifestations such as these are less common. It is thought that these may be due to secondary compression neuropathies. Although lower motor neuron damage is associated with lead exposure, no causal relationship has been demonstrated between lead exposure and development of idiopathic motor neuron syndromes such as amyotrophic lateral sclerosis.11 Chronic lead exposure causes systemic toxicity in addition to neuropathy. Microcytic, hypochromic anemia is often seen, as are GI disturbances including constipation and abdominal pain. Other less common manifestations include renal dysfunction, fatigue, weight loss, and occasionally gout.

Diagnostic considerations. The electrophysiologic findings in lead neuropathy are controversial. In animal models, demyelination has been demonstrated, while in humans the physiology is generally axonal. Electromyographic evidence of active denervation and chronic motor unit reinnervation in distal muscles reflect the presence of axonal degeneration. Nerve conduction studies show evidence of sensory axon loss, even though sensory symptoms are minimal or absent. Compound motor action potential amplitudes are reduced in more severe cases. Nerve conduction studies may be abnormal before the appearance of symptoms.12 The degree of abnormality correlates with the lead burden in the body.13 Somatosensory evoked potential (SSEP) amplitudes are correlated with blood lead levels. SSEP latencies have been reported to be prolonged in both central and peripheral nerve segments. Neuropsychologic testing reveals abnormalities in memory, attention and visuospatial functioning. Confirmation of current or past lead exposure is usually possible. Laboratory evaluation reveals a microcytic, hypochromic anemia with basophilic stippling of erythrocytes. Urinary lead levels can be measured. Chelating agents that draw lead from soft tissues and allow for its excretion may increase the diagnostic senstivity; however, they are not usually necessary for diagnosis. The ratio of the micrograms of lead excreted to milligrams of calcium ethylenediamine tetra-acetic acid (CaEDTA) administered should not exceed 0.6 and the 24-hour urinary lead level after chelation therapy (usually with CaEDTA) should not be greater than 1 mg.14 Treatment. The initial step is removal of the affected individual from further exposure. Treatment of lead intoxication is based on chelation. The goal is the mobilization of lead from the bone to allow elimination. Penicillamine, succimer, CaEDTA and BAL administered in short course are all effective chelating agents and improvement usually begins within 2 weeks (see Chapter 39.8). Oral agents (e.g., succimer) are usually adequate for milder cases, while combination therapy with EDTA and sometimes in combination with BAL is recommended for more severe intoxication that includes encephalopathy. Diazepam should be provided for treatment of seizures associated with lead toxicity although they are often refractory. In those cases that involve brain edema, mannitol, hyperventilation and fluid restriction should all be utilized to lower intracranial pressure. The mortality rate is high in those presenting with seizures and encephalopathy. Complete recovery from neuropathy is usual except in severe cases. The improvement typically begins 2 weeks after initiation of chelation therapy.

Mercury Elemental mercury is used in thermometers, barometers and other gauges. Organic mercurial compounds have been used as a preservative in latex paints, in various disinfectants and are also used as industrial catalysts. Inorganic mercurial salts and elemental mercury are present in the

672 Disorders of the Peripheral Nervous System manufacture of chlorine, in dental amalgams, and in the natural gas industry. Outbreaks of mercury toxicity have been reported due to contaminated fish in Minamata Bay, Japan15 and in Iraq related to application of an organic mercury-containing fungicide to grain.16 Elemental mercury vapor is absorbed by inhalation. Mercury salts are absorbed through the skin and GI tract, and organic mercury is usually absorbed through the GI tract. While there is evidence that inhalation of low levels of mercury vapor occurs with mercury amalgam dental fillings, raising health concerns, no causal relationship between the presence of mercury amalgam dental fillings and clinical toxicity has been established to date.17

ication has been reported to show a motor greater than sensory axonal neuropathy.20 Nerve conduction studies may reveal evidence of a developing neuropathy in exposed workers prior to symptom onset.19

Treatment. Mercury intoxication is primarily treated by removal from exposure. Although excretion of mercury in the urine is increased by chelation with agents such as penicillamine or dimercaprol,21 it is not clear that this speeds recovery from the toxic effects. The prognosis is good for complete recovery in most cases, depending on severity of illness.

Thallium Clinical considerations.

The form of mercury that one is exposed to determines the pattern of nervous system involvement and whether there are associated systemic symptoms. Although CNS dysfunction dominates, there are some reports of PNS effects. Elemental mercury is very lipid soluble and tends to cause more CNS involvement with little or no systemic signs. Low-level toxicity due to elemental mercury, known as micromercurialism, causes tremor, fatigue, GI dysfunction, anorexia and weight loss. Continued exposure leads to more prominent tremor that may involve the head, face and even the eyelids. Personality change, hyperexcitability and insomnia are other possible effects of chronic elemental mercury exposure. Mercury salts also cause systemic toxicity including GI symptoms and nephrotic syndrome. PNS involvement has been reported with exposure to all forms of mercury. The relationship of mercury exposure to neuropathy is better established with mercury vapor than with organic mercury compounds. Mercury vapor exposure is associated with a subacute, motor neuropathy that may be confused with Guillain–Barré syndrome. However, electrophysiology reveals damage to motor axons rather than demyelination. There is usually a preceding irritation of the upper respiratory tract that may be mistaken for an infection. As with other forms of mercury exposure, mercury vapor may also cause neuropsychologic dysfunction. Organic mercury compounds such as methyl mercury cause tremor, hearing loss, constriction of visual fields, mental impairment and dysarthria with prolonged exposure. The other prominent symptom seen more often with organic mercury toxicity is sensory ataxia. This is thought to be due to damage of dorsal root ganglion neurons. Sensory neuron damage may be heralded by distal paresthesias, which progress more proximally and may involve the tongue.18 Complex organic mercurials are also associated with nephrotoxicty.

Diagnostic considerations.

Diagnosis of mercury intoxication is based on eliciting the appropriate exposure history in the setting of a neurologic syndrome as described above. Mercury levels can be measured in urine, blood and hair with limited reliability in the latter. Blood mercury level is a good indicator of recent exposure while urine measurements including serial measurements (after administering penicillamine) are better for chronic exposure.19 Electrodiagnostic testing in cases of mercury intox-

Thallous salts were commonly used as pesticides and rodenticides in the past. Although rarely used now, accidental (mostly children) or homicidal poisoning continues to occur. Occupational thallium exposure is usually low-level and chronic, rather than acute and high level. Occupations that pose some risk include smelting plants, mines and cement factories. Consumption of contaminated food and water may also be a source of intoxication. Thallium is absorbed through the GI tract, by dermal contact or by inhalation. Thallium is distributed throughout the body as potassium would be and substitutes for it in reactions.22,23 However, the mechanism of toxicity has not been clearly elicited.

Clinical considerations.

Thallium intoxication causes a distal, symmetric sensory greater than motor peripheral neuropathy. Pain is a prominent feature. Degeneration affects the distal portion of the longest axons initially and the large diameter sensory fibers are most susceptible. Involvement of small unmyelinated fibers may lead to a delayed autonomic neuropathy.24 The manifestation of thallium neuropathy depends on the temporal course and intensity of exposure. The most common scenario is a massive, acute ingestion. Vomiting, abdominal pain and diarrhea occur within hours but may be delayed up to a day. Severe, burning distal paresthesias in the legs develop within 2–5 days and are accompanied by intense joint pains. Large and small fiber sensory modalities are affected and the sensory abnormalities include the hands and trunk relatively early. Although weakness is not usually a prominent complaint, it is frequently present on examination. Surprisingly, deep tendon reflexes tend to be preserved early in the disease process in spite of large fiber sensory loss. Cardiac and respiratory failure leading to death may also occur with massive ingestion. The mental status may decline to lethargy or coma. Although alopecia is the classic sign of thallium intoxication, it does not appear until approximately 15–39 days after ingestion and is not helpful in the acute setting. Alopecia is not specific for thallium poisoning and does not always occur. Renal insufficiency and paralytic ileus are other systemic manifestations sometimes seen with acute thallium poisoning. The presentation of an acute neuropathy with abdominal pain needs to be distinguished from acute intermittent porphyria and other metal poisonings (e.g., arsenic and lead). The recovery in acute thallium poisoning tends to be

Toxic and Occupational PNS Disorders 673 incomplete. There is often residual CNS dysfunction. The recovery of the peripheral nerves is slow and there is often persistent sensory loss. In those with severe motor axon damage, weakness may also persist. With a smaller initial ingestion, thallium neuropathy may present subacutely. Subacute thallium neuropathy evolves more slowly, beginning more than a week after exposure. Alopecia, hyperkeratosis and Mee’s lines are more common. Other neurologic features reported with subacute thallium intoxication include cranial neuropathies, chorea and ataxia.22 The neuropathy is characterized by defects in all sensory modalities. Walking may be affected early on because of painful paresthesias in the feet. Although distal weakness is usually detected, the patient does not often complain of it and it is usually not severe. The deep tendon reflexes are slightly reduced or normal. Autonomic dysfunction may lead to hypertension and tachycardia. Subacute neuropathy tends to have a much better prognosis. With termination of exposure, most patients recover within 6 months. Hair regrowth begins earlier at about 10 weeks after withdrawal.

Diagnostic considerations.

Thallium levels can be measured in urine, blood or body tissues. Sensitive methods capable of detecting microgram quantities are available. Although established blood levels that indicate toxicity are available, they do not reflect the total body burden, as thallium is sequestered in tissue. In cases without alopecia, a potassium chloride challenge will cause the urinary excretion of thallium to rise even when the baseline level in blood and urine is normal. Cerebrospinal fluid protein is normal in cases of thallium neuropathy. Nerve conduction studies demonstrate reduced sensory potential amplitudes with conduction velocity slowing. EMG shows evidence of acute and chronic motor axon degeneration. The electrophysiologic abnormalities can be used to follow the degree of peripheral nerve damage.25

Treatment. GI elimination is enhanced by administration of Prussian blue or activated charcoal, which bind thallium in the gut.26,27 Laxatives are also helpful since constipation and paralytic ileus are frequently present. Forced diuresis or administration of potassium chloride enhances urinary excretion. Thallium has a half-life of 30 days in the body if no measures are taken to promote excretion.

Pesticides, herbicides and rodenticides Organophosphates Organophosphorous (OP) compounds are used as insecticides, antioxidants, petroleum additives, flame retardants, lubricants and plastic modifiers. The best characterized peripheral neuropathy caused by these compounds in humans is that due to triorthocresylphosphate (TOCP), although exposure to other OP such as parathion, chlorpyrifos, mipafox, trichlorfon and leptophos may cause a similar neuropathy. Intoxication is usually due to acci-

dental exposure from agricultural pesticide spraying. Individuals affected may be those mixing or applying the pesticide or those working in the fields shortly after spraying. OPs are absorbed through the respiratory and GI tracts as well as through the skin.28 The action of OP is to irreversibly inhibit acetyl cholinesterase (AChE) in erythrocytes and nervous tissue by phosphorylation. Acetylcholine accumulates due to lack of degradation, leading to excessive stimulation of both muscarinic and nicotinic receptors. Most OP esters are quickly degraded in the environment. Pesticides containing OP are also intentionally ingested in suicide attempts.

Clinical considerations. The acute or type I OP syndrome is primarily due to excessive muscarinic cholinergic stimulation.29 The effects are frequently seen within hours of exposure and are always present within one day. The specific OP and the degree of exposure determine the intensity of the acute syndrome. Characteristic symptoms include nausea, vomiting, diarrhea, bronchospasm, bradycardia, sweating, salivation, and micturition (see Chapter 48) . Extreme intoxication leads to CNS involvement with fatigue, nervousness, emotional lability, decreased alertness, cognitive impairment, convulsions and coma (see Chapter 28.1). Prior OP exposure may increase susceptibility to developing the acute syndrome on subsequent exposure because of a decrease in functional AChE. The acute syndrome may have prominent mental status changes and other drug or toxin ingestion needs to be considered. The type II or intermediate OP syndrome is the result of overstimulation of nicotinic ACh receptors in skeletal muscle.30 The onset occurs within 12–96 hours of exposure. There may be a symptom-free interval of 1–4 days between the acute and intermediate syndromes. The initial feature is usually respiratory insufficiency. Proximal muscle and neck flexor weakness follow and distal extremity strength is usually preserved. Cranial muscles, including extraocular muscles, may be involved. Sensory function is unaffected. Dystonic posturing is occasionally seen. Recovery begins in the cranial muscles 5–15 days after exposure. It then proceeds from the respiratory muscles to the proximal muscles and lastly the neck flexors. Since atropine is specific for muscarinic receptors, it does not prevent or treat the intermediate syndrome. The differential diagnosis of the intermediate syndrome includes Guillain–Barré syndrome, periodic paralysis or a severe attack of myasthenia gravis. Regardless of whether exposure has caused the type I or type II syndromes, a central-peripheral axonopathy may develop with exposure to some OPs. Central-peripheral axonopathy refers to a process that affects the distal portions of peripheral axons initially, but, with continued exposure, leads to damage of the distal portions of central axons. The delayed appearance of this neuropathy, 7–21 days after exposure, led to the phrase organophosphateinduced delayed polyneuropathy (OPIDP). This neuropathy is not related to the inhibition of AChE but rather to a distinct esterase localized to nervous tissue, termed neuropathy target esterase. Although the OPIDP is less frequent than

674 Disorders of the Peripheral Nervous System the cholinergic syndromes, it causes significant morbidity. Most agricultural OPs do not cause the OPIDP and those that cause subtle cholinergic symptoms seem more likely to cause delayed neuropathy. OPIDP frequently occurs in the setting of low-level, chronic exposure. Although most toxic central-peripheral axonopathies are chronic, OPIDP is subacute. Symptoms are usually maximal within two weeks after onset. Initial manifestations include painful paresthesias in the feet and cramping pain in the calf muscles. Motor symptoms and signs are prominent and there is early weakness of the leg muscles including foot drop. The intrinsic hand muscles become involved next and the proximal muscles are spared until later in the course. Sensory loss can usually be detected with careful examination. Ataxia may be present, that is more severe than expected based on the degree of sensory loss and weakness. Although the ankle reflex is typically absent, the activity of the other reflexes is variable. The OPIDP needs to be distinguished from other toxic causes of a central-peripheral distal axonopathy.

Diagnostic considerations. Electrophysiologic testing is a very sensitive indicator of OP peripheral neurologic effects.31 Shortly after OP exposure, a single stimulus produces spontaneous repetitive motor action potentials (SRMAPs) following the initial compound motor action potential.32 While the ability to elicit SRMAPs is a sensitive marker of OP exposure, there is no correlation with the degree of intoxication. Once muscle weakness becomes evident, repetitive nerve stimulation produces decremental responses. In contrast to myasthenia gravis where the decrement is usually maximal by the fourth response, here it is most prominent by the second. With mild OP intoxication, rapid rates of stimulation may be necessary to demonstrate decrement, and there may be a subsequent incremental response. With high-level exposure, decrement is evident with slow rates of stimulation and SRMAPs may be absent.31 OPIDP is characterized electrophysiologically as a sensorimotor axonal neuropathy. Sensory nerve conduction abnormalities appear earlier and more prominently than motor nerve abnormalities, despite clinical motor symptoms. Sensory nerve action potentials are reduced in amplitude or absent, while motor conduction studies are normal or reveal minimal slowing of conduction velocity.33 Needle EMG reveals evidence for acute and chronic denervation in the distal limb muscles. In the intermediate syndrome, EMG is normal, despite prominent muscle weakness. Routine clinical laboratory findings are usually normal. Recent exposure to OP causes reduction in erythrocyte AChE levels. AChE levels less than 20% of baseline values are frequently associated with severe weakness. The wide range of normal erythrocyte AChE levels makes a single determination difficult to interpret. Serial measurements showing progressive decline in activity are more useful. The CSF protein level in OPIDP is normal or only mildly elevated.

erated by 1 week after exposure, this is usually sufficient for functional recovery. Cognitive and behavioral abnormalities may persist after recovery from the acute syndrome. Patients with mild OPIDP usually have excellent recovery. With more severe initial deficits, residual deficits such as claw hand deformity, foot drop or atrophy may persist. Damage of central axons may only become apparent after recovery from the peripheral neuropathy. Injury of distal portions of the corticospinal and spinocerebellar tracts may lead to spasticity or ataxia.

Vacor N-3-pyridylmethyl-N-p-nitrophenyl urea (PNU) or vacor is a rodenticide that is structurally related to nicotinamide. Accidental or intentional ingestion is the most common form of exposure. This leads to a severe acute distal axonopathy with significant autonomic involvement. Vacor also damages the beta cells of the pancreas leading to acute diabetes mellitus. Massive ingestion leads to limb weakness and impairment of postural reflexes within the first hour. This rapidly progresses to severe generalized weakness including the cranial nerve innervated muscles, urinary retention and diabetic ketoacidosis. Those that survive the acute ingestion frequently continue to have endocrine and autonomic dysfunction. Motor weakness improves gradually over the course of months. The few autopsy studies reported have demonstrated wallerian-like degeneration in the peripheral nerves, roots and dorsal root ganglia.34 Animal studies of vacor-induced neuropathy indicate defects of fast axonal transport in distal nerves.35 There is also abnormal morphology of the neuromuscular junction. This impairment may explain the rapid onset of weakness. Although the precise biologic mechanism of vacor neuropathy is unclear, it can be prevented experimentally by administering nicotinamide.

Solvents Carbon disulfide Carbon disulfide is a clear liquid, which is a vapor at room temperature. Carbon disulfide is absorbed by inhalation, dermal contact and, although less common, can also be absorbed through the GI tract. It is used in the production of cellophane films and viscose rayon fibers (see Chapter 40). It is also a major metabolite in the breakdown of the drug disulfiram (antabuse), which is used as a deterrent for alcohol abuse. The toxicity of this compound is most likely related to its reactivity with amine, sulfhydryl and hydroxyl groups, which results in the formation of reactive sulfur atoms. Isothiocyanates are produced, which can covalently bind to and cross-link cytoskeletal proteins such as neurofilaments. This may be related to the formation of giant axonal swellings seen in experimental studies performed in rats. These are similar morphologically to those seen in hexacarbon and acrylamide neuropathies.

Treatment.

The acute and intermediate OP syndromes have a good prognosis as long as there is adequate supportive care. Although only 40–60% of AChE content is regen-

Clinical considerations. Acute or subacute high-level exposure results primarily in CNS dysfunction including

Toxic and Occupational PNS Disorders 675 confusion, hallucinations, memory impairment, and emotional lability (see Chapter 28.1).36 Chronic, low-level exposure causes a combination of peripheral neuropathy and CNS abnormalities. The neuropathy may be asymptomatic and only detected by electrophysiologic testing when exposure is low but, as the concentration increases, a progressive sensorimotor distal polyneuropathy emerges. This neuropathy produces distal numbness and weakness along with painful cramping in the legs. The knee and ankle reflexes are often absent. Continued exposure leads to involvement of the arms. CNS manifestations of prolonged exposure include headache, dizziness, depression, memory impairment and impaired sexual arousal. Exposed persons may also have extrapyramidal signs of tremor, bradykinesia, and cogwheel rigidity, as well as hemiparesis or spasticity.

Diagnostic considerations. Although CS2 is difficult to measure, urinary levels of its metabolite, 2-triothiazolidine-4-carboxylic acid, have proved to be a sensitive measure of exposure. Nerve conduction studies reveal slowing of conduction velocities and prolongation of motor and sensory latencies. Needle electromyography reveals active and chronic denervation in distal leg muscles, reflecting axonal degeneration and chronic motor unit reinnervation.37,38 The spinal fluid profile is unremarkable. Treatment. Early cessation of exposure is the key to prevent further decline. There may be a role for the use of pyridoxine to treat the neuropathy. The effects may be related to the reactivity of CS2 with pyridoxine. The degree of recovery depends on the severity of dysfunction at the time of removal from exposure. In mild cases almost complete recovery from neuropathy and most of the CNS abnormalities can be expected. Occasionally, CNS recovery is incomplete, probably due to residual spinal cord damage. Severe neuropathy may not completely recover, and as many as a third may have symptoms and signs of neuropathy 10 years later.38

n-Hexane and methyl-n-butyl ketone These compounds are clear, colorless, volatile liquids used as solvents (see Chapter 40). They are metabolized to the toxic compound 2,5 hexanedione.39 Hexacarbons gain entry to the body via inhalation, dermal contact and, rarely, ingestion. Exposure occurs in the petroleum production and refining industries. They are also components of lacquers and glues, which are widely used in the shoe and cabinet making industries. Methyl-n-butyl ketone (MnBK) use in the manufacturing of plastic-coated and colorcoated fabrics prompted an epidemic of peripheral neuropathy.40 Intentional inhalation (glue sniffing) is also a cause of high-level hexacarbon exposure. Methyl ethyl ketone (MEK) is not significantly neurotoxic by itself, but is present in many solvent mixtures with n-hexane and MnBK, and may potentiate their neurotoxicity.

Clinical considerations. Isolated high-level acute exposure causes CNS depression and narcosis. However, repeated

massive exposure such as glue sniffing leads to a subacute, predominantly motor neuropathy with cranial nerve dysfunction.41 This neuropathy may be associated with autonomic dysfunction including impotence, hyper- or anhidrosis, and vasomotor instability. Chronic exposure to lower levels results in a slowly developing central-peripheral axonopathy that affects the sensory and motor systems in a length dependent fashion.42 Nerve fiber degeneration affects the distal portions of peripheral axons first, but with continued exposure leads to damage of distal corticospinal, dorsal column, and other central pathways. All sensory modalities are affected, beginning in the feet with dysfunction gradually progressing proximally with continued exposure. Numbness rarely progresses higher than the knee in spite of the eventual involvement of the hands, and pain is not a complaint. Although ankle reflexes are lost early in the course, the other reflexes are usually spared. Distal leg and arm weakness and atrophy occur with continued exposure. In severe cases, the neuropathy is complicated by malaise, weight loss, abdominal pain, and leg cramps. Worsening of symptoms after removal from exposure, a phenomenon described as ‘coasting,’ is common.

Diagnostic considerations. Electrophysiologic testing demonstrates a sensorimotor axonal neuropathy.43 Active denervation is detected in distal muscles often before nerve conduction studies become abnormal. In severe cases, nerve conduction studies reveal marked slowing of distal motor conduction velocities, which is an unusual finding in other toxic neuropathies. Asymptomatic workers employed in factories where cases of solvent polyneuropathy have occurred have been demonstrated to have slowing of conduction velocities. The nerve biopsy reveals a very characteristic morphologic abnormality known as giant axonal swelling. This arises as a result of the accumulation of neurofilaments.44 Giant axonal swellings are also seen with acrylamide and carbon disulfide exposure, and in the genetic giant axonal neuropathy. The accumulation of neurofilamentous material is most prominent at the paranodal region, and is likely related to cross-linking and disruption of axonal transport. Spinal fluid protein is usually normal unless the nerve roots become involved, in which case CSF protein may be elevated. Because of the subacute presentation and the presence of conduction slowing, the neuropathy associated with repeated high-level hexacarbon exposure needs to be differentiated from Guillain–Barré syndrome. Treatment. The only treatment is cessation of exposure to allow recovery. As mentioned above, symptoms frequently continue to worsen for 1–4 months after cessation of exposure. The ‘coasting’ phenomenon is possibly due to the lipid solubility of hexacarbons, which delays their elimination from the body. Measurement of 2,5hexanedione in the urine can help detect exposure before significant toxicity occurs. Recovery depends on the severity of neuropathy. Patients with mild neuropathy usually make a complete recovery in less than a year. Residual

676 Disorders of the Peripheral Nervous System distal atrophy, weakness and sensory loss are not uncommon with severe neuropathies. The effects of central damage (spasticity, long-tract weakness) may only become evident after resolution of the neuropathy.

Trichloroethylene The solvent trichloroethylene (TCE) is used as a degreasing agent, a cleaner for photographic equipment and lenses, and in the extraction of fats and oils from vegetables (see Chapter 40). It is also used in the dry cleaning and rubber production industries, and previously was used as an anesthetic agent. Toxicity may be related to a breakdown product, dichloroacetylene (DCA), rather than TCE itself. TCE is usually absorbed by inhalation and is very lipid soluble.

Clinical considerations.

Acute exposure is the most common scenario and trigeminal nerve dysfunction is a frequent manifestation. The sensory loss is usually in the distribution of all three trigeminal divisions and there may be weakness of mastication. Other cranial neuropathies affecting the facial, optic, oculomotor and glossopharyngeal nerves have been reported. There is a questionable relationship to a distal sensorimotor peripheral neuropathy. Acute exposure to TCE also causes CNS dysfunction with altered mental status and ataxia. Chronic exposure has also been reported to cause cognitive dysfunction and peripheral neuropathy.

Diagnostic considerations. Trigeminal somatosensory evoked potentials have been used to measure the effects of TCE exposure. Abnormalities are reported in asymptomatic individuals exposed to TCE. Slowing of nerve conduction velocity in facial, trigeminal (blink) and extremity peripheral nerves has been reported. Pathologically, axon degeneration, cell loss in the cranial nerve nuclei, and myelin degeneration have been reported in autopsy studies. Treatment. Removal of the individual from the exposure source is the first step. If ingested, lavage should be performed. The mental status changes usually resolve rapidly but facial numbness tends to persist. In one case, long-term follow-up 18 years later revealed continued facial sensory disturbance.

Allyl chloride Allyl chloride is a reactive halogenated hydrocarbon that is used in the manufacture of glycerin and epoxy resin. Exposure to high levels for long duration results in a distal symmetric neuropathy. This is gradual in onset and presents with distal numbness and weakness along with reduced ankle reflexes.45 Nerve conduction studies reveal prolonged distal latencies. EMG demonstrates active and chronic denervation in distal muscles. Cessation of exposure usually results in good recovery. Animal studies reveal abnormal accumulation of neurofilaments. There is also degeneration in the distal terminals of both peripheral and central axons.

Mixed solvents A variety of solvent mixtures, including toluene, have been associated with electrophysiologic changes in peripheral (particularly autonomic) nerve function. Further research is require to assess the clinical and pathological impact of solvents on the peripheral nervous system. Electrophysiologic studies and trial removal from exposure should be considered in exposed individuals with persistent neuropathic symptoms.

Plastics Acrylamide Acrylamide is used in grouting agents for soil and sealing applications, and polyacrylamide is used as a flocculator in waste water treatment plants. Although acrylamide monomer is the toxic form, polyacrylamide (which is innocuous) may be contaminated by up to 2% monomer and therefore may be a source of toxicity. Absorption takes place by dermal contact, inhalation or ingestion. Acrylamide neuropathy is a popular experimental animal model for studying the processes of axonal transport, dyingback neuropathy and axonal swelling. Acrylamide appears to interfere with axonal transport46,47 resulting in an accumulation of neurofilaments and axonal swelling.48,49 The swelling is most prominent in the paranodal region, possibly due to the constriction of the axon at that point.

Clinical considerations. The manifestations of acrylamide toxicity depend on the degree and duration of exposure. The usual route of exposure is through the skin and thus a contact dermatitis is usually present prior to the clinical symptoms of neuropathy. In the setting of acute exposure, malaise, dizziness, anorexia and headache are often present. With high-level acute exposure, the neurologic picture includes encephalopathy with seizures and truncal ataxia followed by peripheral neuropathy. Early behavioral changes may be less apparent to the patient than others. In chronic, low-level exposure, the dermatitis persists but the CNS effects are not as prominent. The neuropathy resulting from acrylamide is a classic example of a central-peripheral distal axonopathy.50 Initial clinical manifestations include toe numbness and widespread hyporeflexia. Large fiber sensory dysfunction with loss of vibration and proprioception is common while pain and paresthesias are rare. Acute, high-level exposure often results in widespread autonomic dysfunction such as impairment of reflex changes in heart rate and blood pressure, vasomotor changes in fingers and toes, and excessive sweating. Overt autonomic dysfunction rarely occurs in chronic exposure and may be limited to excessive sweating of the hands and feet. Although sensory complaints dominate, motor and cerebellar deficits may be evident on physical examination. Cranial nerve function is unaffected. Diagnostic considerations. Neurophysiologic testing reveals reduced amplitude sensory responses with preservation of motor amplitude and conduction velocity.51

Toxic and Occupational PNS Disorders 677 These findings are characteristic of a distal axonopathy and in some instances the electrophysiologic abnormalities may precede the development of symptoms. The sural nerve biopsy correlates with the physiologic and clinical manifestations, showing reduced numbers of large diameter, thickly myelinated fibers.

Treatment. Preventing further exposure to acrylamide is the primary treatment modality. Acute ingestion should prompt gastric lavage to reduce levels of intoxication. Liver and renal failure may complicate recovery. Removal from exposure usually results in recovery if the neuropathy is mild. Some residual loss of vibratory sensation may be apparent. However, in the case of severe neuropathy, spasticity, ataxia, more profound sensory dysfunction and memory problems may remain. Central nervous system dysfunction, such as spasticity and upper motor neuron weakness may be obscured initially by peripheral nerve dysfunction. As nerve recovery ensues, clinical dysfunction may remain due to unresolved central nervous system dysfunction. Coasting, the worsening of symptoms after termination of exposure, may occur.

Dimethylaminoproprionitrile (DMAPN) DMAPN had been used as a catalyst in polymerization reactions until 1978. It was found to be responsible for an outbreak of toxic axonopathy in the production of polyurethane foams. Since that time, it is no longer used and no additional cases have been reported.

most cases, the sensory potential amplitudes are reduced in the lower extremities. Severe cases may demonstrate slowing of motor conduction velocity.

Treatment. Removal from exposure is the only form of treatment. The prognosis for recovery is good in young patients. Older individuals tend to have more persistent bladder and sexual dysfunction.

Gases Methyl bromide Methyl bromide is used in insecticides, fire extinguishers, refrigerants and fumigants. Acute, high-level exposure may be fatal. Chronic, high-level exposure to methyl bromide causes multifocal neurologic dysfunction involving the peripheral nerves, pyramidal tracts and cerebellum. The neuropathy usually appears after 3–7 months of exposure and is a distal sensorimotor polyneuropathy. Paresthesias in the distal extremities are the first symptoms. Pain and ataxia develop later. The optic nerves may be involved and color vision loss may reveal toxic exposure to methyl bromide at an early stage. Electrophysiologic testing reveals a distal, motor predominant neuropathy.53 Sural nerve biopsy reveals loss of large myelinated axons. The cerebrospinal fluid is normal. The prognosis appears to be good in most cases, although recovery may take 6–8 months.

Ethylene oxide (EtO) Clinical considerations. The peripheral neuropathy associated with DMAPN is very distinct in that there are prominent urinary symptoms that manifest prior to the onset of sensory or motor complaints.52 The degree of exposure determines the rapidity of onset of symptoms. The exposed individual often complains initially of urinary hesitancy and abdominal pain. This progresses to reduced frequency of urination, decreased urinary stream and incontinence. Sexual dysfunction follows with partial or complete impotence. About the same time, sensory symptoms develop in the feet. The sensory symptoms progress to involve the proximal legs and the hands while weakness develops in the distal legs. Abnormality of sensation to pain, temperature and touch in the sacral dermatomes is characteristic of DMAPN neuropathy. There is also loss of distal vibratory sensation but the reflexes are surprisingly well preserved. The preservation of reflexes, autonomic features and preferential loss of pain and temperature sensation suggest more prominent involvement of the small nerve fibers. However, the available morphologic data have not borne this out. Other than bladder and sexual abnormalities, other autonomic function is preserved. The cranial nerves are not affected. The differential diagnosis is primarily that of diabetic small-fiber neuropathy or amyloid neuropathy.

Diagnostic considerations. Urodynamic studies reveal hypocontractility of the bladder, consistent with autonomic nerve dysfunction. In mild cases of DMAPN neuropathy, electrodiagnostic testing may be normal. In

EtO is used in the sterilization of medical equipment, particularly that which is heat sensitive. After sterilization, the object must be allowed to aerate or residual EtO may be present. EtO is used in the production of ethylene glycol (antifreeze) and other chemical compounds, as well as the production of polyester fibers and polyethylene films. As this is a gaseous compound, the primary route of exposure is via inhalation.

Clinical considerations. Acute exposure causes respiratory tract and mucous membrane irritation as well as nausea, vomiting, headache, dizziness and anorexia. EtO exposure results in a distal sensorimotor axonopathy.54 This may be seen after long duration, low-level exposure or with subacute, higher level exposure. Clinical symptoms include distal numbness and weakness, incoordination and ataxia. Sensory loss begins in the feet and, with continued exposure, begins to involve the hands. Weakness also begins distally and there is diffuse reduction of reflexes with the ankle jerks usually absent. Chronic exposure also causes CNS dysfunction including memory and concentration problems, as well as dysarthria and increased tone. Removal from exposure is the only therapy. There is usually a gradual recovery from the neuropathy once exposure is terminated. Diagnostic considerations.

Electrophysiologic testing is consistent with a distal axonopathy. EMG reveals active and chronic denervation changes in distal muscles. Nerve

678 Disorders of the Peripheral Nervous System conduction studies demonstrate reduced amplitude sensory and motor potentials with normal conduction velocity or only mild slowing. Nerve conduction abnormalities may appear prior to the onset of symptoms in exposed individuals. Sural nerve biopsy shows evidence of axon degeneration.

Polychlorinated biphenyls (PCBs) PCBs have previously been used in electrical insulation as well as plasticizers (see Chapter 45). Although there is a lot of discussion about the toxicity of PCBs, neurotoxicity associated with specific congeners has not been well characterized. The use of PCBs was banned in the 1970s but contaminated products remain. Although PCB contamination of waterways and marine life is the most common source of exposure, there are no reported cases of toxicity as a result. Neuropathy caused by PCBs has been reported in an outbreak caused by contaminated cooking oil in Taiwan. Most cases reported more recently have been electrical workers involved in the removal of PCB contaminated transformers. PCBs are very lipid soluble and may remain in the tissues long after removal from exposure.

Clinical considerations. The peripheral neuropathy associated with PCB exposure is a distal symmetric sensorimotor neuropathy.55 Encephalopathy is usually coexistent with neuropathy. Worsening of symptoms after termination of exposure has been described. Removal from exposure usually results in improvement. Most patients have some persistent deficit several years later. Diagnostic considerations.

Nerve conduction studies demonstrate significant slowing of sensory and motor nerves. Although PCB levels in the blood can be obtained, they do not correlate with the presence of neurologic symptoms.

FOCAL NERVE INJURIES

Entrapment may also occur in the region of the supinator muscle by fibrous bands on the surface of the muscle (arcade of Frohse) or within the muscle itself by tumors, ganglia, or inflammatory conditions. A painful elbow may result from entrapment of the posterior interosseous nerve. The superficial radial nerve is vulnerable to a host of injuries, including lacerations or compression from tight wristwatches, handcuffs, or plaster casts, owing to its superficial position at the wrist.

Posterior interosseous nerve compression The posterior interosseous nerve may be compressed as it enters the supinator muscle under a fibrous arch (arcade of Frohse) or within the substance of the muscle.56 Common causes of injury include ganglia, tumors (especially lipomas), fibrous bands, fractures or dislocations of the radius. The condition may occur spontaneously without an identifiable cause. Many so-called idiopathic cases have been found to be the result of entrapment at the arcade of Frohse. The clinical syndrome of ‘resistant’ tennis elbow is, in many cases, a result of posterior interosseous nerve entrapment, most often by the arcade of Frohse. It often is misdiagnosed as tennis elbow caused by lateral epicondylitis because both conditions result in lateral elbow pain. Posterior interosseous nerve entrapment is characterized by pain on firm palpation of the extensor muscle mass distal to the radial head; in contrast, lateral epicondylitis has marked tenderness with palpation of the lateral epicondyle. Posterior interosseous nerve lesions result in an inability to extend the fingers and abduct the thumb; wrist extension is usually less affected as a result of sparing of the extensor carpi radialis. Radial deviation may occur with wrist extension because of extensor carpi ulnaris weakness. Sensation is spared over the dorsum of the hand. The degree of weakness is variable and may not affect all distal radial muscles to the same extent. In patients with rheumatoid arthritis, posterior interosseous nerve lesions must be distinguished from rupture of the extensor tendons to the thumb and fingers.

(See also Chapter 23.3)

Radial nerve Source of injury Most radial nerve injuries result from external trauma. The radial nerve may be damaged in the axilla by lacerations, crutch use, missile wounds, or pressure from the head of a sleeping person (‘lover’s Saturday night palsy’). Proximal nerve injury affects the triceps muscle and weakens forearm extension. The most frequent radial nerve lesion is compression of the nerve against the middle third of the humerus. These are usually Class I lesions with predominantly conduction block. Severe weakness with total paralysis of wrist and finger extensors may occur. Radial nerve entrapment may result from a fibrous band derived from the lateral head of the triceps. This usually occurs in heavily muscled individuals involved in activities requiring prolonged and vigorous elbow extension.

Treatment and prognosis of specific radial nerve injuries Radial nerve injuries in the upper arm.

Most radial nerve lesions resulting in wrist drop should be treated with a cock-up wrist splint, described later, that maintains mild wrist extension. Mild cases, in which weakness is not complete, may need no specific therapy. Complete recovery, over weeks to a few months, generally occurs in most radial nerve lesions resulting from compression. The ultimate degree of recovery depends on the severity and duration of compression. Patients with radial nerve entrapment within the triceps muscle usually recover spontaneously. Rarely, exploration and decompression may be needed in cases of progressive weakness or poor recovery.57 Radial nerve injury secondary to fractures of the humerus have an excellent prognosis for spontaneous recovery. Mild cases of posterior interosseous nerve entrapment often recover spontaneously. A period

Focal Nerve Injuries 679 of observation ranging from 3 to 6 months is generally prescribed to allow spontaneous recovery. More severe cases, in which weakness is present or with poor recovery, may require surgical exploration and removal of identified masses (tumors, lipomas, or ganglia) or constricting bands. Entrapment under the arcade of Frohse, if present, is released. Nerve compression caused by radial fractures should be observed initially for 2–3 months. Regardless of the cause of the radial nerve palsy, most cases of wrist and finger drop require a proper splint to protect against hyperextension of the paralyzed wrist extensor muscles and shortening of the flexor muscles.58 A static cock-up splint, maintaining the wrist at about 15–30 degrees of extension, is usually all that is required in cases with mild wrist extensor muscle weakness. Palmar sensation should not be obscured by a palmar pad. The metacarpophalangeal (MCP) joints may need to be supported in extension, with the thumb extended and radially abducted. Finger joints must be regularly exercised because they rapidly become stiff when immobilized for even a short time. This type of splint supports the hand in a position of function, thereby encouraging continued use. If the hand is able to be opened sufficiently by interphalangeal (IP) joint extension, MCP joint support may be unnecessary. In more severe cases, dynamic splints are useful. Such splints utilize a finger slip around the proximal IP joint to support the MCP joints in slight extension, thereby allowing active finger flexion while offering protective positioning at rest. During the stage of paralysis, daily passive exercises should be used to maintain adequate range of joint movement, including large arm excursion. After voluntary activity returns, specific exercises are employed to strengthen the wrist and finger extensors. Intrinsic hand muscles become weak and inefficient, owing to dependence on the synergistic action of the wrist extensors, and may need strengthening to restore adequate hand function.

Median nerve Source of injury The superficial position of the median nerve in the upper arm renders it vulnerable to missile wounds and lacerations. It is better protected in the forearm where it lies deep within the pronator teres muscle. Median nerve compression has been attributed to crutches pressing on the axilla, tourniquets, rifle sling palsy, and pressure from the head of a sleeping partner on the brachioaxillary angle or medial aspect of the arm. Median nerve damage secondary to bony fracture is less common than with other upper limb nerves because it is protected by overlying arm muscles. Fractures of the elbow occasionally result in median nerve damage. The anterior interosseous nerve may be damaged by forearm fractures because the nerve is closely related to the radius and ulna as it descends on the interosseous membrane. Chronic median nerve entrapment may result from chronic compression in well-defined fibro-osseous tunnels

or following repeated episodes of strenuous muscle activity.59 The most common nerve entrapment in the upper limb is median nerve compression under the transverse carpal ligament (CTS). Other median nerve entrapments include the pronator syndrome, the anterior interosseous syndrome, and entrapment by the supracondylar ligament.

Carpal tunnel syndrome CTS is by far the most common entrapment neuropathy in the arm. The usual presentation is with acroparesthesias, numbness, tingling, and burning sensations, usually in the lateral three fingers. Nocturnal exacerbation of pain and paresthesias is characteristic and may either wake the patient from sleep or be prominent on awakening in the morning. Although pain may involve the forearm and shoulder region, it is rare for patients to describe numbness or paresthesias radiating proximal to the wrist. Shaking the hand frequently relieves pain. Repetitive tasks, be they occupationally related or hobbies, such as knitting and sewing, often precipitate or aggravate symptoms. Objective sensory loss most consistently involves the second and third fingers, occasionally there is splitting of the fourth finger with sensory loss on the lateral but not the medial aspect. The palm is usually spared. Raynaud’s phenomenon may occasionally be present. If the condition is left untreated, weakness and wasting of median innervated thenar muscles eventually develops, particularly thumb abduction and opposition. Eventually, the thumb is unable to be maintained in the opposed and abducted position (post position). As a result, the thumb is restricted to the plane of the palm, precluding it from acting as a post against which the other fingers can push. Thenar atrophy may occasionally precede weakness. Percussion of the median nerve at the wrist may reproduce sensory symptoms in the median nerve distribution (Tinel’s sign). Forced hand flexion (Phalen’s test) or extension (reverse Phalen’s test) may reproduce the sensory symptoms. Asymptomatic motor and sensory signs may rarely be discovered, particularly in older individuals. A discussion of work-related factors, including forceful and repetitive hand use, in the development of CTS is provided in Chapter 23.3. CTS may develop acutely after a prolonged episode of unaccustomed hand use, such as house-painting. In patients with pre-existing CTS, vigorous hand activity can aggravate symptoms. Much of the original support for CTS as a repetitive movement injury can be traced to studies using poor case definition or workers’ compensation claims, and the specific role of chronic repetitive hand use in CTS remains unsettled.60 Two longitudinal studies of median nerve conductivity in workers engaged in a variety of industrial activities61,62 failed to demonstrate any deterioration of median nerve function over a period of greater than 5–10 years. It has been suggested that a substantial increase in intracarpal pressure results from wrist flexion and extension. In addition, there is some evidence to suggest that patients with reduced cross-sectional diameters of their carpal bones may be at increased risk of CTS. However, in

680 Disorders of the Peripheral Nervous System at least one study, routine measurements of carpal tunnel dimensions did not predict the likelihood of developing CTS in an occupational setting. Although CTS most commonly involves the dominant hand, bilateral involvement is extremely common. In many cases, only one side is symptomatic, the asymptomatic lesion evident only by electrodiagnostic studies. Most cases of CTS are probably the result of non-specific tenosynovitis of the flexor tendons. Other causes include tuberculous tenosynovitis, rheumatoid arthritis, osteoarthritis of the carpus, pregnancy, hemodialysis, myxedema, acromegaly, and infiltration of the transverse carpal ligament in primary amyloidosis. CTS is common in patients with generalized peripheral neuropathies, such as uremia and diabetes. In such patients, it is believed that the generalized neuropathy predisposes the median nerve to compression. Acute CTS can be caused by hemorrhage or infection in the carpal tunnel and constitutes a medical emergency. CTS should be considered for any unexplained pain or sensory disturbance in the hand.

Electrodiagnostic studies. In many cases, the diagnosis is obvious from the clinical signs and symptoms. Diagnostic confirmation in atypical cases and an estimate of severity can be obtained by nerve conduction studies. Confirmatory electrodiagnostic studies should be obtained in all patients undergoing surgery. The most sensitive physiologic parameter is sensory and mixed nerve conduction across the carpal ligament. Median motor conduction is less often abnormal than sensory conduction, but may show prolonged distal latencies and reduced compound motor action potential amplitudes, the latter usually reflecting axonal degeneration. Comparison of distal motor latencies between the median innervated second lumbrical muscle and the ulnar interosseous muscle has proven to be sensitive and specific for mild CTS. Treatment and prognosis. Conservative medical therapy is appropriate when there are mild sensory symptoms without weakness or atrophy, intermittent symptoms, or acute CTS related to a specific injury or overactivity.58 Occupationally related symptoms often respond to a switch in jobs to one not requiring repetitive hand movements. Nocturnal immobilization, using a volar wrist splint that maintains the wrist in the neutral position, is usually the initial treatment. Splinting is most useful when symptoms are intermittent and nocturnally exacerbated. The splint should be worn for as long as it is effective; if symptoms recur, further splinting is usually not helpful. Occasionally, the splint may be worn during the day at work to prevent the wrist from assuming a potentially aggravating position. If possible, repetitive actions that aggravate symptoms should be avoided. Splinting, in general, usually offers only temporary or minimal relief. In patients without marked sensory loss, thenar weakness, or muscle wasting, a local injection of a mixture of lidocaine (Xylocaine) and methylprednisolone (Solu-Mediol) can be tried. Low-dose oral prednisone or diuretics (especially in cases complicating pregnancy) may

occasionally be effective. Eventually, however, many patients require surgical release. Carpal tunnel surgery is indicated when conservative therapy fails to alleviate abnormal sensations or when thenar weakness or atrophy (evidence of axonal injury) is present. Surgery may also be needed in patients with occupations that chronically aggravate the symptoms. In such cases, the patient is often faced with a surgical option or the necessity to find an alternative form of work. Surgery is usually effective in relieving pain and stopping the progression of weakness in almost all cases. The long-term prognosis on returning to the same exposure setting is guarded, and may result in aggravation of the condition. Surgery is almost always successful in halting the progression of weakness. If the lesion is not too advanced, recovery of strength usually occurs. Most cases of failed carpal tunnel release result from either incomplete transection of the transverse carpal ligament or faulty initial diagnosis. As indicated previously, preoperative electrodiagnosis is essential to avoid this latter mistake. Reported complications of surgery include neuroma formation, a tender dysesthetic scar, transection of the palmar cutaneous branch, infection, incomplete release, development of a complex regional pain syndrome, and damage to the superficial palmar arch.

Median nerve entrapment at the elbow (pronator syndrome and anterior interosseous syndrome) The pronator syndrome results from compression of the median nerve as it passes between the two heads of the pronator teres muscle and under the fibrous arch of the flexor digitorum superficialis muscle.56,59 The most frequent cause of entrapment is fibrous bands in the substance of the muscle or passage through a tight flexor superficialis arch. Repeated pronation-supination activity may precipitate symptoms in patients with hypertrophied volar forearm muscles. The nerve may occasionally be compressed under the lacertus fibrosus, a fascial band extending from the biceps tendon to the forearm fascia. The pronator syndrome is characterized by diffuse forearm aching and paresthesias in the median nerve distribution over the hand. The degree of weakness varies, ranging from no weakness, to mild involvement of thenar and forearm musculature. Various tests have been advocated to localize the level of nerve entrapment within the pronator muscle mass. Pain in the proximal forearm induced by forced wrist supination and wrist extension suggests compression within the pronator teres. Pain with forced forearm pronation of the fully supinated and flexed forearm suggests entrapment under the lacertus fibrosus. Compression of the median nerve under the flexor superficialis arch is suggested by pain on forced flexion of the proximal interphalangeal joint of the middle finger. Anterior interosseous nerve compression causes weakness of the flexor pollicis longus, pronator quadratus, and the median innervated flexor digitorum profundus muscles. The resulting clinical deficit is impaired flexion of

Focal Nerve Injuries 681 the terminal phalanx of the thumb and index finger. There is no associated sensory loss. The anterior interosseous nerve may be damaged by forearm lacerations or fractures, fibrous bands within the pronator teres, entrapment by the fibrous arch of the flexor superficialis, and as a manifestation of acute brachial neuritis. Occasionally, anterior interosseous nerve dysfunction may occur without precipitating events or follow vigorous forearm muscular activity. Median nerve entrapment at the elbow may infrequently occur from an anomalous fibrous band that extends from the medial epicondyle to a bony spur on the anteromedial surface of the humerus (ligament of Struthers). The resultant weakness involves all median innervated muscles, including the pronator teres, and is accompanied by loss of the radial pulse when the arm is extended. Radiographs often demonstrate the anomalous bony spur.

Electrophysiologic studies. Nerve conduction studies in proximal median nerve compression syndromes are frequently normal. In more severe cases, the amplitudes of distal motor and sensory potentials may be reduced. Distal motor latencies may occasionally be prolonged with stimulation proximal to the elbow, but forearm median motor conduction velocities are usually normal. The most consistent and sensitive physiologic findings are neurogenic changes in median nerve innervated forearm and hand median muscles on needle EMG. In mild cases in which pain but not weakness is present, all electrophysiologic studies may be normal. Treatment and prognosis. Conservative treatment of the pronator syndrome and spontaneous anterior interosseous nerve entrapment initially involves resting the arm by avoiding elbow flexion and pronation.58 Gentle splinting of the arm in supination occasionally relieves symptoms but may also aggravate the condition. Corticosteroid injections into the pronator teres muscle may be of temporary benefit. In mild cases, non-steroidal anti-inflammatory medications may be useful. In both syndromes, persistent or progressive symptoms require exploratory surgery. Occasional cases are the result of schwannomas or other nerve tumors, which can sometimes be visualized by MRI imaging. Surgical excision of the spur and ligament in median nerve compression by the ligament of Struthers is usually successful. Most entrapment injuries to the median nerve do not require splinting. Traumatic injury, however, frequently results in extensive and severe hand weakness for extended periods. In such cases, proper splinting of the thumb is essential to prevent deformity and preserve hand function. Regardless of the site of median nerve injury, thumb movement will be impaired. Loss of thumb abduction leaves it lying adjacent to the index finger, which predisposes to thumb web adductor contractures. A C-bar or wooden dowel, inserted between the thumb and second metacarpal, maintains thumb abduction, while an opponens bar on the proximal phalanx of the thumb stabilizes it in opposition to the index and middle fingers, thereby allowing full wrist movement and posting of the thumb.

Ulnar nerve Source of injury56,59 Ulnar nerve entrapment is second to CTS as the most common nerve entrapment in the arm. Ulnar compression may occur in the axilla or upper arm, elbow region (either ulnar groove or cubital tunnel), or distally at the wrist or hand. The nerve is relatively exposed as it passes around the elbow and at the wrist, making it vulnerable to mild compressive or penetrating injuries. In the forearm, the muscle mass of the flexor carpi ulnaris affords the nerve relatively safety, save for severe penetrating wounds. Most compression injuries of the ulnar nerve are Class I and II. Improper positioning during anesthesia may result in an ulnar palsy. Occasionally, forceful, repetitive, flexionextension movements of the arm may precipitate symptoms of ulnar nerve dysfunction. It is not clear whether ulnar neuropathy can be caused by repetitive arm movements. The relative infrequency of occurrence, in contrast to CTS, has prevented there being a sufficient number of prospective studies, with electrophysiologic verification, to examine this issue. Dislocations or fractures of the elbow or chronic compression resulting from habitual leaning against the elbows may damage the ulnar nerve. Entrapment may occur in the cubital tunnel, where the nerve lies under the aponeurotic band between the two heads of the flexor carpi ulnaris. Occasionally, ulnar nerve dysfunction becomes evident many years after a supracondylar fracture of the humerus, leading to an increased carrying angle at the elbows (tardy ulnar palsy).

Ulnar entrapment at the elbow The exposed position of the ulnar nerve at the medial epicondyle leaves the nerve susceptible to minor trauma or chronic compression. The nerve enters the forearm through a narrow opening (cubital tunnel) formed by the epicondyle, the medial collateral ligament of the joint, and the firm aponeurotic band to which the flexor carpi ulnaris is attached. This entire structure, formerly known as the cubital tunnel, has recently been renamed the humeroulnar aponeurotic arcade (HUA). It is estimated that up to one-half of normal individuals have thickened and enlarged ulnar nerves, predisposing to compression either in the ulnar groove or HUA. Elbow flexion reduces the size of the opening under the aponeurotic band while extension widens it. Ulnar nerve dysfunction may therefore occur secondary to HUA narrowing during elbow flexion, without additional external trauma or local pathologic changes, or may be the result of any condition that increases nerve size or reduces the space available for the nerve. Some patients give no history of recent ulnar injury but report a previous elbow fracture or traumatic injury. This tardy ulnar palsy is frequently insidious in onset and probably results from narrowing of the HUA secondary to osteoarthritis or an increased carrying angle at the elbow. The clinical deficits resulting from ulnar nerve lesions at the elbow are variable. Sensory symptoms usually precede

682 Disorders of the Peripheral Nervous System weakness. Numbness, paresthesias, or pain in the fourth and fifth fingers are most common. Symptoms are occasionally positionally provoked, especially by prolonged elbow flexion, as during sleep or while talking on the phone. Although a chronic ache in the elbow is common, it is unusual for sensory symptoms and signs to extend proximal to the wrist. Cutaneous sensory appreciation is impaired in the fifth finger with occasional splitting of the fourth finger. Objective sensory impairment may involve the dorsum of the hand but does not extend proximal to the wrist crease. Because patients with ulnar neuropathies may not be bothered by paresthesias and pain, they may present with an advanced degree of weakness and wasting. The first dorsal interossei are usually the earliest and most severely affected muscles; weakness and wasting of other ulnar hand and forearm muscles may follow. Severe weakness results in a claw hand deformity with variable flexion of the distal digits, depending on the degree of profundus muscle weakness. Ulnar neuropathy results in the loss of power grip and impaired precision movements.

Electrodiagnostic studies.

Slowed motor or sensory nerve conduction across the elbow, relative to the forearm segment, is a clear localizing sign. Unfortunately, focal slowing of motor conduction is present in only 50–60% of cases. Absolute slowing of motor conduction across the elbow segment, regardless of forearm conduction, has been reported in 65–85% of patients with motor and sensory signs and about 50% of patients with only sensory impairment. Sensory potentials are frequently reduced in amplitude, including the dorsal cutaneous branch. Motor potential amplitudes are diminished in lesions resulting in motor axon degeneration. Nerve conduction studies may be normal in mild cases without weakness, but if axonal degeneration is present, needle EMG will demonstrate active and/or chronic denervation in ulnar-innervated muscles.

Treatment and prognosis.58

Conservative therapy is reserved for three groups of patients: (1) those in whom symptoms are only posturally precipitated, (2) those mild cases in whom symptoms are aggravated by job-related activity, and (3) those who demonstrate only sensory symptoms without substantial progression. Therapy consists of avoiding aggravating movements, such as repeated elbow flexion and extension or habitually resting on the elbows. Splinting the elbow in extension for prolonged periods, especially during sleep, may occasionally be helpful. Elbow pads can be used in those who habitually rest on their elbows. Conservative therapy should be continued for at least 2–3 months or for as long as symptoms remain intermittent or mild, and weakness is absent. Careful follow-up is important to detect a progressive lesion that may result in substantial axonal degeneration. Non-steroidal anti-inflammatory agents are occasionally helpful. Progressive sensory symptoms, new weakness, or worsening electrophysiologic deterioration dictates surgical inter-

vention. Most surgeons now recommend either simple decompression within the HUA, or anterior transposition of the ulnar nerve, deep to the flexor forearm muscle mass. Clinical improvement can be expected in about 75% of cases. The severity of the preoperative lesion is important in predicting recovery; earlier intervention results in better recovery than waiting until severe wasting has occurred. Complications include neuroma formation, recurrent scarring around the nerve, and persistent pain, possibly caused by interruption of the nerve’s blood supply. Splints for ulnar lesions are designed to prevent hyperextension of the ring and little finger’s metacarpophalangeal (MCP) joints. A restraint on the dorsum of the proximal phalanx supports the MCP joint. Should IP joint flexion persist, a dynamic splint may be used to extend the distal IP joint.

Ulnar nerve lesion at the wrist The ulnar nerve may be lacerated at the wrist or compressed either within Guyon’s canal or more distally within the palm. The most common cause of distal ulnar compression is chronic repeated trauma to the palmar area, such as that which occurs among heavy laborers or cyclists. Other etiologies include ganglions that compress the nerve, either within Guyon’s canal or distally along the deep terminal motor branch, fractures of the carpal bones, lipomas and other tumors, and rheumatoid arthritis. Acute trauma, such as a fall on the outstretched hand, may damage the nerve as it passes between the pisiform bone and the hook of the hamate. Compression within Guyon’s canal may result in any combination of weakness in hypothenar and thenar ulnar innervated muscles and sensory loss in the medial two fingers. Sensory innervation to the palmar and dorsal surfaces of the hand is spared. Distal ulnar compression should be suspected when ulnar innervated thenar muscles are weaker than hypothenar muscles. Other clinical scenarios include selective weakness of the hypothenar muscles and isolated sensory loss of the medial two fingers. Rarely, a more proximal wrist lesion may additionally affect the palmar sensory branch.

Electrodiagnostic studies.

The most specific electrodiagnostic finding is a prolonged distal motor latency to the first dorsal interosseous muscle compared with the abductor digiti minimi muscle. The digital sensory potential to the fifth finger is of low amplitude or absent; dorsal cutaneous sensory potential is spared. Depending on the site of lesion, needle EMG may demonstrate active or chronic denervation in either hypothenar or thenar muscles with sparing of ulnar-innervated forearm muscles.

Treatment and prognosis. Conservative therapy is indicated when there are sensory symptoms alone. Maneuvers responsible for or aggravating the injury should be corrected. Surgery is indicated when conservative therapy has failed to relieve discomfort, when motor or sensory dysfunction progresses, or when the deficit has no clear cause and may be the result of a mass.

Toxic Neuromuscular Disorders 683

TOXIC NEUROMUSCULAR DISORDERS The incidence of clinical neuromuscular transmission dysfunction resulting from toxins is infrequent compared with that of peripheral neuropathy. Recognition of a toxic etiology is vital, however, because complete function is usually returned after the offending agent(s) is(are) identified and eliminated. A discussion of the physiology of neuromuscular transmission is beyond the scope of this chapter, and the reader is referred to one of the many excellent texts dealing with this subject. The following is a brief description of the various toxins that have been identified to disrupt neuromuscular transmission.

Snake envenomation The venom of certain poisonous snakes may produce acute, widespread weakness that clinically resembles a myasthenic crisis.63 Two families of snakes have venom with a predilection for the neuromuscular junction, i.e., the Elapidae (coral snakes, cobras, mambas, and kraits) and the Hydrophiidae (sea snakes). Envenomation with cobra venom causes symptoms and signs within minutes to hours. The symptoms include ptosis, oculomotor paralysis, lower bulbar nerve dysfunction (resulting in lingual, laryngeal, and pharyngeal dysfunction), diffuse weakness, and eventually, respiratory compromise. In addition to neuromuscular blockage, local muscle necrosis may result. Impaired neuromuscular transmission from snake venom results from: (1) postsynaptic acetylcholine receptor blockade, which may be irreversible or partially reversible, and (2) presynaptic inhibition of acetylcholine release. Treatment of snake envenomation includes administration of anticholinesterases to antagonize postsynaptic receptor blockade, specific antivenom, debridement of necrotic muscle tissue, and mechanical respiration, when needed.

Arthropod envenomation Impaired neuromuscular transmission may result from the toxins of black widow spiders, scorpions, female ticks, and wasps.64 Black widow venom increases acetylcholine release and, probably, also prevents endocytosis of acetylcholine vesicles to nerve terminal membranes. The eventual result is depletion of such vesicles and impaired neuromuscular transmission. Clinical symptoms begin within 15–60 minutes of toxin injection and predominantly involve severe muscle cramps of the abdomen and limbs. Treatment is aimed at reducing muscle cramping by warming, infusing calcium gluconate, and administering magnesium sulfate to antagonize acetylcholine release. Atropine may help antagonize the cholinergic symptoms, and the administration of 2.5 mL of reconstituted antiserum is recommended. Acute weakness may also be caused by the toxin from North American ticks (Dermacentor andersoni and D. variabilis). Tick paralysis may clinically resemble acute inflammatory demyelinating neuropathy (AIDP) and may result in

respiratory paralysis.65 The exact site of the pathologic change remains uncertain; terminal motor nerve endings and neuromuscular junctions have both been implicated. Removal of the tick is curative, often leading to rapid clinical recovery. Anticholinesterase therapy is usually ineffective. Scorpion toxin indirectly affects the neuromuscular junction by inducing AChE release, caused by repetitive nerve terminal impulses.66 These repetitive action potentials are believed to result from a toxin-induced delay in the normal sodium channel inactivation, thereby prolonging the nerve action potential. Clinical neuromuscular junction dysfunction is usually minimal. Weakness simulating ocular myasthenia may result from the stings of wasps, bees, and hornets. Ocular symptoms may persist for weeks and are responsive to anticholinesterases and prednisone. Generalized weakness is rare.

Toxic myopathies A number of toxic chemicals may primarily affect muscle.67 The most common is alcohol, which produces a number of clinical scenarios, resulting from a combination of a primary toxic effect on muscle, related nerve disease, and metabolite disturbances, especially potassium. Most other toxic myopathies result from pharmaceutical agents, rather than industrial or occupational exposure. In many cases, the relationship between toxin and muscle weakness is not initially apparent. Most toxic myopathies are generalized and symmetric in nature, affecting proximal muscles most severely. Cranial nerve musculature is usually spared. Focal myopathies are usually caused by the local injection of toxic agents.

Focal myopathies Focal muscle dysfunction, secondary to toxins, usually results from intramuscular injections, either from a local effect of the needle or the administered drug. Drugs with a known local toxic effect include diazepam (Valium), lidocaine, digoxin (Lanoxin), chloroquine (Aralen), opiates, and chlorpromazine (Thorazine). Paraldehyde and cephalothin sodium (Keflin) produce local irritation and abscess formation. Clinical weakness is usually minimal. The most common manifestation is an elevated creatine kinase (CK) level. Occasionally, severe induration and contractures may result. This is especially common with intramuscular injections of antibiotics in children or with meperidine (Demerol) or pentazocine (Talwin).

Acute/subacute myopathies Many drugs can cause rapid-onset, symmetric, proximal muscle weakness, which is often painful.67 The underlying mechanism is usually related to either potassium deficiency or an inflammatory myopathy. Occasionally, as with Ltryptophan, inflammation may predominate in the fascial tissue rather than the muscle itself. The serum CK concentration is usually markedly elevated, and myoglobinuria may be present. Drugs implicated in such myopathies include clofibrate (Atromid-S), epsilon-aminocaproic acid (Amicar),

684 Disorders of the Peripheral Nervous System emetine hydrochloride, and vincristine (Oncoven). Clear inflammatory myopathies have been reported in patients taking procainamide (Provestyl), levodopa (Larodopa), and D-penicillamine. Patients ingesting certain brands of the amino acid L-tryptophan have developed severe myalgias and occasionally weakness. Pathologic changes have included a marked perimysial and perivascular inflammatory response with varying degrees of muscle necrosis. Recovery was often exceedingly prolonged and far outlasted the period of exposure. In most of these inflammatory myopathies, spontaneous recovery often accompanies removal of the drug, but steroids additionally may be needed. Acute muscle weakness caused by hypokalemia is usually generalized and often painful. Frank muscle necrosis may result, and elevated CK levels and myoglobinuria may be present. Hypokalemia may result from diuretics, purgatives, amphotericin B (Fungizone), and carbenoxolone. Amphotericin B and carbenoxolone, along with amphetamines, have been implicated in producing a severe necrotizing myopathy with rhabdomyolysis.

Chronic myopathies Many drugs produce a chronic painless myopathy presenting with limb-girdle weakness. Proximal leg and pelvic muscles are usually affected to a greater degree than are arm muscles. CK levels may be normal. The most commonly implicated drugs are corticosteroids, which produce weakness through their interference with oxidative metabolism and inhibition of protein synthesis. Other drugs include chloroquine, which can produce a clinical picture identical to that from steroids, heroin, and chronic hypokalemia. Muscle weakness has also been ascribed to perhexiline, colchicine, and rifampin (Rimectane).

Alcoholic myopathy The deleterious effect of alcohol on muscle may be through a direct toxic effect or associated malnutrition and electrolyte disturbance. Alcohol may produce acute, subacute, and chronic weakness.67 The acute variety usually follows binge drinking and is associated with painful swollen muscles and myoglobinuria. An acute but painless variety of muscle weakness may also occur, associated with marked hypokalemia, elevated CK levels, and muscle necrosis. Recovery proceeds over several weeks to months, but the problem may recur with subsequent drinking bouts.

5.

6.

7. 8.

9. 10. 11.

12. 13.

14. 15.

16. 17. 18.

19.

20.

21.

22.

23.

References 1. Schaumburg HH, Wisniewski HM, Spencer PS. Ultrastructural studies of the dying-back process. I. Peripheral nerve terminal and axon degeneration in systemic acrylamide intoxication. J Neuropathol Exper Neurol 1974; 33:260. 2. Spencer PS, Schaumburg HH. Central-peripheral distal axonopathy – the pathogenesis of dying-back polyneuropathies. In: Zimmerman H, ed. Progress in neuropathology, vol. 3. New York: Grune and Stratton, 1976;253. 3. Schaumburg HH, Spencer PS. Recognizing neurotoxic disease. Neurology 1987; 37:276. 4. Klassen C. Heavy metals and heavy metal antagonists. In: Goodman, Gilman A, Rall TW, Nies AS, Taylor P, eds.

24.

25. 26.

27.

28.

Goodman and Gilman’s the pharmacological basis of therapeutics. Elmsford, New York: Pergamon Press, 1990;1602. O’Shaughnessy E, Kraft GK. Arsenic poisoning: long-term follow-up of a non-fatal case. Arch Phys Med Rehab 1976; 57:403. Windebank AJ, McCall JT, Dyck PJ. Metal neuropathy. In: Dyck PJ, Thomas PK, Lambert EH, Bunge RP, eds. Peripheral neuropathy, 2nd edn. Philadelphia: WB Saunders, 1984;2133. Oh SJ. Electrophysiologic profile in arsenic neuropathy. J Neurol Neurosurg Psych 1991; 54:1103. Heyman A, Pfeiffer JB, Taylor HM. Peripheral neuropathy caused by arsenical intoxication: a study of 41 cases with observation on the effects of BAL (2,3-dimercapto-propanol). N Engl J Med 1956; 254:401. Feldman RG. Occupational and environmental toxicology. Philadelphia: Lippincott-Raven Publishers, 1999;30. Whitfield CL, Ch’ien LT, Whitehead JD. Lead encephalopathy in adults. Am J Med 1972; 52:289. Chancellor AM, Slattery JM, Fraser H, Warlow CP. Risk factors for motor neuron disease: a case-control study based on patients from the Scottish Motor Disease registry. J Neurol Neurosurg Psych 1993; 56:1200. Seppalainen AM, Hernberg S. Sensitive technique for detecting subclinical lead neuropathy. Br J Ind Med 1972; 29:443. Seppalainen AM, Tola S, Hernberg S, Kock B. Subclinical neuropathy at ‘safe’ levels of lead exposure. Arch Environ Health 1975; 30:180. Markowitz ME , Rosen JF. Need for the lead mobilization test in children with lead poisoning. J Pediatr 1991; 119:305. Kyrland LT, Faro SN, Siedler H. Minamata disease. The outbreak of a neurologic disorder in Minamata, Japan and its relationship to the ingestion of seafood contaminated by mercuric compounds. World Neurol 1967; 1:370. Bakir F, Damluji SF, Amin-Zaki L, et al. Methylmercury poisoning in Iraq. Science 1973; 181:230. Brune D, Evje DM. Man’s mercury loading from a dental amalgam. Sci Total Environ 1985; 44:51. Takeuchi T. Pathology of Minamata disease. In: Kutsuma M, ed. Minamata Disease. Kumamoto: Study Group of Minamata Disease, Kumamoto University, Japan, 1968;141. Albers JW, Kallenbach LR, Fine LJ, et al. and the Mercury Studies Group. Neurological abnormalities associated with remote occupational elementary mercury exposure. Ann Neurol 1988; 24:651. Singer R, Valciukas JA, Rosenman KD. Peripheral nerve toxicity in workers exposed to inorganic mercury compounds. Arch Environ Health 1987; 42:181. Campbell D, Gonzales M, Sullivan JB Jr. Mercury. In: Sullivan JB Jr, Kreiger GR, eds. Hazardous materials toxicology: clinical principles of environmental health. Baltimore: Williams and Wilkins, 1992;824. Davis LE, Stadefr JC, Kornfeld M, Abercrombie DM, Butler C. Acute thallium poisoning: toxicological and morphological studies of the nervous system. Ann Neurol 1981; 10:38. Douglas KT, Bunni MA, Baindur SR. Minireview: thallium in biochemistry. Int J Biochem 1990; 22:429. Nordentoft T, Anderson EB, Morgensen PH. Initial sensorimotor and delayed autonomic neuropathy in acute thallium poisoning. Neurotoxicol 1998; 19:421. Dumitru D, Kalantri A. Electrophysiologic investigation of thallium poisoning. Muscle Nerve 1990; 13:433. Moore D, House I, Dixon A. Thallium poisoning: diagnosis may be elusive but alopecia is the clue. BMJ 1993; 306(6891):1527. Stevens W, van Peteghem C, Heyndricks A et al. Eleven cases of thallium intoxication treated with Prussian blue. Int J Clin Pharmacol 1974; 10:1. Metcalf DR, Holmes JH. EEG, psychological and neurological alterations in humans with organophosphorous exposure. Ann NY Acad Sci 1969; 160:357.

Toxic Neuromuscular Disorders 685 29. Wadia RS, Sadagtopan C, Amin RB, Sardesai HV. Neurological manifestations of organophosphorous insecticide poisoning. J Neurol Neurosurg Psych 1974; 37:841. 30. DeBleeker J. The intermediate syndrome in organophosphate poisoning: an overview of experimental and clinical observations. Clin Toxicol 1995; 33:683. 31. Wadia RS, Chitra S, Amin RB, Kiwalkar RS, Sardesai HV. Electrophysiologic studies in acute organophosphate poisoning. J Neurol Neurosurg Psych 1987; 50:1442. 32. Besser R, Gutman I, Dillmann U, Weilemann LS, Hopf HC. End-plate dysfunction in acute organophosphate intoxication. Neurology 1989; 39:561. 33. Misra UK, Nag D, Khan WA, Ray PK. A study of nerve conduction velocity, late responses and neuromuscular synapse function in organophosphate workers in India. Arch Toxicol 1988; 61:496. 34. LeWitt P. The neurotoxicity of the rat poison vacor. N Engl J Med 1980; 302:73. 35. Watson DF, Griffin JW. Vacor neuropathy: ultrastructural and axonal transport studies. J Neuropathol Exp Neurol 1987; 46:96. 36. Gordy ST, Trumper M. Carbon disulfide poisoning with a report of six cases. JAMA 1938; 110:1543. 37. Chu C-C, Huang C-C, Chen R-S, Shih T-S. Polyneuropathy induced by carbon disulfide in viscose rayon workers. Occup Environ Med 1995; 52:404. 38. Seppalainen AM, Tolonen M. Neurotoxicity of long term exposure to carbon disulfide in the viscose rayon industry. A neurophysiologic study. Work Environ Health 1974; 11:145. 39. O’Donoghue JL, Krasavage WJ. Hexacarbon neuropathy: a gamma-diketone neuropathy? J Neuropath Exp Neurol 1979; 38:333(abst). 40. Mendell JR, Saida K, Ganansia MG, et al. Toxic polyneuropathy produced by methyl n-butyl ketone. Science 1974; 185:787. 41. Schaumburg HH, Spencer PS. Environmental hydrocarbons produce degeneration in cat hypothalamus and optic tract. Science 1978; 199:199. 42. Korobkin R, Asbury AK, Sumner AJ, Nielsen SL. Glue sniffing neuropathy. Arch Neurol 1975; 32:158. 43. Davenport JG, Farrell DF, Sumi SM. ‘Giant axonal neuropathy’ caused by industrial chemicals: neurofilamentous masses in man. Neurology 1976; 26:919. 44. Asbury AK, Gale MK, Cox SC, Baringer JR, Berg BO. Giant axonal neuropathy: a unique case with segmental neurofilamentous masses. Acta Neuropathol 1972; 20:237. 45. He F, Zahng S. Effects of allyl chloride on occupationally exposed subjects. Scand J Work Environ Health 1985; 11:43. 46. Gold BG, Griffin JW, Price DL. Slow axonal transport in acrylamide neuropathy: different abnormalities produced by single-dose and continuous administration. J Neurosci 1985; 5:1755. 47. Miller MS, Spencer PS. Single dose of acrylamide reduce retrograde transport velocity. J Neurochem 1984; 43:1401. 48. Schaumburg HH, Arezzo JC, Spencer PS. Delayed onset of axonal neuropathy in primates after prolonged low-level administration of a neurotoxin. Ann Neurol 1989; 26:576.

49. Spencer PS, Schaumburg HH. Ultrastructural studies of the dying back process. III. The evolution of experimental peripheral giant axonal degeneration. J Neuropathol Exp Neurol 1977; 36:276. 50. Schaumburg HH, Wisniewski HM, Spencer PS. Ultrastructural studies of the dying back process. Peripheral nerve terminal and axon degeneration in systemic acrylamide intoxication. J Neuropathol Exp Neurol 1974; 33:260. 51. Fullerton PM. Electrophysiologic and histologic observations on the peripheral nerves in acrylamide poisoning. J Neurol Neurosurg Psych 1969; 32:186. 52. Pestronk A, Keogh J, Griffin JG. Dimethylaminopropionitrile intoxication: a new industrial neuropathy. Neurology 1979; 29:540. 53. Cavalleri F, Galassi G, Ferrari S. Methyl bromide induced neuropathy: a clinical, neurophysiological and morphological study. J Neurol Neurosurg Psychiat 1995; 58:383. 54. Gross JA, Haas ML, Swift TR. Ethylene oxide neurotoxicity: report of four cases and review of the literature. Neurology 1979; 29:978. 55. Murai Y, Kuroiwa Y. Peripheral neuropathy in chlorobiphenyl poisoning. Neurology 1971; 21:1173. 56. Stewart JD, ed. Focal peripheral neuropathies. Philadelphia; Lippincott, Williams and Wilkins, 2000. 57. Mitsunage MM, Nakano K. High radial nerve palsy following strenuous muscular activity. Clin Orthop 1988; 234:39. 58. Berger AR, Schaumburg HH. Rehabilitation of focal nerve injuries. J Neurol Rehab 1988; 2:65. 59. Dawson DM, Hallett M, Wilbourne AJ, eds. Entrapment neuropathies. Philadelphia; Lippincott, Williams and Wilkins, 1999. 60. Berger AR, Herskovitz S. Cumulative trauma disorder. In: Rosenberg NL, ed. Occupational and enviromental neurology. Boston: Butterworth-Heinemann, 1995;235. 61. Nathan PA, Keniston RC, Myers LD, Meadows KD, Lockwood RS. Natural history of median nerve sensory conduction in industry: relationship to symptoms and carpal tunnel syndrome in 558 hands over 11 years. Muscle Nerve 1998; 21:711. 62. Nilsson T, Hagberg M, Berstrom L, Lundstrom R. A five-year follow-up of nerve conduction over the carpal tunnel. Stockholm workshop 94. Hand-arm vibration syndrome. Arbete Och Halsa Vetenskaplig skriftserie 1995; 5:117. 63. Campbell CH. The effects of snake venoms and their neurotoxins on the nervous system of man and animals. In: Hornabrook RW, ed. Topics in tropical neurology. Philadelphia: FA Davis, 1975;259. 64. Gilbert EW, Stewart CM. Effective treatment of arachnoidism by calcium salts. Am J Med Sci 1935; 189:532. 65. Cherington M. Botulism: ten-year experience. Arch Neurol 1974; 30:432. 66. Warnick JE, Albuquerque EX, Diniz CR. Electrophysiologic observations on the action of the purified scorpion venom, Tityustoxin, on nerve and skeletal muscle of the rat. J Pharmacol Exp Ther 1976; 198:155. 67. Griggs RC, Mendell JR, Miller RG, eds. Evaluation and treatment of myopathies. Philadelphia: FA Davis, 1995.

28.3 Psychosocial Stressors and Psychiatric Disorders in the Workplace Nancy Fiedler, Lawrence B Stein Most adults spend a significant part of their daily life in the workplace and research has demonstrated that a stimulating job contributes to a more meaningful life, higher self-esteem, and opportunities to increase socioeconomic status. However, psychosocial and physical work stressors also contribute to a decrease in work function and to psychologic disturbances ranging from stress reactions to diagnosable psychiatric illnesses. For example, some occupations such as emergency workers may be more prone to stress-related illnesses due to the unrelenting demands that are a routine part of their occupation. Psychiatric illness (e.g., depression and anxiety) is recognized as a leading occupational health problem by the National Institute of Occupational Safety and Health1 and often occurs from the interaction of work and non-work related environmental stressors and individual predispositions. Therefore, to the extent that the occupational practitioner can identify work-related factors contributing to psychiatric illness and develop appropriate intervention strategies, the impact on the individual and the organization can be mitigated. This chapter offers an overview of the literature on work-related psychosocial and physical stressors and individual susceptibilities associated with psychiatric symptoms and illness. Following this overview and in the context of a public health prevention model, methods for identifying risks within the workplace and preventing the effects of psychiatric illness on the individual and the organization are presented.

OCCUPATIONS, PSYCHIATRIC DISORDERS, AND PRODUCTIVITY Psychiatric disorders impair productivity through several avenues that include reduced labor supply, absenteeism, poor morale, and reduced quality of work. Statistics validating the overall economic impact of psychiatric disorders within the United States abound; depression is estimate to cost $43.7 billion, while alcoholism costs approximately $89 billion per year.2 This amount includes costs of lost productivity, as well as healthcare. Aside from alcoholism, drinking behavior of undiagnosed employees, such as coming to work ‘hung over’, has also been associated with problems at work.3,4 In short, the costs of behavioral problems and psychiatric disorders affect not only workers and their families, but also managers, coworkers, employers and insurance companies. As an initial step toward understanding occupational risk factors that contribute to psychiatric disorders, epidemiologic studies have compared prevalence rates across occupations. For example, the Epidemiologic Catchment Area (ECA) community survey of psychiatric disorders found, after controlling for demographic differences among 104

occupations, that lawyers, teachers, counselors (not college), and secretaries had elevated rates of major depression compared to the overall rate for employed individuals; movers (freight, stock, and material), transport and material moving occupations, handlers, equipment cleaners and laborers, janitors and cleaners, and waiters and waitresses had elevated rates of alcoholism.5 The ECA study also found, however, that the risk for a diagnosis of alcoholism increased for managers and construction laborers when they were unemployed, confirming the overall protective nature of being employed. Comparisons of psychiatric disorders across occupations, while intriguing and heuristically valuable, do not specify what factors within these occupations contribute to risk. Thus, a significant literature has developed in which the behavioral consequences of psychosocial and physical risk factors have been explored.

WORK-RELATED PSYCHOSOCIAL RISKS FOR PSYCHIATRIC DISORDER The mechanism for higher rates of psychiatric disorder among occupations varies with both health-enhancing and deleterious characteristics of the job and individual susceptibilities. Moreover, job characteristics may contribute to the type of psychiatric disorder. With regard to job qualities, evidence appears to be converging to support the significance of job demands and control over decisions as proposed by Karasek.6 For example, employees in lower status positions, who often experience little control over decisions, also have higher rates of psychiatric disorders than those in higher positions (i.e., more depression and anxiety). Although common sense suggests that high demands would cause increased illness, both Mandell et al.5 and Hemmingson & Lundberg7 reported greater risk of alcoholism with lower or no workload. The latter is also consistent with the reported higher rates of alcoholism among unemployed workers and suggests that the relationship between job demands and psychiatric disorder may be curvilinear. That is, low work load (or unemployment) or working in an environment with strenuous demands and little or no control may have serious implications for an individual. In addition, moderators mitigate the negative effects of job demands. Stansfeld et al.8 reported that high social support and skill discretion were protective against absence due to psychiatric illness. Social support consisted of a high level of support from colleagues and supervisors coupled with clear and consistent information from supervisors, and skill discretion referred to job variety and the opportunity to use skills at work. More recently, a nationwide study of Swedish males found that a combination of low work control, low work demands,

Physical Risk Factors 687 and low work social support was related to later alcoholism after controlling for other risk factors.7 This finding further supports the association between low work demands and alcoholism and incorporates other important psychosocial moderators. In a follow-up interview for the ECA study, Mausner-Dorsch & Eaton9 found support for the association between major depressive episode, depressive syndrome, and dysphoria and job strain defined as a combination of high demands and low control over decisions. Moreover, they found that women were at greater risk both because they were more susceptible to job strain and were probably more exposed to it in their work environment. The latter illustrates the interactive effect of individual susceptibility and workrelated stressors on eventual psychiatric illness. In summary, the relationship between job demands and psychiatric illness is moderated, particularly when demands are high, by perception of control over decisions. That is, if the individual has control, demands may not have the same negative impact on psychiatric function.

Work–family conflict In addition to work stressors, pressures external to work also impact the incidence of psychiatric illness. For example, work–family conflicts, defined as the incompatibility between the role demands of a career and family (e.g., long work hours preventing time with spouse or other family members), have been associated with lower job, marital, and life satisfaction.7 More specific to the workplace, individuals who are experiencing high work–family conflict are more prone to job turnover and decreased job productivity. Some gender differences have been found; women show elevated risk of disability when they report both work and marital interpersonal conflict, whereas interpersonal conflict was not predictive of work disability for men.10 Instead, life dissatisfaction, neuroticism, monotonous work, and stress of daily activities predicted work disability for men. Duxbury & Higgins11 report that men experience work–family conflict due to work expectations, while women experience more work–family conflicts due to family expectations. This overview suggests that greater risk for psychiatric disorder occurs when the following work and non-work related conditions converge: (1) the employee is either insufficiently challenged (low job demands) or has too many demands with little control, (2) social support is low

at work, (3) work and family roles conflict, (4) the employee is female. Since women are more often in occupations with less control and simultaneously have more family responsibilities,12 gender may be a surrogate for differences in exposure to stressors at work and in the home. Table 28.3.1 summarizes the work-related psychosocial risk factors and associated psychiatric symptoms.

PHYSICAL RISK FACTORS Physical aspects of the work environment also contribute to the occurrence of stress and psychiatric disturbance. Exposure to toxic chemicals, poor indoor air quality, noise, shiftwork, and trauma are some of the physical stressors associated with psychiatric illness.

Neurotoxicant exposure Numerous studies document psychiatric symptoms associated with exposure to lead, organic solvents, carbon monoxide, and mercury. Lead exposure, whether acute or chronic, may result in non-specific symptoms often found in psychiatric disorders, such as fatigue, decreased libido, restlessness and depression.13,14 Organic solvent exposure is well known to produce a variety of psychiatric symptoms ranging from mild mood disturbances to severe psychoses. Acute solvent exposure is most often followed by mood changes, transient euphoric reactions, or complaints of mental confusion.15 The organic affective syndrome, identified by the World Health Organization and by NIOSH, is associated with chronic solvent exposure and is characterized by symptoms such as irritability, poor concentration, and loss of interest, symptoms also seen in several psychiatric disorders.16,17 Post-traumatic stress disorder,18 somatoform disorder,19 schizophreniform disorder,20 and panic disorder21 have been documented to occur following long-term exposure to organic solvents. Similarly, carbon monoxide (CO) exposure from motor vehicle exhaust or malfunctioning heating systems can produce symptoms such as fatigue, apathy, emotional lability, lowered frustration tolerance, impulsivity, irritability, and at times, psychosis.22-25 The symptoms of CO exposure may be delayed and may occur from 3 to 240 days after recovery from acute intoxication;26 50–75% of exposed patients recover from this delayed syndrome within one

Psychosocial risk factors

Symptoms

High job demands with low decision latitude7,48 Lack of social support53

Feelings of helplessness and hopelessness in both home and work environments; lack of sense of personal accomplishment; and irritability; emotional exhaustion; feelings of depersonalization Increased conflict at home; decreased feelings of self-efficacy in decision making; decreased commitment to organization; increased absenteeism Decreased marital satisfaction; decreased job productivity; increased sense of guilt Increased use of sick leave time; decreased self-esteem; emotional exhaustion; feelings of depersonalization

9-11

Work–family conflict

Interpersonal conflicts with coworkers and supervisors48

Possible psychiatric illness Depression and anxiety Depression and anxiety Depression; anxiety; and heavy alcohol consumption Depression; anxiety; somatic symptoms

Table 28.3.1 Psychosocial risk factors, symptoms associated with these factors, and possible psychiatric illness that may result from prolonged exposure to stressors

688 Psychosocial Stressors and Psychiatric Disorders in the Workplace year.27 Exposure to mercury, which occurred when hatmakers used mercury to process felt (hence the phrase, ‘mad as a hatter’), has been found to result in both mania and chronic depressed mood with apathy and extreme shyness and withdrawal.28 Chronic mild exposure may cause irritability, nervousness, fatigue, and depression.29 Because the clinical presentation of neurotoxicant exposure has significant overlap with symptoms arising from many other factors to include work and non-work related psychosocial stressors, a good exposure history is essential to develop appropriate intervention. For example, if the health practitioner is presented with psychiatric symptoms without exploration of neurotoxicant exposures, behavioral therapy and psychotropic medications may be prescribed when removal or reduction of exposure is the appropriate intervention. Thus, inquiry about the chemicals used at work and at home is an important part of an evaluation for psychiatric disorder.

Indoor air quality Sick building syndrome or non-specific building-related illness (NSBRI) presents as a constellation of symptoms that overlap to some extent with symptoms of stress and psychiatric illness. Symptoms include mucous membrane irritation, headache, fatigue, shortness of breath, rash and abnormal odor perception.30,31 In at least one cross-sectional study, workers in a sick versus control building reported an increased number of psychologic symptoms which did not account for NSBRI but appeared to be independent consequences of poor indoor air quality.32 Several work-site factors are hypothesized to account for the symptoms associated with poor indoor air quality to include direct toxic effects of the chemical mixtures, psychosocial job stress (e.g., poor supervision, high job demands) and noxious odors. Most studies suggest that symptoms likely result from an interaction of these chemical and psychosocial stressors, with each workplace presenting a unique combination of factors.

Noise and shiftwork It has been noted that factory and construction workers exposed to high levels of noise have demonstrated a wide variety of complaints beyond hearing.33 For example, individuals exposed to excessive noise have reported symptoms of depression, anxiety, insomnia, and weight loss.34 Exposure to loud and unsystematic noise has also been demonstrated to create an uncomfortable workplace, decrease worker production, and is viewed by the worker as a significant physical, social, and cognitive stressor.35 Shiftwork represents another source of physical and psychological stress for employees. Little evidence exists regarding a causal relationship between shiftwork and psychiatric disorders, with the exception of individuals with shift maladaptation syndrome (see Chapter 38). Nonetheless, shiftworkers report lower subjective levels of physical health and wellbeing.36-38 In addition, they have higher rates of alcohol and substance abuse39 as compared to daytime workers and high rates of neuroticism.40,41 It has

also been noted that the ‘graveyard’ shift (e.g., midnight to 8:00 am) is associated with the most problems (e.g., higher accident rates, lower performance quality); however, the swing shift has the most negative impact on social patterns and interpersonal interactions.42 At present, studies investigating the prevalence of depression in shiftworkers are contradictory and inconsistent, which reflects the necessity for further research to determine more thoroughly the potential psychological consequences of the worker’s schedules on psychological functioning.

Workplace trauma Workplace trauma can occur under several conditions: (1) physically hazardous working conditions where safety procedures are inadequate or not practiced, (2) the nature of the work is inherently traumatic (e.g., emergency response teams), and (3) work settings with risk for violence. The construction trades have a higher rate of injury and death than many other trades43 and workers who are injured or who witness the injury or death of a coworker are at risk for post-traumatic stress disorder. Similarly, emergency response workers such as police, firefighters, and emergency medical technicians, by the nature of their work, encounter and manage violence, death, and injury on a routine basis.44 Finally, employees in service positions may encounter violence under the following conditions: exchanging money, interacting with the public, working at night or in the early morning, delivering goods, and working alone. In work settings where trauma is likely to occur, preventive measures should be emphasized, and emergency procedures need to be implemented following a trauma to prevent psychiatric disorder and disability.

INDIVIDUAL SUSCEPTIBILITY Risk factors for psychiatric disorders do not confine themselves to work stressors but also lie within the individual. Many of these factors may be psychosocial while others may be biologic or genetic. For example, compared to men, women are exposed to more job strain9 and may also be more sensitive to it and to interpersonal conflict than men.10 Gender is another well-known risk factor for specific psychiatric disturbance, with women being more vulnerable to depression and men abusing alcohol more than women.45 Individuals who have a history of previous psychiatric disturbance or trauma are also at greater risk either for exacerbation of current psychiatric illness or recurrence with exposure to work-related stressors. Similarly, individuals with lower socioeconomic status and those who are unemployed are more at risk for a wide variety of psychopathology including depression, anxiety, substance abuse, and schizophrenia. On the other hand, being employed in a job that is satisfying and productive may act as a buffer against these types of problems. Although in the early stages of investigation, genetic factors may contribute to risk particularly for susceptibility to neurotoxicant exposures.46,47

Recognizing and Assessing Stress, Burnout, and Psychiatric Illness in the Workplace 689

RECOGNIZING AND ASSESSING STRESS, BURNOUT, AND PSYCHIATRIC ILLNESS IN THE WORKPLACE Prevention of psychiatric disorder requires assessment of the job characteristics that contribute to symptoms and early diagnosis to reduce individual and organizational disabilities. The following section provides an overview of psychiatric disorders most frequently seen in the workplace and measures to screen for these symptoms (Table 28.3.2).

Stress symptoms and burnout Stress-related problems often present as physical symptoms such as gastrointestinal distress, chronic headache, low back pain, and fatigue. Symptoms of anxiety and depressed mood that fall short of diagnostic criteria for depression and anxiety disorders, manifest themselves as stress syndromes. Job dissatisfaction and poor morale are organizational manifestations of stress. For example, machine-paced assembly work has been associated with somatic complaints, job dissatisfaction, anxiety, irritability, and depression.35 Stress symptoms, while overlapping, can be distinguished from psychiatric disorders based on three criteria: number of symptoms, chronicity of symptoms, and effects on function. Typically, to meet criteria for a psychiatric disorder, the individual must report a specified number and type of symptoms that have persisted over a given time period (e.g., 2 weeks) and that interfere in the performance of significant life functions at home and/or work. For example, feeling depressed and worried yet not reporting significant effects on interpersonal or work function such as conflicts with coworkers or impaired work performance would probably not meet criteria for a psychiatric disorder.

Measures

However, within the work environment, if these symptoms occur in several individuals, a net effect on productivity and morale may be observed. Thus, assessing job-related stress and intervening can prevent more serious individual and organizational disorder if these symptoms persist. Job burnout is probably the extreme example of a stressrelated syndrome. Although it is not an official medical or psychiatric diagnosis, it is often thought of as a precursor to psychiatric disorder. Maslach48 proposed a tripartite model of burnout that encompasses both job and personal characteristics. Included in this model are emotional exhaustion, cynicism, and a lack of sense of personal accomplishment. It has been duly noted that increased levels of emotional exhaustion, which refers to feelings of ‘being overextended, drained, or used up’, is correlated with physical symptoms such as gastrointestinal disorders, chronic fatigue, hypertension, headaches, sleep disturbances, and flu/cold symptoms. On the other hand, chronic feelings of cynicism (i.e., negative or detached feelings concerning job and workrelated behaviors) and ineffectiveness (i.e., decreased feelings of competency) are thought to contribute to other problems such as despondency, hopelessness, and, in extreme cases, depression and anxiety. Workers experiencing high levels of burnout in any one of these three domains tend to have a higher rate of job turnover, higher rates of absenteeism, are more inflexible about work-related rules and procedures, and are more dissatisfied with their job when compared to their more engaged counterparts. Although burnout is seen across jobs and work settings, it is more likely when the job involves frequent contact with people who are in need of help (e.g., health and service fields) and when a job supervisor is perceived as being unresponsive to employee needs. When either the type of work or the supervisory climate promote burnout, assessment for burnout is the first step to prevent disability and high worker turnover (Table 28.3.2).

Description of measure

Stressors Job content questionnaire76 74

Hassles and uplifts scales Stress symptoms Maslach burnout inventory48

Symptom checklist-90-R73 Beck depression inventory72 Impact of event scale75 Michigan alcohol screening test77 Beck anxiety inventory71

Administration time (minutes)

Psychometric properties

Test-retest reliabilities ranging from .84 to .87 Test-retest reliabilities are .79 for frequency of uplifts and .60 for intensity of hassles

Assesses psychologic demands, decision latitude, social support and job insecurity. Global measure of stressful life events. Helps to identify work and non-work related stressors along with positive aspects of work and personal life.

30

Measures three aspects of burnout: (1) emotional exhaustion; (2) lack of sense of personal accomplishment; and (3) depersonalization. A multidimensional measure of symptoms associated with psychologic distress and a wide variety of psychiatric disorders. A unidimensional scale that measures severity of depression.

30

Test-retest reliability ranging from .79 to .89

15

A multidimensional scale used to assess symptoms associated with post-traumatic stress disorder. A scale that measures past and present alcohol consumption. A unidimensional scale that measures severity of anxiety.

15

Internal consistency ranges from .77 to .90; test-retest reliability is .78 to .90 Test-retest reliability of .90, internal consistency of .86, and coefficient alpha of .94 Coefficient alpha of .42 to .82

15

Coefficient alpha of .95

15

Test-retest reliability of .89, internal consistency of .85, and coefficient alpha of .87

Table 28.3.2 Instruments used to screen for stressors and psychiatric symptoms in the workplace

30

15

690 Psychosocial Stressors and Psychiatric Disorders in the Workplace

Psychiatric disorders Depressive and anxiety disorders Comparisons across occupations show that depressive disorders, in particular, are prevalent in the workplace. Diagnostic criteria common to all of the depressive disorders (major depression, dysthymia) include depressed mood and a loss of interest or pleasure in usual activities with symptoms of decreased or increased appetite, sleep disturbance, poor concentration, fatigue, and thoughts of death. Anxiety disorders include generalized anxiety disorder, panic disorder, and phobias and are characterized by excessive worry and a number of psychologic and physical symptoms such as shortness of breath, heart pounding, sweating, chest pain, nausea, and fear of losing control. Phobias can range from those that are specific to situations or things to a fear of being outside of the home (i.e., agoraphobia). The same screening tools suggested under stress disorders can also be useful to screen for depression and anxiety disorders with severity greater in the individual who has a diagnosable psychiatric disorder. Post-traumatic stress disorder (PTSD) is an anxiety disorder specifically related to traumatic exposures. In the workplace, traumatic events include death or serious injury to self or others and workplace violence. For example, NIOSH data reveal that homicide is the second leading cause of occupational death, following work-related motor vehicle accidents, and surpassing machine-related deaths.49 The individual’s response to a traumatic event such as seeing a coworker seriously injured or killed involves intense fear, helplessness, and horror. Symptoms of PTSD include intrusive thoughts/dreams and recollections of the trauma, reexperiencing the trauma, and avoidance of stimuli that arouse recollection of the trauma. PTSD is not inevitable following a traumatic experience and can be prevented with debriefing and counseling immediately following an event.50

Substance abuse and dependence A vast literature documents the occurrence and deleterious effects of substance abuse in the workplace. Workers showing a pattern of poor performance that includes absenteeism, lateness, conflicts with coworkers, and frequent illness may alert the occupational health professional to a substance abuse problem. However, as is evident from the list, these work-related symptoms are not unique to substance abuse and may have numerous causes. Thus, diagnosing a substance abuse problem as a non-mental health professional, and in the absence of information about the individual’s pattern of substance use, is difficult and potentially harmful. Denial is a major psychologic dimension of substance abuse and therefore, if the diagnosis is suggested to the individual with inadequate documentation, it may serve to reinforce denial. Diagnosis of substance abuse includes an inability to control the use of a mood-altering substance and is not confined to alcohol or illicit drugs but also includes prescription and non-prescription medications. Abuse is

not determined by quantity of use but rather by the fact that the individual continues to use the substance despite adverse medical, social, family, or occupational consequences. Thus, the diagnosis is behavioral and does not depend on the amount or type of substance used. To make this point more clearly, virtually the same criteria are applied to diagnose pathologic gambling.45

Psychoses Psychotic disorders are characterized by delusions, hallucinations, incoherent speech, disorganized behavior, and flat or inappropriate affect. Since psychoses are frequently diagnosed in young adulthood, some individuals may have their first episode while employed. Psychotic individuals are often not employed, but with medication may be returned to a workplace with proper accommodations.

STRATEGIES FOR PREVENTION Primary prevention What can we do to make work less stressful and individuals less vulnerable to what may, to some extent, be an inevitable part of work? Jobs where workers have little control over decisions seem to produce greater risk for psychiatric disorder. Therefore, programs to enhance employee participation in decisions are interventions that may prove to reduce psychiatric morbidity. For example, Karasek recommended that specific Quality of Work Life programs (QWL) be used by organizations to ensure that employees feel a sense of empowerment. More specifically, this process enables workers to set their own goals, make decisions, and solve problems within their sphere of responsibility. Although QWL interventions have focused primarily on lower-level employees, Quality Circles (QC), which are often found within various organizational strata (e.g., middle management, blue-collar workers), include groups of volunteers who work together on a particular job, meet regularly, and discuss job-related problems and possible solutions. Porras and Silvers51 have noted that QCs have a positive impact on employee attitudes, which may decrease job strain and burnout. However, the effect of QCs on job productivity is more equivocal.52 On an individual basis, wellness programs that enhance resilience through better diet, exercise, and relaxation may also improve mental health. These programs typically include stress management workshops, health risk assessments, exercise facilities, subsidized cafeterias, individual counseling, and seminars and lectures. Programs that reduce work–family conflict can also reduce the incidence of psychiatric disorder. For example, employee-sponsored daycare centers have been demonstrated to increase job satisfaction and decrease work-related stress on working mothers.53 Other factors such as flexible work hours54 and parental leave programs55 have observed similar findings. Perhaps the existence of these programs helps the employer or organization to acknowledge and support the challenges that employees often face. Also, these programs help

Strategies for Prevention 691 employees feel more empowered and valued as individuals who make significant contributions at the workplace. Some preliminary research indicates that when upper level managers support wellness programs, and when employees have easy access to these services, notable decreases in medical expenses and increases in productivity56 are typically seen. Accurately assessing physical and psychosocial hazards within the workplace is an important step toward preventing psychiatric disorders. Reducing exposure to neurotoxicants and to hazardous working conditions need to be considered when increased rates of symptoms and psychiatric disturbance are noted. For example, reducing exposure to neurotoxicants through engineering controls and programs to train employees in the use of protective equipment will reduce exposure and disability. NIOSH has provided specific recommendations for shift workers to reduce the health effects of working non-daytime hours. Also, debriefing programs are standard operating procedures for emergency workers such as police following traumatic exposures. Such prevention programs are examples of proactive efforts to reduce psychiatric morbidity based on knowledge of the inherent risks of the work.

Secondary prevention Employee Assistance Programs (EAP) are offered in over 20,000 US companies. The purpose of these programs is to detect and treat individuals with psychiatric disorders.57 Initially, EAPs were developed in recognition of the impact substance abuse has on work productivity. However, it became apparent that other psychiatric disorders and psychosocial problems such as elder care, child care, and financial management also significantly impact work function. Therefore, EAPs broadened their scope to include evaluation and referral services for any personal problem. These ‘broad brush’ programs encourage employees to seek services on their own or as ‘self-referrals’ rather than wait until job performance suffers. Some investigators have reported significant improvement in absenteeism, lost time, warnings and supervisors ratings of performance58-62 when comparing these indicators before and after EAP counseling. However, despite the proliferation of EAPs and widespread claims of their cost-effectiveness, more evaluation is needed. Although EAPs have increased access to psychiatric services, there is a paucity of literature to document the effects of psychiatric treatment on work-related variables. Mintz et al.63 reviewed the literature addressing the effect of psychiatric treatment on the capacity to work for those diagnosed with drug addiction, alcoholism, anxiety and affective disorders, gambling, and schizophrenia, and concluded that most attention has been given to work outcomes for substance abusers. The authors found that long-term treatment was not more beneficial than standard alcohol treatment regimens and that successful treatment aimed at reducing abusive drinking increased productivity. In a separate review, data on occupational outcomes from 10 treatment studies for depression were analyzed.

Symptom improvements occurred more rapidly than improvements in work-related variables, such as missed time, lower productivity, and interpersonal problems, and were not affected by treatment duration. Work outcomes improved, however, as treatment duration increased, with maximum benefits achieved at 4–6 months. For schizophrenia, neuroleptic drugs reduce symptoms, but some studies suggest that they may also adversely affect work capacity by interfering with the learning process.64 Overall, the most striking finding was the lack of attention in the psychiatric outcome research to the effects of treatment on functional work capacity, despite the stated importance of occupational impairment inherent in the DSM-IV criteria for most psychiatric disorders.45 In sum, psychiatric disorders and associated behavioral problems, such as alcohol consumption, significantly impact productivity – regardless of their cause or relationship to worksite factors and stressors. From the data available, either through program evaluation of EAPs or in the general treatment outcome literature, it appears that when employed individuals are treated, their work improves. This finding is encouraging and further supports the importance of health insurance benefits that include psychiatric treatment.

Tertiary prevention – fitness for duty When an employee has been out of work for psychiatric treatment (e.g., depression or anxiety) or a question arises about the employee’s ability to function on the job, a fitness for duty evaluation may be requested. Fitness for duty is defined as the individual’s ability to perform a job based on the specific job requirements. Therefore, this evaluation requires a detailed understanding of the job duties. Often this can be problematic since job descriptions are not necessarily informative or sufficiently behavioral in their descriptions (see Table 28.3.3 for guidelines). Ancillary materials such as interviews with workers in similar positions or with supervisors may be needed to understand the essential behaviors expected to perform the job. Fitness for duty can never be based solely on psychiatric diagnosis but rather must be based on a behavioral analysis of the employee’s abilities. Past job performance is the best predictor of future job performance. Further, determining a global assessment of functioning can be useful as a behavioral guide for the individual’s current level of function and ability to perform daily tasks related to work.65 Overall, matching an assessment of the employee’s current behavioral function with the essential functions required to perform a job, along with consideration of the employee’s premorbid level of function on the job, will give the best prediction of an employee’s fitness for return to a job.

Accommodation in the workplace Since the Americans with Disabilities Act of 1990 (ADA) was passed, employers have been under pressure to employ

692 Psychosocial Stressors and Psychiatric Disorders in the Workplace Steps involved in constructing an observational and behavioral based performance analysis

Sample occupation: Executive Assistant

Sample occupation: Baker

Step 1: Observation: observe a worker complete job tasks from preparation to finish Step 2: Describe: develop a description of job tasks in observable and behaviorally oriented terms

Watch a sample of administrative assistants perform daily work duties

Step 3: Rank-order: by consulting supervisors and workers, rank-order from highest to lowest the tasks that are most crucial to a job

Through both direct observation and through interviews with present workers and supervisors, determine the most important aspects of the job; determine if others are available to help if parts of job cannot be completed

Step 4: Develop worksheet: construct a worksheet to be given to individuals conducting fitness for duty evaluations which include key components of performance analysis

Give list to the individual conducting the assessment to ensure that they have an understanding of duties required to work effectively as an administrative assistant

Verbal fluency skills to communicate effectively on phone and faceto-face with coworkers; attention and concentration to organize complex schedule; motor skills needed to type, learn and remember software programs (e.g., word processors. spreadsheets, etc.)

Watch a sample of bakers perform the duties involved in the production of bread Read recipe card; motor skills and coordination to handle oven and mixing; psychomotor speed to safely retrieve finished product from oven Through both direct observation and interviews with present workers and supervisors, determine the most important aspects of the job; determine if others are available to help if parts of job cannot be completed Give list to the individual conducting the assessment to ensure that they have an understanding of duties required to work effectively as a baker

Table 28.3.3 Observationally and behaviorally based performance

and accommodate individuals with disabilities, including psychiatric illness. The number of discrimination claims against employers based on emotional/psychiatric impairment has also increased since the passage of this legislation. For example, in 1997 the Equal Employment Opportunity Commission (EEOC) reported that 15% of discrimination claims were due to emotional/psychiatric impairment, which represented the largest category of claims in that year. The ADA prohibits discrimination based on disability and provides that employers must make ‘reasonable accommodations’ to the disabilities of ‘qualified’ applicants so long as this does not impose ‘undue hardship’. ‘Qualified’ means that the individual can perform the essential functions of the job, except for the disability. ‘Reasonable accommodation’ refers to any modification or adjustment to a job or work environment that will allow the qualified employee with the disability to perform the job functions. ‘Undue hardship’ refers to an action requiring significant difficulty or expense.66 Employers are not allowed to inquire about a disability prior to employment, and the applicant does not have to reveal a psychiatric history at time of hire. Moreover, if a long-term employee who was previously performing the job develops a psychiatric disorder, the employer is obligated to make accommodations.67 For people who are hospitalized for psychiatric diagnoses such as schizophrenia, employment rates have traditionally been low, with less than 20% of this group employed.67 The best predictors of future work performance among the chronically mentally ill seem to be ratings of the individual’s work adjustment in a sheltered job site, the ability to function socially with others, and prior employment history.68,69 Thus, diagnosis alone (e.g., psychotic vs. nonpsychotic) is not as predictive of work capacity as is objective behavioral performance. While these findings apply

Management strategies Review strengths and weaknesses with employees Set behavioral goals for job performance Deliver positive feedback and criticism in constructive manner Meet regularly with employees Be flexible in administrative policies (e.g., allow relaxation breaks for anxiety) Accommodations for workers with psychiatric disabilities Utilize flextime when needed Consider job sharing Arrange environment to reduce excess noise or visual distractions Extend leave time Allow workers to call supportive individual during day (e.g., family, friends) Join meetings between employer, supervisor and employment service * As noted at http://janweb.icdi.wvu.edu/kinder/pages/psychiatric.html

Table 28.3.4 Employing and accommodating workers with psychiatric disabilities*

specifically to the psychoses, the same guidelines are applicable for any physical or psychiatric illness. The Mental Health Law Project’s guidebook70 on the ADA provides a helpful document outlining reasonable accommodations for individuals with psychiatric disabilities. Accommodations include analysis of the individual employee’s behavioral problems, such as anxiety, and sensitivity to criticism, followed by the development of accommodations based on individual needs (see Table 28.3.4 for guidelines).

SUMMARY AND CONCLUSIONS With increasing attention to psychiatric disorders in the workplace and growing concerns about how working conditions contribute to psychiatric illness, the occupational

Summary and Conclusions 693 health practitioner will be called upon to prevent, identify, and manage psychiatric disorders. Developing effective screening programs to identify stressful working conditions and early signs of psychiatric symptoms can help reduce lost productivity and disability. As for most occupational health programs, these efforts will require close coordination between the occupational health professional, management, and employees to develop meaningful programs that will succeed in reducing psychiatric disability. Although sometimes less apparent to the observer, the disabilities associated with psychiatric illness deserve accommodation and will require creative approaches to help managers overcome the biases toward mental illness that still exist in the greater community and the workplace.

References 1. Millar JD. Mental health and the workplace: an interchangeable partnership. Am Psychol 1990; 45:1165–6. 2. Hurley J, Horowitz J. Alcohol and health. New York: Sphere Publishing, 1990. 3. Harwood JJ, Kristiansen P, Rachal JV. Social and economic costs of alcohol abuse and alcoholism. Issue Report No. 2. Research Triangle Park, NC: Research Triangle Institute, 1985. 4. Ames GM, Grube JW, Moore RS. The relationship of drinking and hangovers to workplace problems: an empirical study. J Stud Alcohol 1997; 58:37–47. 5. Mandell W, Eaton WW, Anthony JC, et al. Alcoholism and occupations: a review and analysis of 104 occupations. Alcohol Clin Exp Res 1992; 16:734–46. 6. Karasek RA. Job demands, job decision latitude, and mental strain: implications for job redesign. Adm Sci Q 1979; 24:285–308. 7. Hemmingsson T, Lundberg I. Work control, work demands, and work social support in relation to alcoholism among young men. Alcohol Clin Exp Res 1998; 22:921–7. 8. Stansfeld SA, Fuhrer R, Head J, et al. Work and psychiatric disorder in the Whitehall II Study. J Psychosom Res 1997; 43:73–81. 9. Mausner-Dorsch H, Easton WW. Psychosocial work environment and depression: epidemiologic assessment of the demand–control model. Am J Public Health 2000; 90:1765–70. 10. Appelberg K, Romanov K, Heikkila K, et al. Interpersonal conflict as a predictor of work disability: a follow-up study of 15,348 Finnish employees. J Psychosom Res 1996; 40:157–67. 11. Duxbury LE, Higgins CA. Gender differences in work–family conflict. J Appl Psychol 1991; 76:60–74. 12. Hochschild A, MacHung A. The second shift. New York: Avon Books, 1990. 13. Eskanazi B, Maizlish N. Effects of occupational exposure to chemicals on neurobehavioural functioning. In: Tarter RE, VanThiel DH, Edward KL, eds. Medical neuropsychology, the impact of disease on behavior. New York: Plenum, 1988;223–63. 14. Schottenfeld RS, Cullen MR. Organic affective illness associated with lead intoxication. Am J Psychiatry 1984; 141:1423–6. 15. Johnson BL, Baker EL, El Batawi M, et al. Prevention of neurotoxic illness in working populations. New York: John Wiley and Sons, 1987. 16. NIOSH. Bulletin 48. Organic solvent neurotoxicity. Washington DC: National Institute for Occupational Safety and Health, 1987. 17. World Health Organization. Chronic effects of organic solvents on the central nervous system and diagnostic criteria. Copenhagen: WHO Regional Office for Europe, 1985.

18. Morrow LA, Ryan CM, Goldstein G, et al. A distinct pattern of personality disturbance following exposure to mixtures of organic solvents. J Occup Med 1989; 31:743–6. 19. Schottenfeld RS, Cullen MR. Recognition of occupationinduced post traumatic stress disorders. J Occup Med 1986; 28:365–9. 20. Goldblum D, Chouinard G. Schizophrenia psychosis associated with chronic industrial toluene exposure: case report. J Clin Psychiatry 1985; 46:350–1. 21. Dager SR, Holland JP, Cowley DS, et al. Panic disorder precipitated by exposure to organic solvents in the work place. Am J Psychiatry 1987; 144:1056–8. 22. Ely EW, Moorehead B, Haponik EF. Warehouse workers’ headache: emergency evaluation and management of 30 patients with carbon monoxide poisoning. Am J Med 1995; 98:145–55. 23. Lezak MD. Neuropsychological assessment. New York: Oxford University Press, 1995 24. Meredith T, Vale A. Carbon monoxide poisoning. Br Med J 1988; 296:77–9. 25. Min SK. A brain syndrome associated with delayed neuropsychiatric sequelae following acute carbon monoxide intoxication. Acta Psychiatr Scand 1986; 73:80–6. 26. Ernst A, Zibrak JD. Carbon monoxide poisoning. N Engl J Med 1998; 339:1603–8. 27. Choi IS. Delayed neurologic sequelae in carbon monoxide intoxication. Arch Neurol 1983; 40:433–5. 28. Maghazaji HI. Psychiatric aspects of methylmercury poisoning. J Neurol Neurosurg Psychiatry 1974; 37:954–8. 29. Gross LS, Nagy RM. Neuropsychiatric aspects of poisonous and toxic disorders. In: Yudofsky SC, Hales RE, eds. American Psychiatric Press textbook of psychiatry, 2nd edn. Washington DC: American Psychiatric Press, 1992. 30. Menzies D, Bourbeau J. Building-related illnesses. N Engl J Med 1997; 337:1524–31. 31. Redlich CA, Sparer J, Cullen MR. Sick-building syndrome. Lancet 1997; 349:1013–6. 32. Bauer RM, Greve KW, Besch EL, et al. The role of psychological factors in the report of building-related symptoms in sick building syndrome. J Consult Clin Psychol 1992; 60:213–9. 33. Loewen L, Suedfeld P. Cognitive and arousal effects of masking office noise. Environ Behav 1992; 24:381–95. 34. Bing-shuang H, Yue-lin Y, Ren-yi W, et al. Evaluation of depressive symptoms in workers exposed to industrial noise. Homeostasis Health Dis 1997; 38:123–5. 35. Dejoy DM. Information input rate, control over task pacing, and performance during and after noise exposure. J Gen Psychol 1985; 11:229–42. 36. Akerstedt T. Psychological and psychophysiologic effects of shift work. Scand J Work Environ Health 1990; 16:67–73. 37. Frese M, Semmer N. Shiftwork, stress, and psychosomatic complaints: a comparison between workers in different shiftwork schedules, non-shiftworkers, and former shiftworkers. Ergonomics 1986; 29:99–114. 38. Verhaegen P, Dirkx J, Maasen A, et al. Subjective health after twelve years of shift work. In: Haider M, ed. Night and shiftwork: longterm effects and their prevention. Frankfurt Am Main: Verlag Peter Lang, 1986;67–74. 39. Costa G, Apostali P, D’Andrea F, et al. Gastrointestinal and neurotic disorders in textile shift workers. In: Reinberg A, Vieux N, Andlauer P, eds. Night and shift work: biologic and social aspects. Oxford: Pergamon Press, 1981;187–96. 40. Harma M, Illmarinen J, Knauth P. Physical fitness and other individual factors relating to the shiftwork tolerance of women. Chronobiol Int 1988; 5:417–24. 41. Tasto DL, Colligan MJ, Skjel EW, et al. Health consequences of shift work (SRI Project URU-4426). Washington DC: US Department of Health, Education and Welfare, NIOSH, 1978. 42. Cole R, Loving R, Kripke D. Psychiatric aspects of shiftwork. In: Scott A, ed. Occupational medicine: state of the art reviews. Philadelphia: Hanley & Belfus, 1990.

694 Psychosocial Stressors and Psychiatric Disorders in the Workplace 43. Keyserling WM. Occupational safety preventing accidents and overt trauma. In: Levy BS, Wegman DH, eds. Occupational health: recognizing and preventing work-related disease and injury, 4th edn. Philadelphia: Lippincott Williams & Wilkins, 2000;181–94. 44. Regehr C, Hill J, Glancy G. Individual predictors of traumatic reactions in firefighters. J Nerv Ment Dis 2000; 188:333–9. 45. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th edn. Washington DC: APA, 1994. 46. Kawamoto T, Koga M, Murata K, et al. Effects of ALDH2, CYP1A1, and CYP2E1 genetic polymorphisms and smoking and drinking habits on toluene metabolism in humans. Toxicol Appl Pharmacol 1995; 133:295–304. 47. Soderkvist P, Ahmadi A, Akerback A, et al. Glutathione Stransferase M1 null genotype as a risk modifier for solventinduced chronic toxic encephalopathy. Scand J Work Environ Health 1996; 22:360–3. 48. Maslach C, Jackson SE, Lieter MP. Maslach burnout inventory manual. Palo Alto: Consulting Psychologists Press, 1996 49. NIOSH. Bulletin 57. Violence in the workplace: risk factors and prevention strategies. Washington DC: National Institute of Occupational Safety and Health, 1996. 50. Tucker P, Trautman R. Understanding and treating PTSD: past, present, and future. Bull Menninger Clin 2000; 64:37–51. 51. Porras JL, Silvers RC. Organizational development and transformation. Ann Rev Psychol 1991; 42:51–78. 52. Cordery KL, Mueller WS, Smith LM. Attitudinal and behavioral effects of autonomous group work: a longitudinal field study. Acad Manage J 1991; 34:464–76. 53. Warren J, Johnson P. The impact of workplace support on work–family role strain in family relations. J Appl Fam Child Stud 1995; 44:163–9. 54. Barling J, Barenbug A. Some personal consequences of ‘flextime’ work schedules. J Soc Psychol 1984; 123:137–8. 55. McGovern P, Dowd B, Gherdingen D, et al. Time off work and the postpartum health of employed women. Med Care 1997; 35:507–21. 56. Caudron S. The wellness payoff. Pers J 1990; 69:54–60. 57. Adamson DW, Gardner MD. Employee assistance programs and managed care: merge and converge. In: Sauber SR, ed. Managed mental healthcare: major diagnostic and treatment approaches. Bristol: Brunner/Mazel, Inc., 1997;67–82. 58. Cooper CL, Sadri G. The impact of stress counselling at work. J Soc Behav Pers 1991; 6:411–23. 59. Guppy A, Marsden J. Assisting employees with drinking problems: changes in mental health, job perceptions and work performance. Work Stress 1997; 11:341–50.

60. Mitchie S. Reducing absenteeism by stress management: valuation of a stress counselling service. Work Stress 1996;10:367–72. 61. Ramanathan CS. EAP’s response to personal stress and productivity: implications for occupational social work. Soc Work 1992; 37:234–9. 62. Walsh DC, Hingson RW, Merrigan DM, et al. A randomized trial of treatment options for alcohol-abusing workers. N Engl J Med 1991; 325:775–82. 63. Mintz J, Mintz LI, Arruda MJ, et al. Treatments of depression and the functional capacity to work. Arch Gen Psychiatry 1992; 49:761–8. 64. Hogarty GE, McEvoy JP, Munetz M, et al. Dose of fluphenazine, familial expressed emotion, and outcome in schizophrenia: results of a two-year controlled study. Arch Gen Psychiatry 1988; 45:797–805. 65. Sperry L. Psychiatric consultations in the workplace. Washington DC: American Psychiatric Press, 1993. 66. United States Department of Justice. The Americans with Disabilities Act: Questions and Answers. Washington DC: Civil Rights Division, US Department of Justice, 1991. 67. Carling PJ. Reasonable accommodations in the workplace for individuals with psychiatric disabilities. Consult Psychol J 1993; 46–62. 68. Anthony WA, Jansen MA. Predicting the vocational capacity of the chronically mentally ill. Am Psychol 1984; 39:537–44. 69. Massel HK, Liberman RP, Mintz J, et al. Evaluating the capacity to work of the mentally ill. Psychiatry 1990; 53:31–43. 70. Mental Health Law Project. Mental health consumers in the work place: how the Americans with Disabilities Act protects you against employment discrimination. Washington DC: Mental Health Law Project, 1992. 71. Beck AT. Beck anxiety inventory. San Antonio: The Psychological Corporation, 1993. 72. Beck AT. Beck depression inventory. San Antonio: The Psychological Corporation, 1978 73. Derogatis LR. SCL-90-R administration, scoring and procedures manual-III. Towson: Clinical Psychometric Research, 1983. 74. Dohrenwend BS, Krasnoff L, Askenasy AR, et al. Exemplification of a method for scaling life events: the peri life events scale. J Health Soc Behav 1978; 19:205–29. 75. Horowitz M, Wilner N, Alvarez W. Impact of event scale: a measure of subjective stress. Psychosom Med 1979; 41:209–18. 76. Karasek R, Theorell T. Healthy work. New York: Basic Books, 1990 77. Selzer ML, Vinokur A, Van Rooijen L. A self-administered short Michigan screening test. J Stud Alcohol 1975; 36:117–26.

Chapter 29 Dermatologic Diseases 29.1 Contact Dermatitis Kalman L Watsky, Christina A Herrick, Elizabeth F Sherertz, Frances J Storrs The types of contact dermatitis that may be associated with the workplace are listed in Table 29.1.1 and are the focus of this chapter. By far the most common type is irritant contact dermatitis (ICD), representing approximately 80% of occupational contact dermatitis, with the remaining 20% being allergic. Other occupational skin diseases are discussed in Chapters 20.2 and 21.9. The patient does not walk in with an established diagnosis, however, and one needs to start with a systematic approach to evaluate the individual with skin disease.

APPROACH TO THE WORKER WITH SKIN DISEASE The key elements to consider in evaluating a patient with skin disease are summarized in Table 29.1.2. It is most useful to see the patient at a time when the dermatitis is active, but this is not always possible. Speaking directly to the patient rather than relying on the history gathered by the nurse, the physician, or the supervisor is helpful.

History of dermatitis The patient’s description of events at the onset of the skin problem can be very important to the ensuing investigation. The date of onset generally should be during employment at the job in question. However, a person may have worked at a previous job with similar tasks and exposures, and developed contact allergy or irritation that could recur or be aggravated by the current job. If the person had a previous dermatologic disease (e.g., atopic dermatitis, psoriasis), a history of the time course of worsening of the dermatitis should be sought. The rapidity with which symptoms and signs occurred can give important clues. Immediate (or within hours) onset of itching and swelling with hive-like lesions suggests contact urticaria. Burning, stinging, and a red, dried appearance at the time of an exposure suggest an acute irritant reaction. Chronic irritant dermatitis can take much longer to develop. Approximately 3 months of constant wetting and drying is the typical time course in wet-work jobs.1 Allergic contact dermatitis (ACD) may occur in a less predictable manner; it may occur within weeks of exposure to a new material, or it may develop after months or years. Itching, blistering, and spreading of lesions are common at some time in the course of contact allergic dermatitis (see later). The anatomic site of the lesions at the onset of dermatitis also is important. The hands are the most likely site, and it should

be noted which hand and which site on the hand (dorsum, fingers, palm) was first involved. In general, the initial site of involvement should correspond to sites exposed during a job task to provide evidence for occupational dermatitis. Later in the course of the condition, other body sites may become involved that may not clearly be related to direct exposure. The course of the dermatitis related to time away from work is important in trying to establish evidence for workrelatedness. Documenting improvement away from the initial worksite is supportive evidence that work exposures are playing a role. It is important to note treatments undertaken during the time away from work, because some treatments (e.g., systemic corticosteroids) may suppress an ongoing problem and thus confuse the temporal relationship. Lack of improvement away from work is more difficult to assess: some chronic dermatoses may be very slow to improve even with time off from work, therapy, and protection. Activities undertaken away from work (household chores, hobbies) also may cloud the issue. At times, workers are moved to light duty or modified type of work after a skin problem develops. It is important to try to determine what differences there are between initial and modified duties, particularly with regard to the type and frequency of materials contacted, hand washing, and protective equipment. These differences also can give clues to which procedure may be contributing to the dermatitis. The time course of recognition and treatment of the dermatitis also is important, as this may modify the appearance of the dermatitis even before a physician has seen the patient. Depending on the initial severity of a dermatitis, an employee may treat himself or herself for weeks or promptly seek medical evaluation. Encourage the patient to bring pharmacy printouts, bags of medications, notes from other physicians, or any other information to try to compile a history of topical and systemic medications. Topical treatments are of particular importance, because sometimes ingredients in these products can cause irritation or allergic reactions themselves, thus perpetuating what may have been a self-limited problem. Treatment with systemic corticosteroids that leads to clearing of the dermatitis and is followed by a subsequent flare of the dermatitis despite continued absence from work suggests that non-work factors may be playing a role. These issues further complicate assessment of the patient.

Occupational history General principles for taking an occupational history are discussed in Chapter 1. When considering patients with

696 Contact Dermatitis Irritant contact dermatitis Single exposure dermatitis Cumulative irritant dermatitis Frictional irritant dermatitis Psoriasiform dermatitis Fiberglass dermatitis Low humidity dermatitis Post-traumatic eczema Allergic contact dermatitis Contact urticaria* Contact photodermatitis Phototoxic Photoallergic

II.

III.

*Urticaria appears on the list as it may eventuate into dermatitis (see text).

Table 29.1.1 Occupational contact dermatitis

I.

History A. Present illness Date of onset Body site at onset Patient description Onset – abrupt or gradual Appearance, spread Frequency Effect of treatment Course of disease Effect of weekend, vacation Work procedure change Treatment and effect on dermatitis B. Occupational information Current employer Employment dates Job title At time of onset Description of job tasks Materials contacted Protection Water exposure Hand washing Clothing/equipment Protective creams/cleansers Skin cleaning Method and frequency Other workers affected Job since dermatitis Previous job tasks or jobs Episodes of dermatitis Second job Dates of disability Date of job changes C. Personal history Other exposures Animals Foods Plants Clothing Personal care products Hobbies Past history of skin disease Plant dermatitis Hand dermatitis Psoriasis Athlete’s foot History of atopy Personal/family Atopic dermatitis Hay fever Asthma Medical problems Medications

IV.

Prescribed Over-the-counter Physical examination Lesion type Secondary changes Distribution Other skin disease Photographic documentation Diagnostic techniques Skin scrapings Fungus Fibers Culture Skin biopsy Patch test Contact urticaria test Photopatch test Supplemental information Material safety data sheets Medical records Workplace Other physician

Table 29.1.2 Clinical evaluation of the worker with skin disease

skin problems, it is useful to have patients explain their routine, having them use their hands to demonstrate how they handle materials. Ask specifically about the amount of time spent each day with wet exposure, wearing protective clothing, and frequency of contact with irritating chemicals. If items or material safety data sheets (MSDS) are brought from the workplace, go over each item with the patient to determine the type of contact and the frequency of exposure. Repetitive motion tasks or handling materials that could cause friction or trauma to the skin should be noted. Information about the physical environment in the workplace, such as temperature, humidity, patterns of airflow, and exposure to ultraviolet light, should be sought. It also is important to ask about other intermittent job tasks that may have a relationship to the onset of dermatitis, such as periodic cleaning routines, machinery maintenance, changing cutting oils, and overtime work. There may be a discrepancy between the specific type of protective clothing (e.g., gloves) recommended and the actual use of these items in the workplace. Does the worker follow the recommended use, is protection optional, or does the worker have problems with consistent usage because he or she perceives the protection as inadequate (e.g., gloves tear, get saturated, or hamper dexterity)? Asking whether other workers are affected who work at the same job can be helpful, although a site visit or examination of the other workers, when feasible, is more valuable. The sudden development of dermatitis by many workers suggests a breakdown in housekeeping or the recent introduction of an irritant into the industrial process. It also is helpful to ask the worker what he or she believes is the cause. The question may narrow down the relevant part of the job task. Inquiring whether or not the patient likes the job and wants to continue (if the dermatitis can be improved) may be significant, because motivation may be a helpful factor in predicting prognosis, depending on the type of dermatitis involved. This question also may help uncover the rare malingerer. For example, if a diagnosis of

Approach to the Worker with Skin Disease 697 ICD is made and a specific glove is recommended, the motivated employee may be more likely to use the protective glove to reduce the dermatitis and stay on the job. With regard to skin disease, it also is important to inquire about previous work descriptions, exposures, and episodes of dermatitis, as well as second jobs or combined schooling with a job. One of our patients was concerned that her hand dermatitis was due to her office job. She also was going to school to become a hairdresser, and the wetwork exposures in this setting were much more likely to be the cause of her dermatitis.

Personal history Obtaining a history of previous skin disease should be a focal point in talking with the patient. This history may provide clues to the current skin problem and may have prognostic implications.

Atopic history Childhood atopic dermatitis (atopic eczema) is fairly common (greater than 10% of children in the United States), and is characterized by dry, pruritic skin, usually appearing in flexural areas. The skin in an atopic individual is more susceptible to irritants, such as rough fibers, and to changes in the environment, such as wet–dry and hot–cold changes. Determining an atopic background most often is done by eliciting a personal and/or family history (in first-degree relatives) of atopic dermatitis (eczema), hay fever and allergic rhinitis, or childhood asthma. A history of one or more of these conditions may be elicited in up to 25% of individuals, indicating that atopy is common. A history of childhood eczema is a common factor in adults who develop hand dermatitis. In his study of compensated skin disease in South Carolina, Shmunes found that the relative odds of developing work-related skin disease were 13.5 times greater for an individual with a history of atopic skin disease.2 Further, the course of work-related dermatitis may be more prolonged in an atopic patient. Psoriasis is a common skin disorder that may have potential for aggravation or development of new lesions in response to occupational factors, especially friction or repeated trauma involving the hands. Psoriasis on the hands as a result of work-related trauma may appear to be dermatitis at first glance. The presence of psoriasis should not fool the physician because ACD may coexist with psoriasis.3 A history of skin contact allergy to jewelry, rubber articles, plants such as poison ivy, or other materials may provide a clue to the current dermatitis. Other information in the personal history that should be sought is listed in Table 29.1.2. The patient’s personal hygiene routine may give hints of other potential contact allergen exposure. Oral medications may predispose the person to special problems. For example, use of a medication that is potentially phototoxic (e.g., tetracycline) while continuing to work at a job with significant ultraviolet exposure could lead to severe sunburn. Hobbies are often sources of contact with potential allergens and irritants. Emphasis on pre-existing skin diseases as well as non-work

exposures is essential in that surveys of skin diseases occurring in the workplace have shown that a minority of them are fully related to the patient’s work.4

Physical examination The physical examination should encompass not only the affected area of skin but also hair, nails, and sites not directly affected by workplace exposure. Evidence of other skin disease, especially flexural eczema, psoriasis, dermatophyte fungal infection, acne, and acute or chronic sun damage, should be noted. The size, color, and type of primary lesion (macule, papule, vesicle, wheal, pustule), as well as the presence of secondary lesions, such as scales, crusts, fissures, erosions, or ulcerations, should be noted. The configuration of lesions, linear grouping, or cut-off at sun-exposed sites or sites of protective equipment such as gloves or facemasks should be noted. Patterns of localization of dermatitis on the hands can occasionally be useful. Primary web space involvement is usually, but not always, irritant. Vesicles localized on the fingertips often are associated with allergy. Note whether both dorsal and palmar surfaces are involved. Nail changes may indicate chronicity of lesions. See the discussion about specific disorders for more detail on clinical clues. Often, however, it is not possible to distinguish clinically the type of dermatitis without further diagnostic testing. Recording the distribution of lesions on a drawing of the body or hands is helpful.

Diagnostic techniques A number of techniques can be helpful in the evaluation of a patient with possible occupational dermatitis. In general, an experienced physician (e.g., dermatologist) should perform and interpret these tests. If scaling is present, a skin scraping to which 10–20% potassium hydroxide (KOH) is applied may be examined under a microscope for evidence of fungal hyphal elements. The same technique may be used to look for irritating fibers (e.g., fiberglass). Cultures should be considered if a primary or secondary infection is suspected. Bacterial superinfection (impetigo) is common in crusted eczematous dermatoses, and a cotton swab sample taken from beneath a crust or a ruptured pustule may be submitted for bacterial culture, in the search for Staphylococcus aureus or group A β-hemolytic Streptococcus in particular. Marginated scaling lesions that are suspected to be the result of dermatophyte or Candida infection may be cultured from scrapings of scale using fungal media. If herpetic viral infection is suspected, cultures can be obtained from vesicular lesions if the appropriate medium is available. Culture or typing for other viral infection, such as wart papilloma virus, can only be performed if special facilities are available. At times, a skin biopsy may be helpful to characterize the microscopic inflammatory pattern, confirm the presence of neoplastic lesion, or establish a cutaneous diagnosis of a nondermatitis type. Skin biopsy of dermatitis and eczema may be non-specific. The biopsy technique usually is performed with local anesthesia, a small punch biopsy, and suture closure.

698 Contact Dermatitis The following diagnostic techniques can be especially useful in evaluating skin disease that may be wholly or partially occupational in origin. These techniques in particular should be reserved for physicians trained in their usefulness, pitfalls, methodology, and interpretation.

Patch testing Diagnostic epicutaneous patch tests can help establish contact allergy (delayed contact hypersensitivity) to a given material. The concentration of the material to be tested should be standardized. Physicians with little experience

in patch testing should only test with known substances in concentrations that will not give an irritant reaction when applied to the skin. Commercially available patch tests that have been developed under Food and Drug Administration (FDA) guidelines are available in the United States for common cutaneous allergens such as rubber additives, formaldehyde and related preservatives, nickel, and neomycin. In Canada and Europe, a wider selection of allergens can be purchased. Common allergens in occupational ACD are listed in Table 29.1.3. Many of these allergens also are common outside the workplace. There

Allergen

Patch testing concentration**

Acrylates Balsam of Peru Benzocaine‡ Benzoyl peroxide medication Black rubber chemicals‡

1–5% 25% 5% 1% 0.6%

2-Bromo-2-nitropropane-1,3-diol (Bronopol)*† Carbamates‡ Chloroxylenol (P-chloro-) m-xylenol

0.5%

Cinnamic aldehyde‡ Cobalt chloride‡

1% 1%

Diaminodiphenylmethane

0.5%

Diazolidinyl urea (Germall II)*† Disperse yellow 3 Epoxy resin‡ Ethylenediamine dihydrochloride‡ Formaldehyde‡ Fragrance mix Glutaraldehyde Glyceryl monothioglycolate Imidazolidinyl urea*†‡ 5-Chloro-2-methyl-4-isothiazolin-3one and 2-methyl-4-isothiazolin-3one (Kathon CG)* Lanolin (wool wax alcohol)* Mercaptobenzothiazole‡

1% petrolatum or water 1% 1% 1% 1% water 8% 1% water 1% 2% petrolatum or water 0.1% water or petrolatum

P-Methylaminophenol sulfate (Metol) Neomycin‡ Nickel sulfate‡ Para-tertiary-butylphenol formaldehyde resin‡ Phenol formaldehyde resin Para-phenylenediamine‡

1%

Topical medications Rubber accelerator (especially gloves), fungicide (veterinarians), anticorrosive Black and white photo developer

20% 2.5% 1%

Topical antibiotic preparations Metal tools and devices, jewelry Neoprene plastics and glues (e.g., shoes)

5% 1%

Potassium dichromate‡ Propylene glycol

0.25% 10% water

Quaternium 15*†‡ (Dowicil 200) Rosin (colophony)*‡ Sesquiterpene lactone mix Thimerosal (merthiolate)

2% 20% 0.1% 0.1%

Tetramethylthiuram disulfide‡ Thioureas Tixocortol-21-pivalate

1% 1% 1%

Plastics, glues (e.g., plywood) Hair dye (humans and animals), para-amino chemical cross-reactor Leather fixative, anticorrosion chemical, paints Topical medicaments, tattoos, foods, brake fluids, antifreeze, plastics Cosmetic and industrial preservative Cosmetics, glues, soldering flux, anti-skid (violinists, athletes) Allergen in chrysanthemum and other weeds Preservative (e.g., vaccines), cross-reacts with some mercury compounds Rubber accelerator (especially shoes and gloves), fungicides Rubber accelerator Marker for corticosteroid allergy

1% 1%

30% 1%

Source examples Glues; paints, cosmetics (nails), dental appliances, artificial hips Perfume screen, plastics, medications Topical anesthetics, para-amino chemical cross-reactions Catalyst for acrylic resins, bleaching agent foods, topical Mixture of P-phenylenediamine related chemicals used in black rubber Cosmetic and industrial preservative Rubber, fungicides Photography, rubber, glues, photocopy, cosmetic and industrial preservative Plastics, flavoring, perfumes Nickel co-reactor, animal feeds, photography, acrylates, paints, glazes Epoxy and polyurethane curing agent, para-amino chemical cross-reactions Cosmetic and industrial preservative Nylon dyes Plastics, paints, glues Fluxes, stabilizer in topical medications Preservative, fabric finishes, plastics Additive in many personal care products Medical and dental sterilizing solutions, leather fixatives Permanent waving solutions Cosmetic and industrial preservative Cosmetic and industrial preservative

*CTFA (Cosmetic, Toiletry and Fragrance Association) name, appears on cosmetic labels. ( ) is trade name and may be used in other industries. †Formaldehyde-releasing preservatives. ‡Commercial allergens commonly available in the United States. **Diluted in petrolatum unless otherwise designated.

Table 29.1.3 Common contact allergens

Irritant Contact Dermatitis 699 are several reference textbooks which discuss common allergens and irritants related to specific occupations.5–7 The tests usually are applied to the upper back on aluminum disc chambers or plastic chambers held in place with non-sensitizing tape. The tests are left in place for 2 days, and then removed and read. A second reading at 3, 4, or 7 days after patch test application is important for a final interpretation. An erythematous, spreading, indurated or vesicular response at an allergen site indicates a positive allergic response. At times, irritant and allergic reactions are indistinguishable. Patch tests should be considered as a confirmatory test when an individual is exposed to a potential allergen or when the clinical presentation suggests that contact allergy is playing a role in the dermatitis. Test results require careful interpretation to discern allergic from irritant responses, as well as to determine the relevance of a positive result to the patient’s dermatitis and exposures. There are nuances to the technique, including timing of testing related to systemic corticosteroid therapy, that dictate that patch testing should be performed only by physicians experienced in the method. These tests are used only to diagnose allergy and are one component of the total evaluation. False-negative and false-positive results can occur. Patch tests often are negative in occupational contact dermatitis. Standard textbooks offer details of patch testing methods.6,7 For many substances that have not been standardized, De Groot’s book provides suggested concentrations and vehicles of patch testing.8

Contact urticaria testing Contact urticaria should be considered when an immediate itching, edematous urticarial eruption or angioedema occurs at the site of contact with a substance. Occupationally, this has been seen mostly with food handlers, particularly meat and fish processors. With increasing use of latex gloves, contact urticarial reactions to latex gloves and other medical devices has been seen more commonly in medical and related personnel and certain patient populations. This has prompted a move to non-latex alternatives, especially in hospitals.9 General guidelines for this type of testing are available in reference texts,6,7 and Hausen and Hjorth give specific suggestions for testing foods.10 Testing for contact urticaria involves applying the substance in question (e.g., saline in which a piece of a latex glove or the glove itself has been soaked, or a piece of meat or fish flesh) to normal skin, previously affected skin, or skin adjacent to active dermatitis. Signs of itching, redness, and swelling should be sought within 1⁄2 to 2 hours of application. If no reaction is seen, a small scratch or prick that draws no blood with a sterile needle is induced on normal skin, and the substance is applied to this site, followed by observation. It is important that this testing be performed with positive (histamine) and negative (saline) controls and in an environment where resuscitation equipment is readily available because lifethreatening anaphylaxis may occur. Delayed contact urticaria may occur, so patients should be followed up 1–2 days after testing.

Photopatch testing When allergy to a substance in combination with ultraviolet exposure is being considered, phototesting and photopatch testing are indicated. Phototesting involves exposure of the patient’s skin to measured doses of ultraviolet A (UVA), ultraviolet B (UVB), or both to determine whether or not a sunburn reaction occurs at less than the predicted dose of ultraviolet light. Photopatch tests are performed by applying two sets of allergens to a patient’s back. After 1 day, one set is exposed to a measured dose (usually 10 joules) of ultraviolet A, and final readings are taken at 2 and 3 days to determine whether there is a reaction to any allergen, with or without the ultraviolet exposure.11

Supplemental information and diagnostic criteria The uses and limitations of MSDS are summarized elsewhere in this text. Medical records from the workplace and private physician can help with details of clinical description of the acute events, chronology of medication use, and diagnostic work-up previously performed. Combining the history, physical findings, and results of the diagnostic evaluation can be complicated. Criteria for determining occupational causation and aggravation for contact dermatitis are summarized in Table 29.1.4. An affirmative answer to at least four criteria strengthens the case for workplace exposure substantially contributing to the dermatitis. The level of diagnostic certainty required, as with any occupational illness, depends on the circumstances, the standard for most compensation systems being ‘more probable than not’.

IRRITANT CONTACT DERMATITIS Irritant contact dermatitis (ICD), an inflammatory disease caused by skin contact with material that inflicts damage through a non-immunologic mechanism, is the most common occupational skin disease. The material that Questions* 1. Is the clinical appearance consistent with contact dermatitis? 2. Are there workplace exposures to potential cutaneous irritants or allergens? 3. Is the anatomic distribution of dermatitis consistent with cutaneous exposure during the job task? 4. Is the temporal relationship between exposure and onset consistent with contact dermatitis? 5. Are non-occupational exposures excluded as possible causes? 6. Does dermatitis improve away from work exposure to the suspected irritant or allergen? 7. Do patch or provocation tests identify a probable causal agent? *Answering yes to at least four questions may provide adequate probability for workplace exposure. From Mathias CGT. Contact dermatitis and workers’ compensation: criteria for establishing occupational causation and aggravation. J Am Acad Dermatol 1989; 20:842–8, with permission from The Association for Professionals in Infection Control and Epidemiology.

Table 29.1.4 Contact dermatitis: Mathias criteria for probable occupational causation

700 Contact Dermatitis a

Damage from irritant Threshold for dermatitis Dermatitis present Dermatitis absent Irritant exposure

Strong irritant Single exposure

Damage

Petroleum distillates Fuels Organic solvents Turpentine Chlorinated hydrocarbons Alcohols Alkalis/acids Glycols Oxidizing agents Reducing agents Esters Ketones Dimethyl sulfoxide Water (frequent or intermittent) Soaps/detergents/cleaning agents Cutting fluids Plants and animal products Table 29.1.5 Examples of industrial materials that can cause skin irritation

Time b Weak irritant Repeated exposures

Damage

Cleaner Housekeeping Construction Food service Medical/dental Engineer Hairdresser Mechanic Printer Butcher Agriculture/gardening Machinist Table 29.1.6 Occupations at high risk for irritant contact dermatitis

contacts the skin usually is a chemical that is toxic to the skin in single or in cumulative (repetitive or frequent) exposure. Examples of industrial irritants are listed in Table 29.1.5, and occupations at high risk for ICD are listed in Table 29.1.6. Lists of irritants associated with specific jobs are available from several sources.1,5 Physical factors such as mechanical forces (friction, repetitive motion) and ambient environment (heat, cold, humidity) also can lead to ICD.

Pathogenesis The pathophysiology is not well understood. There is evidence that both external environmental factors and host-related factors contribute to the likelihood that an irritant skin reaction may occur, particularly when exposure to weak irritants is being considered. Strong irritants elicit symptoms of burning and stinging as well as signs of redness and swelling in most exposed individuals within a short time period. The progression to blistering dermatitis or chemical burn depends on the concentration of the irritant and the duration of skin contact. Healing begins to occur after the irritant is removed from the skin, but during this healing phase there is a reduced threshold for developing dermatitis from the same or weaker irritants. Weak irritants tend to cause damage in a subtler manner. The stinging or burning sensation may not occur; thus, there may be no subjective signal that damage is being done. Repeated exposure to a weak

Time Figure 29.1.1: (a) Schematic for the time course of dermatitis with a single strong irritant exposure. Note rapid healing and recovery of threshold expected when exposure is stopped. (b) Time course of repeated insult with a weak irritant leading to dermatitis. The lowered threshold for dermatitis that results leads to perpetuation of dermatitis, even with less frequent damaging individual exposures. (Adapted from Malten KE. Thoughts on irritant contact dermatitis. Contact Derm 1981; 7:238–47; and from Dahl MV. Chronic irritant dermatitis: mechanisms, variables, and differentiation from other forms of contact dermatitis. Adv Dermatol 1988; 3:261–76. © 1988 Munksgaard International Publishers Ltd, Copenhagen, Denmark.)

irritant may be necessary to produce clinical evidence of dermatitis, so-called cumulative insult damage. The latency period for development of ICD of this type may vary from days, if contact with the weak irritant is constant, to years if contact is intermittent. Further complicating the picture is the fact that additional weak irritants, even water and soap, add further insult and contribute to the prolonged healing times necessary for dermatitis of this type. The schematic diagrams shown in Figure 29.1.1 contrast the time course of skin damage and dermatitis from strong versus weak irritants, and the effect on the threshold for developing dermatitis.12,13 Environmental or exogenous factors that may predispose a person to the development of cutaneous irritation

Irritant Contact Dermatitis 701 from a given material include the type of irritant and the amount that penetrates the skin (which may depend on the concentration and vehicle for the irritant and time of contact with skin). The body site where contact occurs also makes a difference: eyelid skin is more prone to air-borne irritants than is palmar skin. Percutaneous penetration also varies with anatomic site; for example, the dorsal surfaces of the hands may be more likely to develop dermatitis than the palms due to differences in stratum corneum penetration by the irritant. Temperature and humidity also may be important. Working in a hot or cold, dry environment may cause chapping and water loss from the skin that could predispose the skin to penetration by some irritants, whereas heat, sweat, and occlusive clothing also could contribute to irritant effects of a material on skin. Perhaps the single most important environmental factor in ICD is the role of water or wet work.14,15 Hydration of the skin increases the potential for the penetration of many substances. Intermittent wet and dry exposure leads to water loss, which can generate breaks in the skin barrier that solvents can traverse. The addition of soap or detergent to the water exposure, either in the hygiene routine or the industrial process itself, can lead to loss of superficial skin lipids, alkaline damage to the skin, or abrasive action on the stratum corneum that also contributes to irritant skin damage.16 Water exposure itself, or in combination with soap, can act as a weak irritant that lowers the threshold for dermatitis from other substances that might be otherwise tolerated, as shown in Figure 29.1.1(b).17,18 For this reason, avoidance of wet work is often recommended during the healing phase of contact dermatitis of any cause, and it also is recommended for individuals with endogenous factors (see later) that might make them more likely to develop ICD. Atopic dermatitis has a strong association with the development of occupational ICD, as noted earlier. Individuals with atopic dermatitis have a low threshold for irritation and require a long time to heal. Individuals with a positive atopic history but no apparent dermatitis also may have a higher risk of hand dermatitis.17,19 Separating the endogenous atopic dermatitis from the relative importance of external irritants may be difficult. A history of new dermatitis at sites not previously involved but compatible with jobrelated exposure or a history of significantly more severe involvement (aggravation) at sites of pre-existent dermatitis suggests that the exogenous exposures are contributing to an ICD flare of atopic dermatitis. The link between atopy and potential for ICD is a powerful case for preplacement evaluation and job modifications, because preventative measures could have a significant impact on reducing the number of cases of occupational skin disease. Other host factors for ICD include skin care regimen and skin types. The point of injury by irritants at the cellular level is variable. The stratum corneum barrier may be compromised or cell membranes may be disrupted, particularly by detergents. Damage to the epidermis can trigger the release of various cytokines and inflammatory mediators that potentiate the damage and inflammation. Some irritants have direct effects on blood vessels, causing vasodilatation,

and others may directly stimulate polymorphonuclear leukocyte chemotaxis or mast cell degranulation. The degree of inflammation seen clinically or histologically depends on the specific irritant involved, its concentration, and the type of exposure.20

Clinical course The clinical manifestations of ICD usually reflect the epidermal damage: a glazed look to the skin, with dryness of the surface ranging from a chapped scaling appearance to cracking or fissuring (Fig. 29.1.2). Erythema is variable and may be very faint when it is caused by weak irritants. Burning and stinging may be more prominent than itching. The web spaces of the fingers may be affected first, especially in ICD associated with wet work. In general, the presence of multiple small vesicles, especially around the nail fold, is not common in ICD and its presence should suggest the possibility of concomitant contact allergy to

A

B Figure 29.1.2: Contact irritant dermatitis. Irritant contact dermatitis usually begins in the web spaces of the fingers and than spreads to involve the adjacent skin. The dorsum of the hand may be involved more than the palm. Redness, scaling, and itching without obvious vesiculation are most characteristic. This sort of dermatitis is common in people who do wet work, and is especially common in people who have a history of childhood eczema and who also do wet work.

702 Contact Dermatitis the same or another offending agent.21 Large blisters of acute onset may occur with a strong irritant, causing the equivalent of a chemical burn, but these blisters usually resolve rapidly with removal of the irritant and wound care. With epidermal disruption, secondary infection may occur and lead to amber-colored crusting of the involved sites. With cumulative irritant dermatitis or chronic irritant exposure, the skin becomes hyperkeratotic. At sites like the hands, this thickened skin is prone to fissures.

Diagnosis The major considerations in the differential diagnosis of ICD are atopic dermatitis (Fig. 29.1.3) and ACD, and it may be extremely difficult to distinguish among these conditions clinically. Table 29.1.7 contrasts features of irritant and ACD. In chronic cases, other disorders such as psoria-

Figure 29.1.3: Atopic hand eczema. This hairdresser is not allergic to anything in her workplace on patch testing. The frequent wetting and drying of her hands in her work has given her chronic dermatitis on both her palms and her dorsal hands. Notice the distal distribution, which is characteristic of irritant dermatitis caused by wet work exposures in atopics.

Variables Who Where When Job-relatedness Atopy Morphology Distribution Intermittent and frequent water exposure Immunology Histology Chronicity Diagnosis Examples

sis and acquired keratoderma also should be considered. There is no routine laboratory test to confirm a diagnosis of ICD: skin biopsy results are non-specific, and patch tests are used to confirm or exclude contact allergy. The history of exposures, job description, and clinical presentation lead to a clinical diagnosis of ICD.

Treatment and prognosis Avoidance of strong irritants usually leads to prompt improvement of an acute case of ICD. Oral antibiotic treatment of secondary infection, soaks and debridement of crusts or bullae, and topical emollients often are adequate in this setting. If a chemical burn has occurred, specific emergency treatment should be instituted depending on the nature of the irritant chemical involved. For dermatitis due to weaker irritants or cumulative or multiple exposures, it is helpful to move the patient away from the job task associated with those or other irritants. As mentioned earlier, the threshold for further irritation is lowered, so that ideally all irritant exposures (occupational, domestic and hobby, and therapeutic) should be minimized. In the workplace, reducing exposure to the irritant through job modification or appropriate protective clothing may be helpful. If a similar problem has been seen among a number of workers at a given job, consideration should be given to replacement of the offending agent with a less irritating material. The proper glove may be helpful but barrier creams generally are not beneficial.22,23 Skin cleaning should be performed as gently as possible, and abrasives, solvents, or harsh detergents should be avoided. If there is a marked inflammatory reaction, a topical corticosteroid in a simple ointment base (e.g., petrolatum) may be of benefit. Corticosteroid creams contain preservatives that can sensitize the skin. Ointments do not contain these additives or water in the base, which can be drying. Lack of response or worsening of symptoms with use of a topical corticosteroid may rarely indicate contact allergy to the steroid moiety itself.24 Bland emollients such

Irritant

Allergic

Many people Localized Soon with strong irritants (minutes to hours) Late with weak irritants (months) Improves with long (3 weeks) vacations Predisposes Erythema, scale, fissures Hands 90% May cause and perpetuate

Few people Spreads 24–72 hours

Not known to be relevant (1991)

Type IV (cell-mediated immunity, delayed hypersensitivity) Spongiosis with lymphocytic cellular infiltrate or inseparable from irritant If diagnosed early clears, can be chronic History, physical, and patch testing Poison oak/ivy, nickel, chrome

Spongiosis or upper epidermal damage, some lymphocytes Often History and physical observation Hairdressers and machinists chronic hand dermatitis

Table 29.1.7 Generalizations about irritant versus allergic contact dermatitis

May improve even on weekends No predisposition Vesicles or inseparable from irritant Hands 90% May perpetuate

Irritant Contact Dermatitis 703 as white petrolatum are preferable to popular commercial moisturizing lotions, which can occasionally lead to further irritation.25 Systemic corticosteroids may alleviate symptoms of inflamed ICD, but they should be reserved for short-term therapy of more severe cases. Oral antibiotics to cover Staphylococcus aureus and group A β-hemolytic streptococcal infection, in particular, should be used if crusted fissures are evident. As the dermatitis improves, the strength and dosage interval of the topical corticosteroid should be tapered. The time course to improvement can be variable. As a general rule, the longer the duration of dermatitis, the longer the recovery time. The presence of atopy also prolongs healing time. The recommended length of time away from exposures or with modified duties is variable. Photographs of the dermatitis at the beginning and end of the time-off period may be of great value. If the dermatitis does not improve with the avoidance and treatment outlined, then other diagnoses should be considered and the therapeutic regimen should be scrutinized. The prognosis of ICD depends on exogenous factors, such as the ability of the person to avoid irritants, and endogenous factors, such as the presence of atopy. For an isolated episode of an acute case of ICD to a strong irritant, the prognosis for recovery is good. Chronic ICD may not have such an optimistic prognosis. In fact, occupational contact dermatitis in general carries a poor prognosis, which is frustrating to the patient, employer, physicians, and insurance carriers.26 For example, a follow-up study of patients with soluble-oil dermatitis (metal-working industry) indicated that the patients who experienced the resolution of their dermatitis after discontinuing oil contact did so within 3 months.27 If there was no improvement within that time, the patients were not likely to improve whether they avoided soluble-oil contact or not. Overall, approximately 70% of these patients had not healed 2 years after being diagnosed.27 Strides made in industrial hygiene and dermatologic diagnosis and treatment in the last 20 years have not led to an improvement of the prognosis. In long-term followup investigations of patients diagnosed with occupational contact dermatitis, 25% of patients have persistent dermatitis, 50% of patients had periodic flares of dermatitis, and only 25% completely healed. These percentages are similar for patients who change or leave their jobs owing to their skin disease; the majority continues to have dermatitis.28 Factors to be considered in persistent occupational dermatitis are summarized in Table 29.1.8. Perhaps the most difficult of these factors to determine is the development of secondary dermatoses due to a lowered irritant threshold or other factors, or the concept that dermatitis begets dermatitis. It may not be possible to predict when and how much improvement will occur for a given patient with contact dermatitis.

Incorrect dermatologic diagnosis Proper diagnosis of occupational contact dermatitis, but Failure to establish specific cause Failure to eliminate exposure to causative agent(s) Improper therapeutic intervention Development of secondary dermatoses Infection Neurodermatitis/Lichen simplex chronicus Reactivation of atopic dermatitis Improper job modification or change Improper cleansing routines Development of Cross-sensitivities Multiple sensitivities New sensitivities Malingering From Birmingham DJ. Prolonged and recurrent occupational dermatitis: some whys and wherefores. In Adams RM ed. Occup Med State Rev 1986; 1:349–56; and Hogan DJ, Dannaker CJ, Maibach HI. The prognosis of contact dermatitis. J Am Acad Dermatol 1990; 23:300–7, with permission from American Academy of Dermatology.

Table 29.1.8 Some reasons for poor outcome of occupational contact dermatitis

Individuals Vocational counseling/pre-placement screening Atopic dermatitis Family history of atopy History of contact dermatitis Leisure activities/hobby exposures Identification of irritants Literature sources Material safety data sheets Predictive tests of skin irritancy and allergenicity Draize test (animals) Cumulative irritation test (human/animals) Work environment General Ventilation Temperature Humidity General housekeeping Airflow patterns Industrial process Reduce skin contact (automation, enclosure) Substitute less irritating chemicals Appropriate protective clothing/gloves Label hazardous materials/work sites Reduce mechanical irritation Education Workers and management Exposures Job procedures/technique training Emergency measures Information on patient’s job exposures Healthcare providers Recognition and treatment of skin disease Skin care program Choice of soap, detergent, cleanser Soft towels Moisturizing creams Assess hand washing routine (avoid excess hand washing) Protective clothing Simple lubricants (e.g., petrolatum)

Prevention

Table 29.1.9 Prevention of contact dermatitis

Given the poor prognosis of ICD, improved preventive strategies are essential.29,30 Prophylactic strategies should

include those outlined in Table 29.1.9. Identification of workplace irritants, modification of work practices and

704 Contact Dermatitis industrial processes, and worker education can reduce irritant exposures and incidence and severity of ICD. Early recognition and appropriate treatment are also essential. Workers with atopic dermatitis should be counseled as to the risks of wet-work careers such as housekeeping, medical fields, and hairdressing. Intermittent wetting and drying is particularly damaging (e.g., hand washing). Preplacement screening for atopy for wet-work or other high-risk occupations might reduce risk but is of unproven value and is ethically and legally problematic. Educational programs should also address possible irritant exposures outside the workplace, as in domestic chores or hobbies, because an accumulation of irritant insults may lead to dermatitis. In the context of the job description, it is important to know which chemicals or other materials are irritating to human skin. Strong irritants may be identified on material safety data sheets, but weak irritants or those that may result in dermatitis from cumulative exposure may not be identified as such. Other sources of information on potential irritancy should be sought in the industrial literature. For new products likely to have significant exposure to human skin, predictive testing of irritant potential can be performed. There are a number of methods advocated for use in experimental animals or humans. These methods are widely used to evaluate skin care products and topical medications, but they could be adapted for evaluation of industrial compounds, if necessary. Further preventive strategies are similar to those discussed later under ACD and are summarized in Table 29.1.9.

LOW HUMIDITY DERMATOSES

of atopic dermatitis are particularly vulnerable, and those who already have some sort of dermatitis will note that it worsens. Persons in whom this problem is likely to be prevalent include any workers in low humidity environments, which are common with current heating and airconditioning systems.

Diagnosis History and suspicion are most helpful initially, but inspection of the workplace, with documentation of low humidity and high temperature and airflow, as well as correction of the skin complaints after remediation of these factors, is diagnostic. Air-borne gaseous and particulate substances and other causes of dermatitis (allergic, contact, seborrheic) may complicate the diagnosis, and they should be sought as well. Low humidity dermatitis may be the actual explanation for some skin diseases that have been attributed to visual display terminals, carbonless paper, and even the sick building syndrome (see Chapters 38 and 23.2), and thus its diagnosis and correction can improve the morale of an entire workforce.33

Treatment and prognosis Raising the humidity and lowering the temperature, when possible, solves most complaints. Moving a worker away from under or beside an air duct helps even more. Simple lubricants, devoid of common sensitizers that can complicate the problem, hasten the recovery of the xerotic skin. In some instances, topical corticosteroids are required briefly.

Pathophysiology At humidities of 50% and higher, the stratum corneum is at a steady state and remains pliable. In situations that are hot and dry and in which there is flowing air, the stratum corneum dries out and fissures, resulting in chapped skin.31 This mechanism plays a role in the development of some cases of allergic contact dermatitis and most irritant dermatitis associated with intermittent wetting and drying.32 It plays a primary role in explaining a dermatitis that is common in hot, dry, well-ventilated workplaces. As with most types of irritant dermatitis, this problem can affect many workers (Table 29.1.7).

Clinical course Patients working in these environments, and especially those sitting exactly at points of maximum airflow, note itching on their exposed skin surfaces (especially the face and neck). This problem is followed by scaling and, finally, by the development of patches of asteatotic eczema (xerosis, erythema craquele, chapping). Although exposed surfaces are involved most often, covered skin may be included as well. Erythema or flushing and even urticaria (perhaps as a result of rubbing and scratching) also may be part of the syndrome. Individuals with a personal history

IRRITANT CONTACT DERMATITIS DUE TO FIBERGLASS Pathophysiology Fiberglass dermatitis is a specific type of ICD in which the mechanical effect of the sharp fiber spicules causes a sensation of itching and, at times, an inflammatory reaction.34 The fiber size (>4.5 μm) is important in determining the likelihood of skin irritation. Textile glass and wool fibers that are used for insulation are the types most often implicated in skin irritation.

Clinical course The skin lesions are small papules that are often crusted due to scratching. Hive-like lesions and linear excoriations may occur, and lesions have a predilection for exposed surfaces as well as areas of the skin where clothing is close to the skin (e.g., the waistline). Intense pruritus may be the chief complaint, and clinically apparent lesions may be absent. Fibers may be found on tape strippings of the skin or skin scrapings examined with potassium hydroxide preparation.

Frictional Contact Dermatitis 705

Diagnosis The differential diagnosis includes infestations by scabies or body lice, irritation from dust exposure, neurodermatitis, and other types of contact dermatitis. A history of exposure such as working with fiberglass insulation or in indoor environments undergoing renovation should be sought.

Treatment and prognosis In fiberglass workers, the symptoms of irritation often are relieved with gentle washing of exposed sites with nonabrasive soap and water. New workers tend to be more affected, and work hardening can occur over time in some workers. Topical corticosteroids are of limited benefit.

Prevention To prevent fiberglass dermatitis, it is helpful for workers to wear loose clothing with long sleeves and pants, avoid rubbing the skin, and wash work clothes separately. Careful skin cleaning after working with fiberglass and use of skin emollients also may help.

POST-TRAUMATIC ECZEMA Pathophysiology and clinical course Post-traumatic eczema is a term used for the occurrence of dermatitis at a site of previous injury to the skin. The pathophysiology is not well understood and may be related to the lowered threshold for irritant dermatitis at sites of skin damage. Mathias has described an idiopathic type and an isomorphic reaction (i.e., the occurrence of a cutaneous lesion of a recognizable skin disease at a trauma site).35 Patients with typical lesions of psoriasis, vitiligo, and lichen planus at other sites may develop new lesions at sites of trauma through the isomorphic reaction. Idiopathic post-traumatic eczema is an inflammatory, scaling reaction limited to the previous trauma site. The eruption usually begins within weeks of the injury and may persist for months to years after the initial injury.

Diagnosis It is essential to exclude infection (bacterial, herpetic, fungal) at the injury site through appropriate laboratory techniques and culture. Contact dermatitis resulting from medications used to treat the wound (e.g., neomycin, adhesive, benzocaine) also should be excluded.

Treatment and prognosis Topical corticosteroids may be helpful in the treatment of post-traumatic eczema, but patients may have intermittent active dermatitis at the trauma site. If the initial injury was occupational, this prolonged dermatitis at the injury site may be considered work related.

FRICTIONAL CONTACT DERMATITIS Pathophysiology and clinical course Frictional dermatitis results at points of chronic irritation and trauma. It is most easily recognized as calluses, and some occupations have distinctively distributed calluses.36 Skin diseases that koebnerize (develop lesions typically at trauma sites [isomorphic phenomenon]) such as psoriasis or lichen planus may involve the palms and present a clinical picture that, at first glance, looks like frictional contact dermatitis and even may have foci of spongiosis histologically. Some people develop similar lesions clinically and histologically, although they do not have psoriasis or a family history. Clinicians may call frictional contact dermatitis psoriasiform dermatitis or hyperkeratotic dermatitis of the palms.37 This entity does not vesiculate, seldom itches, and may involve the weight-bearing portions of the feet if heavy shoes are worn and if the patient is obese. The condition improves dramatically when the mechanical trauma is withdrawn, but it may take a month or longer to do so. Workers with impressive mechanical hand trauma are at greatest risk (e.g., construction workers, machinists, forest industry workers, and baggage handlers). A more subtle but equally distinctive frictional dermatitis can occur in office workers who handle large volumes of papers, and especially carbonless copy paper. This dermatitis is found on the fingertips, where maximum paper contact occurs. It does not vesiculate or itch but rather is characterized by areas of redness, scaling, and painful fissuring. These patients may have no previous history of hand dermatitis and often are not atopic.33

Diagnosis This entity receives little mention in most comprehensive reviews of hand dermatitis. Exclusion of other entities such as irritant and allergic dermatitis, genuine psoriasis, and dermatophyte infection is important. A biopsy can be of great help but will not absolutely exclude genuine psoriasis, which can be similar to frictional dermatitis histologically. Psoriasis may first begin on the palms in the setting of occupational trauma and only with the spread of the disease to distant sites can it be differentiated with confidence from the more nebulous frictional dermatoses. Clearing of the dermatitis with trauma avoidance is of the greatest diagnostic help.

Treatment and prognosis Trauma avoidance, simple lubricants, and phototherapy are helpful.38 Topical corticosteroids are of little help. Chronicity occurs, but clearing with a job change is more common.

706 Contact Dermatitis Industry Textile Painting Printing Agriculture Food preparation Medical/dental/veterinary Hairdressers Metal workers Rubber manufacture Leather tanning Plastics manufacture

Possible sensitizing agents Azo dyes, formaldehyde, resins Epoxy, acrylates Acrylates, epoxy resins Pesticides, plants, rubber Flavorings and preservatives Rubber, acrylates, medications Paraphenylenediamine, nickel, glyceryl monothioglycolate, rubber Biocides in cutting fluids, metals Thiuram, carbamates, mercaptobenzothiazole Chromate, formaldehyde Formaldehyde, phenolic resins, epoxy resins

contact with the allergenic substance (Table 29.1.7). The eruption is worse at sites with the most concentrated allergen exposure, and it may be milder (scattered erythematous papules) on areas where there was minimal exposure or where the allergen was washed off. The distribution of lesions may offer clues to the source of the allergen (e.g., gloves, clothing, air-borne) (Fig. 29.1.4). In severe, acute cases, the lesions may extend beyond sites of direct exposure because of the intensity of the inflammatory reaction and a phenomenon known as an auto-sensitization, or id, reaction. Because id reactions on the hands also are associated with fungal infections, the feet should be carefully examined. Inspecting the feet may provide clues to ACD because exposure may be originating from the person’s

Table 29.1.10 Examples of occupations at risk for contact allergy

ALLERGIC CONTACT DERMATITIS ACD is caused by skin contact with material that triggers an immunologic reaction, leading to inflammatory skin lesions. Approximately 20–25% of cases of occupational contact dermatitis are found to be allergic in origin, although some studies have reported up to 50%.39 Certain occupational groups are at greater risk for developing contact allergy due to occupational exposure to common allergens (Table 29.1.10), with agricultural and manufacturing jobs being most often cited. In California and Oregon, for example, poison oak dermatitis is among the most commonly reported job-related illnesses. Common allergens in occupational ACD are summarized in Table 29.1.3, but these allergens represent only a small portion of known industrial allergens.

A

Pathophysiology ACD is a type of delayed hypersensitivity reaction. A hapten for contact hypersensitivity usually is a low-molecular-weight, lipid-soluble material (to allow penetration through the stratum corneum to viable epidermis). The hapten then binds or complexes with cell surface or structural proteins on various cells, including Langerhans’ cells (epidermal antigen-presenting cells) and keratinocytes. It appears that the Langerhans’ cell processes and presents the antigen to T lymphocytes, and a cascade of events follows involving a number of interleukins and chemotactic factors that leads to a clonal proliferation of sensitized lymphocytes and to the inflammatory reaction seen clinically. There are several recent reviews of the mechanisms of ACD, and our understanding of the intricacies of the processes is still evolving.40–42 Recent data support the role of cutaneous irritancy as a necessary factor in the development of allergic contact dermatitis.43,44

Clinical course The clinical appearance of acute ACD usually is an erythematous, vesicular, edematous eruption at the sites of

B Figure 29.1.4: (a) and (b) Allergic contact dermatitis. Allergic contact dermatitis is characterized in its acute phase by vesicles with crusting and scaling. Erythema and edema are also features of this condition. The patient in (a) used neomycin-containing ointment on irritant dermatitis. The patient in (b) was a hairdresser who became allergic to the rubber gloves she was using to protect her hands from wet work.

Allergic Contact Dermatitis 707 footwear (e.g., rubber chemical). The lesions in acute ACD often occur 1–2 days after exposure, and may take 2 weeks to resolve. Thus, this time course differs from acute irritant exposure. With repeated exposure to an allergen chronic ACD can develop. The clinical picture is variable, depending on the time interval between exposures (daily, weekly), and whether concomitant irritation is occurring with the same or other materials. In chronic ACD, the lesions generally are erythematous, scaling, and thickened. Vesicles may or may not be present, but they should be looked for very carefully. The hands are the most frequent site of involvement in ACD as well as in ICD. The dorsal surfaces may be more involved than the palms. However, the pattern of lesions on the hands cannot be reliably used to clinically distinguish ACD from ICD.45

Diagnosis The diagnosis of ACD is based on the occupational history and results of skin patch testing. A clinical impression alone can be incorrect, particularly with regard to the offending agent. Patch testing is summarized above, and as noted, there are several limitations to patch testing. There are limited commercially available allergens in the United States (Table 29.1.3), and both false-negative and falsepositive results can occur. Adams’ Occupational skin disease5 and other reviews39,46 offer valuable information about patch testing. If patch testing gives a positive result, a determination should be made as to whether or not the allergen is present in the work environment. If so, avoidance of this allergen by the patient should be attempted by wearing adequate protective clothing, changing the industrial process to reduce worker exposure, or substituting a less allergenic material. If these measures are not feasible, then the worker may need to be removed from the job. If patch testing is negative, the person’s clinical history, examination, and exposures should be reviewed. If ACD is still suspected, one must question whether the proper allergens were tested and, if so, were they tested with the proper concentration and vehicle. A visit to the workplace to observe the patient performing his or her job may result in increasing the pool of suspected allergens. After an interval of several weeks, repeat patch testing might be performed with the same and additional allergens. It is preferable to have the patient tested when no systemic corticosteroids are being taken. Ultraviolet light can alter immune responsive cells in the skin; therefore it is best to avoid testing in the setting of recent ultraviolet exposure at the test site.47 If contact allergy is suspected in such an individual (e.g., a brick mason who works shirtless in the summer), it may be worthwhile to repeat patch tests in the winter. Other possibilities to be considered when faced with unexpected negative patch tests are contact urticaria or photocontact dermatitis. Eczematous dermatitis, for example, may result from scratching the area affected by contact urticaria induced by rubber gloves.

Treatment and prognosis Avoidance of the allergen by one of the methods noted earlier is the first step in the management of ACD.48 Unfortunately, these measures alone often do not solve the problem for several reasons. (1) The allergen (e.g., nickel) may be commonplace in both the workplace and the domestic environment, thereby making avoidance extremely difficult. (2) The allergen (e.g., glyceryl monothioglycolate in hairdressing) may be very common in the work environment, not only in direct contact in an industrial step but as residual in other steps of a process. In the example of glyceryl monothioglycolate, even if the affected hairdresser stops using the offending permanent wave solutions on his or her clients, there is still potential for the hairdresser to react to the product on other clients. Further, routine use of gloves may not protect the worker because the allergen can penetrate the gloves.49 (3) In addition, ICD, atopic dermatitis, psoriasis, and another dermatosis may be part of a given patient’s diagnosis, so that avoidance of the allergen alone does not solve the problem. When there is a reduced threshold for irritation, a recovery phase also may occur with ACD. Thus, general protective measures outlined for ICD are useful in treating ACD as well. For acute ACD, removal of the affected individual from work and irritant exposures is indicated. Cool water or astringent agents such as aluminum acetate (Burow’s solution) used as compresses are helpful in the treatment of weeping vesicular dermatitis. Topical corticosteroids applied once or twice daily also are useful. The strength of the corticosteroid chosen should depend on the body site involved and the severity of the reaction. Potent topical corticosteroids generally should not be used on the face or intertriginous sites, such as the groin, for prolonged periods (> 1 week). If medical treatment is sought early in the presentation of severe, acute ACD, systemic corticosteroids may be useful. A 2–3-week course of prednisone (0.5–0.75 mg/kg/day, with optional tapering) often is helpful. Shorter courses, or dose-packs, may result in rebound of signs and symptoms. Patch testing should be deferred until the systemic corticosteroids dose is lowered or completed. With vesicular dermatitis, secondary infection should be suspected and treated systemically, if present. It is prudent to avoid potentially complicating topical therapy of ACD with treatments such as neomycin-containing antibiotic ointments, medicated petroleum jelly, or even rubber gloves as such treatment could become part of the allergic problem if the patient also develops an allergy to one of these materials. Multiple contact allergies can develop. Systemic antihistamines can be useful for symptoms of ACD. Because most antihistamines are potentially sedating, caution should be used in prescribing them to workers who remain on the job. Newer low-sedating antihistamines may be potential alternatives that may not interfere with safety or job performance. For chronic ACD, treatment measures are similar. Phototherapy with psoralens plus ultraviolet A (PUVA) may also be beneficial.50

708 Contact Dermatitis

Prevention Many of the preventive measures discussed under ICD are applicable to ACD as well (Table 29.1.9). With regard to preplacement evaluation, eliciting a history of contact allergy to nickel, gloves, or other allergens relevant to a specific job situation can be useful. For example, an individual may have a past history of nickel allergy that had manifested as a jewelry reaction. The associated rash may have long since resolved with avoidance of the offending jewelry. That individual could develop a new eruption if the new job involved frequent handling of nickel-plated metal parts. Educating the potential worker about the nature of the job and potential allergen exposures (e.g., poison ivy or oak in an outdoor job) also is useful in that this information may help an individual recall a past problem with that allergen. To educate the person effectively, however, the potential allergens in the workplace need to be identified. There are methods of predictive testing available to determine the sensitizing potential of chemicals in contact with human skin. If a situation arises in which a potentially allergenic substance is to be introduced into the workplace, the question may arise regarding testing of employees to determine whether an individual is already allergic to that substance.51 This generally is not recommended for several reasons. An allergy may develop later even if the test is currently negative, and screening patch tests will not predict this possibility. Also, there is a risk of sensitizing the individual to the substance through the 2-day occlusive patch test, which may be a different exposure than might be encountered in the routine job activity. Predictive patch testing generally is limited to pre-marketing product evaluation and involves special techniques. Protective clothing can be useful in minimizing skin contact with an allergen if (1) the clothing (such as gloves) is properly worn and (2) the clothing has been chosen so that it prevents penetration of the offending agent.52 Choosing specific gloves so that a hairdresser avoids contact with glyceryl monothioglycolate is a good example. Reference texts are available that detail penetration times for various gloves and chemicals.53 In general, barrier creams have limited usefulness, but more recently, creams have been developed that aim at specific allergens. Products applied by outdoor workers prior to exposure to poison oak and ivy have been found to be effective in preventing or decreasing the occurrence of poison oak and ivy allergic dermatitis.54,55 Other barrier creams have been developed in an attempt to inactivate an allergen (e.g., chemical reduction of chromium) or to chelate metal allergens.56

CONTACT URTICARIA Immediate urticarial reactions to topical materials in the workplace, or so-called contact urticaria, has been recognized for many years, particularly in workers in the fish and meat industries. There is heightened awareness of contact urticaria in recent years due to the increased

recognition of immediate reactions to latex gloves and products in the healthcare field.9,57

Pathogenesis The wheal and flare reaction of contact urticaria can occur through either allergic (immunologic) or non-allergic (non-immunologic) mechanisms. Allergic contact urticaria usually is mediated by an IgE mechanism, and it is more common in atopic individuals (e.g., latex contact urticaria). As with delayed contact allergy, there may be a history of frequent exposure to an agent over time before symptoms occur, and a low proportion of individuals exposed to the agent develop an allergic reaction. In the non-immunologic type, many exposed individuals may be affected by the offending agent despite no prior exposure, paralleling what can occur with strong irritants. Atopic individuals are not predisposed to non-immunologic contact urticaria. There appears to be a direct effect on blood vessel walls through the release of vasoactive substances, leading to the wheal and flare.

Clinical course The mechanism of contact urticaria (allergic versus nonallergic) may not be distinguished by the clinical presentation. There may be the onset of itching, stinging, or burning immediately or within minutes of contact with the agent. Preservatives (e.g., sorbic acid) or perfume ingredients (e.g., aldehydes) may explain the burning some people experience with cosmetics. This urticarial-like reaction is non-immunologic. The clinical lesions of contact urticaria may range from erythema to urticaria or angioedema, or all three (Fig. 29.1.5). Lesions usually are limited to sites of exposure, and again, the hands and forearms are the sites affected most often. Examples of materials that can cause contact urticaria in the occupational setting are listed in Table 29.1.11. Fish, foods, animal proteins, flavorings, and perfume ingredients are among the agents associated most often with contact urticaria. With the institution of universal precautions in all allied health fields, contact urticaria is important to recognize because life-threatening anaphylactic reactions have occurred in healthcare workers (as well as in patients) exposed to latex products. Contact urticaria to latex (also called latex hypersensitivity) may present as asthma, hand dermatitis, or anaphylaxis depending on the source of exposure, e.g. air-borne latex protein in glove powder, gloves themselves, or mucosal contact with latex products. Groups at greatest risk include healthcare workers, patients with meningomyelocele, and those with atopic dermatitis.9,57 A curious aspect of contact urticaria is that symptoms and signs may not occur with every exposure. In some cases, the reaction occurs only when the material is in contact with previously damaged skin (e.g., irritant dermatitis, atopic eczema), and if contact occurs when the underlying dermatitis is in remission, no urticaria or pruritus may be evident. Further, the clinical lesions of contact urticaria (erythematous wheals) may evolve into a vesicular

Contact Urticaria 709 Foods Apple Beer Carrot Egg Fish Lettuce Meat (poultry, beef) Milk Potato Spices

Medications

Flavorings/preservatives*

Bacitracin Benzocaine Chloramphenicol Neomycin Penicillin Promethazine Streptomycin

Balsam of Peru Benzoic acid Cinnamic acid/aldehyde Sodium benzoate Sorbic acid

Miscellaneous Acrylic monomer Alcohol Aliphatic polyamine Diethyltoluamide Formaldehyde Lindane Latex Metals: nickel Platinum Xylene

* Usually cause non-immunologic contact urticaria.

Table 29.1.11 Sources of contact urticaria in the workplace

A

C

dermatitis resembling typical ACD. For this reason, the time course of symptoms – itching at the time of or soon after contact – is a key point in suspecting the diagnosis of contact urticaria as the triggering factor for the subsequent evolution of the condition into eczematous dermatitis.

Diagnosis and differential diagnosis Idiopathic urticaria usually can be distinguished from contact urticaria by its distribution. Physical urticarias, such as pressure or cold urticaria, may be more difficult to distinguish without contact urticaria testing or other test-

B

Figure 29.1.5: (a)–(c) Contact urticaria to latex. Wearing a latex glove, especially a wet one, for as short a time as 10 minutes can produce urticaria on the skin underneath the glove (a). Sometimes the urticaria is more conspicuous on the wrist than it is on the hand (b). An alternative way of producing contact urticaria is by tying a piece of the suspect glove onto the normal skin for 10–30 minutes and then observing the area of skin where the glove fabric has had contact (c).

ing for physical urticarias. Dermographism, the most common physical urticaria, can be mistaken for type I allergy.58 For example, a worker who cleans and packs fish on ice is exposed to the fish, the cold environment, rubber gloves, and the mechanical pressure of grasping the knife, any one of which could trigger urticarial lesions. Testing to demonstrate that the fish caused urticarial lesions could lead to intervention with protective vinyl gloves that might allow the worker to continue on the job. Irritants can cause immediate symptoms, usually burning more than itching, but there can be an overlap of irritant, allergic, and urticarial reactions triggered by certain agents

710 Contact Dermatitis (e.g., formaldehyde, acrylates, ammonium persulfate in hair bleaches). Testing for immediate reactions to latex proteins begins with a RAST (radio-allergosorbent test) for latex. If this is negative, having the patient wear a latex glove (or portion of glove) on a moistened hand is the next step. The risk of anaphylaxis is lower with this approach than with the skin prick test, still considered the gold standard.57 ACD can be difficult to distinguish in the case of an eczematous reaction developing after immediate symptoms. Contact urticaria testing with a delayed eczematous reaction at the test site occasionally can demonstrate the problem, or performing both contact urticaria testing and standard patch testing at times may be useful. Latex protein can cause a type 4 delayed-type response as well as a type 1 reaction.59,60

Treatment and prognosis As with other types of contact dermatitis, avoidance of contact of the allergen with the skin is important. Oral antihistamines are useful to relieve itching. Systemic corticosteroids or epinephrine should be held in reserve for severe cases. Topical corticosteroids may be useful when eczematous lesions evolve. Protective gloves or sleeves can be useful if they are chosen carefully. As noted, the gloves themselves may be causing the problem. If glove-induced contact urticaria is suspected, it is most often due to the latex component, but immediate reactions to other components such as cornstarch or rubber additives have been reported. It may be important to analyze the situation before trying to recommend a substitute glove.

Prevention Recognition that there are materials in the workplace that could cause both non-immunologic and immunologic contact urticaria is the first step. Worker education and personal protection are important. Because atopic individuals are more likely to develop immunologic contact urticaria, medical surveillance for early diagnosis and exposure modification may be helpful.

CONTACT PHOTODERMATITIS Pathogenesis Some naturally occurring and synthetic chemicals may cause contact dermatitis only in the presence of light, which may be sunlight or artificial light sources that emit specific wavelengths. This general reaction is known as contact photodermatitis, and there are two categories of reaction: phototoxic and photoallergic. Phototoxic reactions have non-immunologic mechanisms, in which a chemical, by the nature of its structure, absorbs a specific wavelength of light and produces a product that causes cellular damage. The mechanism may or may not involve production of reactive oxygen species. Photoallergic reactions involve conversion, by exposure to light, of a chem-

ical into an immunologically reactive hapten that leads to a delayed allergic hypersensitivity reaction. A recent review covers the spectrum of photosensitivity disorders.61

Clinical course It can be difficult to distinguish clinically phototoxic from photo-ACD. In both, exposed areas of the skin such as the face, anterior V of the neck, backs of the hands, and uncovered sites on the arm (to the shirt sleeves) and legs frequently are involved. Hairy areas, upper eyelids, and below the chin may be spared. Phototoxic reactions may appear as a painful, exaggerated sunburn that may develop bullae and subsequent pigmentation. When the offending agent is avoided, the dermatitis usually disappears promptly. Photo-ACD may have many of the features outlined under ACD (itching, vesiculation).

Diagnosis and differential diagnosis The distribution (on sun-exposed surfaces) of the reaction is an important clinical feature in suspecting the diagnosis, and the condition sometimes is difficult to distinguish from air-borne contact exposure. The method for photopatch testing is outlined in a previous section. Examples of substances that can cause contact photodermatitis with topical exposure and occupations at risk are given in Table 29.1.12.

Topical agent Phototoxic Coal-tar derivatives Pitch, creosote Acridine Anthracene Dyes Anthraquinone Eosin Methylene blue Drugs Phenothiazines Sulfonamides Plants and derivatives Compositae (yarrow) Umbelliferae (celery, parsnip) Rutaceae (lime, lemon) Psoralen 8-Methoxypsoralen Photoallergic Antifungal agents Fentichlor Fragrances Methyl coumarin Musk ambrette Halogenated salicylanilide Soap, deodorants Phenothiazines Agricultural Sunscreens PABA esters Benzophenones Whiteners

Occupational example

Railroad, construction Chemical/laboratory Textile, laboratory

Medical, pharmaceutical Agricultural Vegetable workers Bartenders Medical workers

Perfume/cosmetic industry Soap, cosmetic industry Usage exposure Medical Outdoor workers Textile workers

Table 29.1.12 Contact photodermatitis

Contact Photodermatitis 711

Treatment and prognosis Avoidance of contact with the offending agent is the most important measure in prevention and treatment. Protective measures such as clothing and topical sunscreens may be helpful, but it is important to be sure that the individual is not experiencing photodermatitis resulting from a chemical sunscreen ingredient (e.g., benzophenone). Other treatment measures are the same as those outlined under ACD. Prognostically, phototoxic eruptions usually resolve fairly quickly. Chronic photo-ACD may rarely eventuate into a marked and persistent photosensitivity reaction that continues despite avoidance of the triggering agent. Such persistent light reactivity (chronic actinic dermatitis) is difficult to manage and may require immunosuppressive agents.

References 1. Bruze M, Emmett EA. Occupational exposures to irritants. In: Jackson EM, Goldner R, eds. Irritant contact dermatitis. New York: Marcel Dekker, Inc., 1990:81–106. 2. Shmunes E. Predisposing factors in occupational skin diseases. Dermatol Clin 1988; 6:7–13. 3. Sherertz EF, Zanolli MD. Occupational allergic contact and frictional dermatitis leading to plaques of psoriasis: a challenge in diagnosis. Am J Contact Derm 1991; 2:52–55. 4. Meding B. Epidemiology of hand eczema in an industrial city. Acta Derm Venereol Suppl (Stockh) 1990; 153:1–43. 5. Adams RM, ed. Occupational skin disease, 2nd edn. Philadelphia: WB Saunders, 1990. 6. Rietschel RL, Fowler JF. Fisher’s contact dermatitis, 4th edn. Baltimore: Williams and Wilkins, 1995. 7. Rycroft RJG, Menne T, Frosch PJ, Benezra C. Textbook of contact dermatitis, 1st edn. Berlin: Springer-Verlag, 1992. 8. De Groot AC. Patch testing: test concentrations and vehicles for 2800 allergens. Elsevier: New York, 1986. 9. Warshaw E. Latex allergy. J Am Acad Dermatol 1998; 39:1–25. 10. Hausen BM, Hjorth N. Skin reactions to topical food exposure. Dermatol Clin 1984; 2:567–78. 11. DeLeo V, Gonzalez E, Kim J, Lim H. Phototesting and photopatch testing: when to do it and when not to do it. Am J Contact Derm 2000; 11:57–61. 12. Malten KE. Thoughts on irritant contact dermatitis. Contact Derm 1981; 7:238–47. 13. Dahl MV. Chronic irritant contact dermatitis: mechanisms, variables, and differentiation from other forms of contact dermatitis. Adv Dermatol 1988; 3:261–76. 14. Duarte I, Terumi Nakano J, Lazzarini R. Hand eczema: evaluation of 250 patients. Am J Contact Derm 1998; 9:216–23. 15. Smith HR, Armstrong DK, Wakelin SH, Rycroft RJ, White IR, McFadden JP. Descriptive epidemiology of hand dermatitis at the St. John’s contact dermatitis clinic 1983–97. Br J Dermatol 2000; 142:284–7. 16. Mathias CGT. Contact dermatitis from use or misuse of soaps, detergents, and cleansers in the workplace. Occup Med State Art Rev 1986; 1:205–18. 17. Majoie IM, von Blomberg BM, Bruynzeel DP. Development of hand eczema in junior hairdressers: an 8-year follow-up study. Contact Derm 1996; 34:243–7. 18. Warren R, Ertel KD, Bartolo RG, Levine MJ, Bryant PB, Wong LF. The influence of hard water (calcium) and surfactants on irritant contact dermatitis. Contact Derm 1996; 35:337–43. 19. Meding B. Prevention of hand eczema in atopics. Curr Prob Dermatol 1996; 25:116–22. 20. Nickoloff BJ. Immunologic reaction triggered during irritant contact dermatitis. Am J Contact Derm 1998; 9:107–10.

21. Sun CC, Guo YL, Lin RS. Occupational hand dermatitis in a tertiary referral dermatology clinic in Taipei. Contact Derm 1995; 33:414–8. 22. Wigger-Alberti W, Elsner P. Do barrier creams and gloves prevent or provoke contact dermatitis? Am J Contact Derm 1998; 9:100–6. 23. Berndt U, Wigger-Alberti W, Gabard B, Elsner P. Efficacy of a barrier cream and its vehicle as protective measures against occupational irritant contact dermatitis. Contact Derm 2000; 42:77–80. 24. Boffa MJ, Wilkinson SM, Beck MH. Screening for corticosteroid contact hypersensitivity. Contact Derm 1995; 33:149–51. 25. Hannuksela M. Moisturizers in the prevention of contact dermatitis. Curr Prob Dermatol 1996; 25:214–20. 26. Holness DL, Nethercott JR. Work outcome in workers with occupational skin disease. Am J Ind Med 1995; 27:807–15. 27. Pryce DW, Irvine D, English JSC, Rycroft RJG. Soluble oil dermatitis: a follow-up study. Contact Derm 1989; 21:28–35. 28. Lushniak BD. The epidemiology of occupational contact dermatitis. Dermatol Clin 1995; 13: 671–9. 29. Mathias CG. Prevention of occupational contact dermatitis. J Am Acad Dermatol 1990; 23:742–8. 30. Halkier-Sorensen L. Occupational skin diseases. Contact Derm 1996; 35(suppl):1–120 31. Rycroft RJG. Low humidity and microtrauma. Am J Ind Med 1985; 8:371–3. 32. Hosoi J, Hariya T, Denda M, Tsuchiya T. Regulation of the cutaneous allergic reaction by humidity. Contact Derm 2000; 42:81–4. 33. Buring JE, Hennekens CH. Carbonless copy paper: a review of published epidemiologic studies. J Occup Med 1991; 33:486–95. 34. Konzen JK. Fiberglass and the skin. In: Maibach HI, Gellin GA, eds. Occupational and industrial dermatology. Chicago: Year Book Medical Publishers Inc., 1982. 35. Mathias CGT. Post-traumatic eczema. Dermatol Clin 1988; 6:35–42. 36. Ronchese F. Calluses, cicatrices, and other stigma as an aid to personal identification. JAMA 1945; 128:925–32. 37. Hersle K, Mobacken H. Hyperkeratotic dermatitis of the palms. Br J Dermatol 1982; 107:195–202. 38. Schempp CM, Muller H, Czech W, Schopf E, Simon JC. Treatment of chronic palmoplantar eczema with local bathPUVA therapy. J Am Acad Dermatol 1997; 36:733–7. 39. Nethercott VR, Holness DL, Adam RM, et al. Patch testing with routine screening tray in North America, 1985 through 1989: frequency of response. Am J Contact Derm 1991; 2:122–9. 40. Belsito DV. The rise and fall of allergic contact dermatitis. Am J Contact Derm 1997; 8:193–201. 41. Grabbe S, Schwarz T. Immunoregulatory mechanism involved in elicitation of allergic contact hypersensitivity. Immunol Today 1998; 19:37–44. 42. Krasteva M, Kehren J, Ducluzeau MT, et al. Contact dermatitis I. Pathophysiology of contact sensitivity. Eur J Dermatol 1999; 9:65–77. 43. McFadden JP, Basketter DA. Contact allergy, irritancy and ‘danger.’ Contact Derm 2000; 42:123–7. 44. Zhang L, Tinkle SS. Chemical activation of innate and specific immunity in contact dermatitis. J Invest Dermatol 2000; 115:168–76. 45. Cronin E. Clinical patterns of hand eczema in women. Contact Derm 1985; 13:153–61. 46. Storrs FJ, Rosenthal LE, Adams RM, et al. Prevalence and relevance of allergic reactions to patients patch tested in North America, 1984–1985. J Am Acad Dermatol 1989; 20:1038–45. 47. Bergstresser PR. Ultraviolet immunosuppression 2000. Prog Dermatol 2000; 34(3). 48. Woolner D, Soltani K. Management of hand dermatitis. Comprehensive Ther 1994; 20:422–6

712 Contact Dermatitis 49. Storrs FJ. Permanent wave contact dermatitis: contact allergy to glyceryl monothioglycolate. J Am Acad Dermatol 1984; 11:74–85. 50. Zemstov A. Treatment of palmoplantar eczema with bathPUVA therapy. J Am Acad Dermatol 1998; 38:505–6 51. Milkovic-Kraus S, Marcan J. Can pre-employment patch testing help to prevent occupational contact allergy? Contact Derm 1996; 35:226–8 52. Mellstrom GA, Wrangsjo K, Wahlberg JE, Fryklund B. The value and limitations of protective gloves in medical health service: part II. Dermatol Nurs 1996; 8:287–95 53. Forsberg K, Keith LH. Chemical protective clothing performance index book. New York: John Wiley & Sons, 1989. 54. Orchard S, Fellman JH, Storrs FJ. Poison ivy/oak dermatitis. Use of polyamine salts of linoleic acid dimer for topical prophylaxis. Arch Dermatol 1986; 122:783–9. 55. Marks JG, Fowler JF, Sherertz EF, Rietschel RL. Prevention of poison ivy and poison oak allergic contact dermatitis by quaternium-18 bentonite. J Am Acad Dermatol 1995; 33:212–6.

56. Gawkrodger DJ, Healy J, Howe AM. The prevention of nickel contact dermatitis. A review of the use of binding agents and barrier creams. Contact Derm 1995; 32:257–65. 57. Cohen DE, Scheman A, Stewart L, et al. American Academy of Dermatology’s position paper on latex allergy. J Am Acad Dermatol 1998; 39:98–106. 58. Armstrong DKB, Smith HR, Rycroft RJG. Glove-related hand urticaria in the absence of type 1 latex allergy. Contact Derm 1999;41:42. 59. de Groot H, de Jong NW, Duijster E, et al. Prevalence of natural rubber latex allergy (type I and type IV) in laboratory workers in The Netherlands. Contact Derm 1998; 38:159–63. 60. Wilkinson SM, Burd R. Latex: a cause of allergic contact eczema in users of natural rubber gloves. J Am Acad Dermatol 1998; 38:36–42. 61. Gould JW, Mercurio MG, Elmets CA. Cutaneous photosensitivity diseases induced by exogenous agents. J Am Acad Dermatol 1995; 33:551–73.

29.2 Other Dermatoses Christina A Herrick, Kalman L Watsky The spectrum of occupational and environmental skin disease includes numerous other dermatoses in addition to the various forms of contact dermatitis. In this section of the chapter, we deal with non-malignant skin reactions to chemical, physical, and biologic agents. Skin cancer, the first recognized form of occupational skin disease, is discussed in Chapter 30.10. Acute skin injuries, including lacerations and punctures, account for a large proportion of all occupational injuries, but they are outside the scope of this chapter. Skin disease continues to be the second most common form of occupational illness, with only repetitive trauma disorders accounting for more cases each year.1 Data from the 2001 Bureau of Labor Statistics (BLS) annual survey shows that 12% of non-fatal occupational illnesses reported in 2000 in private industry were diseases of the skin. This represents an incidence of 46 cases per 100,000 full-time workers, although the true magnitude of occupational skin disease may be 10- to 50-fold higher than the annual survey suggests due to under-reporting and misclassification. Data from the United States and elsewhere estimate that contact dermatitis accounts for greater than 90% of cases of occupational skin disease.2,3 The remaining disorders represent a wide variety of cutaneous diseases whose prevalence is not well known. This section discusses the manifestations, pathogenesis, diagnosis, treatment, and prognosis of these dermatoses, organized by their causal agents. CHEMICAL Acne and folliculitis Chloracne Disorders of pigmentation PHYSICAL Radiodermatitis Urticaria Photodermatoses Erythema ab igne Miliaria Frostbite and chilblains Vibration-induced white finger disease Occupational acro-osteolysis and scleroderma Foreign body reactions Blisters and calluses BIOLOGIC Bacteria Fungi Viruses Parasites

OCCUPATIONAL ACNE Acne is a disorder of the pilosebaceous unit. Environmental and occupational acne can be divided into three types: oil folliculitis, acne vulgaris induced or exacerbated by environmental exposures, and chloracne. In each

instance, work-related acne is due to the action of chemical and mechanical factors on the follicular wall.

Oil acne (folliculitis) Workers at risk include machinists; oil field workers; oil refiners; auto, truck, aircraft, and boat mechanics; rubber workers; roofers; and road maintenance workers.

Pathogenesis The hair follicle is particularly susceptible to irritation from lipids, which disperse into sebum. Petroleum distillates such as non-synthetic cutting oils, pitch, and tar may alter the keratinization of the wall of the follicle, causing plugging of the follicle (comedo formation) or induce an inflammatory reaction by rupture of the follicular wall (folliculitis).

Clinical course Oil acne presents as acneiform (follicular) lesions at the location of exposure, most commonly on the dorsae of the hands and forearms. Comedones as well as inflammatory folliculitis may occur in exposed sites. Covered areas of the body also may be involved due to saturation of clothing with oils. Pitch or coal tar acne is associated primarily with comedones across the malar area and periorbitally, with few inflammatory lesions.4

Diagnosis The areas of involvement in oil acne are different from those seen in either acne vulgaris or chloracne (Table 29.2.1). The history usually is adequate to confirm the diagnosis.

Prevention Protective clothing should be used, particularly aprons and arm shields. Mandatory daily laundering of work clothes and end-of-shift showers are helpful. If exposure cannot be prevented, then dilute synthetic water-based cutting fluids should be substituted for neat (i.e., straight petroleumderived) oils.

Treatment and prognosis Approaches to treatment are similar to those for routine acne.5 Oral antibiotics, especially tetracycline and erythromycin, reduce inflammation and treat secondary infection. Topical antibiotics (clindamycin; erythromycin) can also be useful. Comedones are more refractory to treatment but may respond to long-term topical retinoids (tretinoin; adapalene). Once the exposure has ended, the eruption gradually subsides.

Acne vulgaris Persons at risk include workers in fast food restaurants, actors, actresses, models, and cosmeticians.

714 Other Dermatoses Age

Distribution

Clinical features

Oil acne (folliculitis) Acne vulgaris

Any age Peak incidence, ages 11 to 20

Exposed sites Face, neck, chest

Chloracne

Any age

Face, especially the malar crescent and auricular creases, axillae, groin; nose spared

Open comedones, pustules Open and closed back comedones, papules, pustules, cysts, scar Open and closed comedones, straw-colored cysts

Associated conditions None None Xerosis, conjunctivitis, actinic elastosis, peripheral neuritis, liver abnormalities

Table 29.2.1 Differential features of acne

Pathogenesis Acne vulgaris can be induced or exacerbated by numerous environmental stimuli in addition to oil. Friction, heat, and sweating also play a role in the development of acne lesions.

Clinical course and diagnosis The lesions typically involve the face, neck, upper chest and back (see Table 29.2.1). The history of exposure to oils and grease, oil-based make-up, or friction (e.g., from a headband) usually is adequate to arrive at the diagnosis.

Prevention Avoiding contact with the offending substance and adequate hygiene usually are helpful in preventing the condition.

Treatment and prognosis Oral antibiotics, especially tetracycline and erythromycin, reduce inflammation and treat secondary infection. Topical antibiotics (clindamycin; erythromycin) can also be useful. Comedones are more refractory to treatment but may respond to long-term topical retinoids (tretinoin; adapalene).

Chloracne Although originally described approximately 100 years ago by Von Bettman,6 chloracne more recently has entered the world’s lexicon as a consequence of Agent Orange use during the Vietnam War and industrial accidents with dioxin and its related compounds.7 Chloracne is believed to be a sensitive indicator of systemic exposure to specific polyaromatic hydrocarbons. Persons at risk include workers in production of halogenated aromatic hydrocarbonbased pesticides and herbicides, electrical workers exposed to older polychlorinated biphenyl (PCB)-type transformer oils, and environmental workers engaged in cleaning up or disposing of previously produced agents. Table 29.2.2 lists the major chemicals that have been incriminated.

Pathogenesis To a much greater extent than other petroleum products, agents that cause chloracne induce a marked dyskeratosis of the follicular epidermis, which causes a non-inflammatory keratin build-up in the follicle referred to as a comedo (Fig. 29.2.1). Comedo formation often progresses to a character-

Polyhalogenated naphthalenes* Polychloronaphthalenes Polybromonaphthalenes Polyhalogenated biphenyls Polychlorinated biphenyls (PCBs) Polybrominated biphenyls (PBBs)† Polyhalogenated dibenzofurans* Polychlorodibenzofurans, especially tri-, tetra-, penta-, and hexachlorodibenzofuran Polybromodibenzofuran, especially tetrabromodibenzofuran Contaminants of polychlorophenol compounds, especially herbicides (2,4,5-T and pentachlorophenol) and herbicide intermediates (2,3,5-trichlorophenol) 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) Hexachlorodibenzo-p-dioxin Tetrachlorodibenzofuran Contaminants of 3,4-dichloroaniline and related herbicides (Propanil, Methazole) 3,4,3′,4′-Tetrachloroazoxybenzene 3,3,3′4′-Tetrachloroazobenzene Others‡ 1,2,3,4′-Tetrachlorobenzene (experimental) Dichlobenil (Casoron)–a herbicide (clinical only) DDT (crude trichlorobenzene) * Polychlorodibenzofurans and hexachloronaphthalenes may occur as contaminants in some PCBs. † Polybromonaphthalenes may occur as contaminants in some PBBs. ‡ Not confirmed as chloracne-producing agents. With permission from Taylor JS. Environmental chloracne: update and overview. Ann NY Acad Sci 1979; 320: 295–307. ©1979 New York Academy of Sciences, USA.

Table 29.2.2 Chloracne-producing chemicals.

istic straw-colored cystic lesion, the chloracne cyst. The development of chloracne can be generally correlated with serum levels of the chloracne-inducing agent, particularly with tetrachlorodibenzodioxin (TCDD).8 The experience with PCBs and pesticides has been less clear because the specific causal metabolite may differ from what is measured (see Chapters 45 and 48). Individual variation in follicles is clearly evident, with some individuals showing higher serum levels of TCDD without chloracne than others who have the cutaneous disease. The follicular level of the agent may be of greatest importance in producing clinical change.

Clinical course Several clinical features distinguish chloracne from acne vulgaris (Table 29.2.1).3,7 The onset of chloracne is seen within 2–8 weeks of exposure to a chloracne-producing

Pigmentary Disorders 715

Treatment The treatment of chloracne is difficult, although some success has been seen with oral antibiotics, topical retinoic acid, and oral isotretinoin.9 Cyst formation may be prevented by early institution of retinoid therapy.

Prevention

Figure 29.2.1: Chloracne: comedones and scarring on the cheek and neck.

Chloracne-producing agents are unwanted byproducts of industrial processes. Even minute exposures must be avoided. Therefore, the following guidelines should be followed at any plant where chloracne-causing agents are produced or used: (1) clean locker room and shower facilities, (2) use disposable impervious clothing for workers with potential for direct contact, (3) routinely monitor for plant contamination using wipe samples, and (4) routinely educate and monitor workers.

PIGMENTARY DISORDERS agent. The disease will continue to progress and subsequently regress over a 4–6-month period, presuming that there is no further exposure to the chloracne-producing agent. If the exposure has been intense, clinical lesions of chloracne may continue to appear many years after the exposure has ended. The primary lesion of chloracne is the comedo as well as the straw-colored cyst. The inflammatory papules and pustules of acne vulgaris are not evident. The distribution of lesions is also of primary importance (see Table 29.2.1). The most uniquely involved areas are the postauricular folds, the malar crescent, and the genitalia. The nose typically is spared. Associated cutaneous features include xerosis, pigmentation, conjunctivitis, and scarring. Non-cutaneous findings are varied following exposure to the chemicals that cause chloracne.3,7 Hepatomegaly may occur, as may liver damage. Liver damage is associated more commonly with chloronaphthalenes than with PCBs. Porphyria cutanea tarda has been associated with exposure to chloracne-producing agents in some cases, but the causal link remains weak. Hepatic porphyria has been attributed to TCDD exposure. Peripheral neuropathy also has been reported. The chloracne-causing agents (Table 29.2.2) vary in their potency, with TCDD causing chloracne at the lowest concentrations. Microgram concentrations of TCDD applied directly to the skin over a period of time induce chloracne.

Diagnosis A history of exposure to a chloracne-producing agent and lesions consistent with chloracne suggest the diagnosis. Serum lipid and adipose levels of suspect compounds and metabolites should be obtained for confirmation of exposure; this measurement must be performed by an experienced laboratory with gas chromatography/mass spectroscopy (GC/MS) capabilities and carefully controlled comparisons. Biopsy may be useful for diagnosis because chloracne-producing agents cause loss of follicular sebaceous glands. It may be difficult to distinguish actinic elastotic comedones from mild cases of chloracne.

Occupational and environmental causes of cutaneous dyspigmentation are varied and may be exhibited as hyperpigmentation or hypopigmentation. The differential diagnosis should always include inherited or idiopathic conditions, and thus, a thorough investigation should be undertaken in each case.

Hyperpigmentation Workers at risk include those exposed to heavy metals and those working with organic nitrogen compounds and dyes. Compounds that may cause hyperpigmentation are listed in Tables 29.2.3 and 29.2.4.

Pathogenesis Mechanisms of hyperpigmentation include: (1) exogenous pigment deposition, (2) deposition in skin systemically, (3)

Chemical Aniline dyes Azo dyes Danthron Dihydroxyacetone 2,4-Dinitrophenol Dinitrosalicylic acid Henna (2,4-dihydroxynaphthoquinone) Nitrazepam Phenazopyridine (Pyridium) Picric acid Sodium nitrite Tetryl (Tetra-nitromethylaniline)

Color Pigmented contact dermatitis Pigmented contact dermatitis Red-brown Yellow-orange Yellow Yellow Reddish orange Yellow Yellow Jaundiced appearance Yellow Canary yellow

Adapted from Lerner EA, Sober AJ. Chemical and pharmacologic agents which cause hyperpigmentation of the skin. In: Fitzpatrick TB, Wick MM, Toda K, eds. Brown melanoderma: biology and disease of epidural pigmentation. Tokyo: University of Tokyo Press, 1986: 215–227.

Table 29.2.3 Hyperpigmentation: nitro compounds and dyes that stain skin

716 Other Dermatoses Substance Arsenic Bismuth (chromium) Copper Gold (chrysiasis) Iron Lead Mercury Silver (argyria)

Color

Location

Bronze Blue-gray Greenish Blue-gray Brown Pallor and vividity Slate gray Slate gray

Diffuse, especially trunk and proximal extremities Diffuse Hair Sun-exposed areas, especially periorbitally Tattoo Generalized, gingival leadline Skin folds Sun-exposed areas

Adapted from Lerner EA, Sober AJ. Chemical and pharmacologic agents which cause hyperpigmentation of the skin. In Fitzpatrick TB, Wick MM, Toda K, eds. Brown Melanoderma: Biology and Disease of Epidural Pigmentation. Tokyo, University of Tokyo Press, 1986 pp 215–227.

Table 29.2.4 Hyperpigmentation: metals that may be systemically or locally deposited in skin

Figure 29.2.2: Argyria: silver tattoos in a silver worker.

Figure 29.2.3: Postinflammatory hyperpigmentation and hypopigmentation following irritant contact dermatitis.

photoeruptions, and (4) postinflammatory hyperpigmentation.10 Exogenous pigment deposition is typically due to a tattoo (e.g., silver shavings in a jewelry worker) or superficial staining by dye-like compounds. The most important of these factors are listed in Tables 29.2.3 and 29.2.4. Systemic toxicity with a variety of heavy metals can lead to cutaneous dyspigmentation. The most striking form of dyspigmentation is argyria due to systemic deposition of silver (Fig. 29.2.2). The most common forms of occupational and environmental hyperpigmentation are photoeruption and postinflammatory hyperpigmentation (Fig. 29.2.3). Photoeruptions include sunburn as well as the exacerbated sunburn response (phototoxic dermatosis) that results from exposure of the skin to furocoumarins and light in the UVA spectrum.11 Postinflammatory hyperpigmentation is typically dermal and can follow any skin injury. Dermal pigment results from deposition of melanin, hemosiderin, or both.

hands. Pigmentation resulting from heavy metal toxicity may be exacerbated by exposure to the sun. Postinflammatory hyperpigmentation occurs at the sites of skin injury. Hyperpigmentation may be epidermal, dermal, or mixed. The type often can be distinguished by a Wood’s lamp examination, which shows epidermal hyperpigmentation to be accentuated and dermal pigmentation to be diminished.

Clinical course The pattern of pigmentation varies depending on the etiology (Tables 29.2.3 and 29.2.4). With photodermatoses, the pigmentary changes are usually noted in sun-exposed sites: the face, the V of the neck, and the dorsa of the

Diagnosis The pattern of pigmentary change often suggests the diagnosis. Wood’s lamp examination may be helpful. Biopsy may give important information in cases of dermal tattoos or systemic toxicity from heavy metals (Tables 29.2.3 and 29.2.4).

Prevention Sunscreens can be effective in limiting photoeruptions. Protective clothing is needed to prevent exogenous pigment deposition. This protection is doubly important in working with organic dye-like compounds because many of these compounds are toxic and are absorbed through intact skin (see Chapter 42). Prevention of systemic absorption of metals requires the range of industrial hygiene practices.

Pigmentary Disorders 717

Treatment and prognosis The hyperpigmentation that results from tattoos and systemic heavy metal toxicity may be irreversible. Exogenous staining of the stratum corneum resolves without treatment within a period of a few weeks. Postinflammatory hyperpigmentation may persist for months, especially in those who are more darkly pigmented. Epidermal hyperpigmentation may respond to treatment with agents containing hydroquinone and retinoic acid. Side effects of hydroquinone include contact dermatitis as well as exogenous ochronosis. Dermal pigmentation has no effective therapy.

Hypopigmentation Hypopigmentation of the skin due to occupational or environmental causes has two principal etiologies: chemical leukoderma and postinflammatory hypopigmentation (Figs 29.2.4 and 29.2.5). Chemical leukoderma can result from skin contact with substances containing hydroquinone or derivatives of alkyl phenols and catechols (Table 29.2.5). Those at risk include rubber workers, photographic developers, hospital housekeepers, printers, and workers in the oil, paint and plastics industries.3 Cutaneous injury, from inflammation or trauma, may also produce a focal loss of skin pigment (Fig. 29.2.5).

Figure 29.2.4: Chemical leukoderma in a worker exposed to phenolic compounds.

Clinical course Chemical leukoderma was first described in 1939 as an epidemic of depigmentation at a tannery due to the use of rubber gloves containing monobenzyl ether of hydroquinone.12,13 Pigment was lost in areas exposed to the compound but also at sites distant from exposed skin. This pattern of pigment loss may be difficult to distinguish clinically from idiopathic vitiligo. Postinflammatory hypopigmentation occurs at the location of previous skin injury, either from trauma or dermatitis.

Pathogenesis

Figure 29.2.5: Postinflammatory hypopigmentation following a chemical burn due to ethylene oxide.

The compounds known to cause chemical leukoderma are very similar in structure to that of melanin and its precursors, and they are known to have a direct cytotoxic effect

Hydroquinone Monobenzyl ether of hydroquinone Monoethyl ether of hydroquinone (hydroxyanisole) p-Isopropylcatechol p-Methylcatechol p-tert-Butylcatechol p-tert-Butylphenol p-tert-Butylphenol p-tert-Amylphenol Mercaptoamines N-(2-Mercaptoethyl)-dimethylamine hydrochloride 2-Mercaptoethylamine hydrochloride N, N’, N′′-Triethylenethiophosphoramide (thiotepa) Reprinted from Bolognia JL, Pawelek JM. Biology of hypopigmentation. J Am Acad Dermatol 1988; 19:217–255, with permission from the American Academy of Dermatology, Inc.

Table 29.2.5 Compounds known to produce chemical leukoderma.

on melanocytes. The effects possibly are due to the formation of free radicals, leading to damage of cellular membranes and formation of antigens, which activate lymphocytes.14 This theory would explain why the loss of pigment cells may occur at sites distant from the sites of exposure. Postinflammatory hypopigmentation likely is due to a defect in the transfer of pigment from melanocytes to keratinocytes.

Diagnosis Wood’s lamp examination is helpful in distinguishing postinflammatory hypopigmentation from chemical leukoderma. Chemical leukoderma typically demonstrates complete depigmentation. Skin biopsy with special stains for melanocytes also may be helpful. It can be difficult to distinguish idiopathic vitiligo from chemical leukoderma, and therefore, a complete history of chemical exposures needs to be obtained in all instances.

718 Other Dermatoses

Prevention Melanocytotoxic agents should be used only with caution and with strict adherence to protective safety measures. Worker education regarding the effects of these chemicals is essential.

Treatment Chemical leukoderma can respond to long-term treatment with psoralen plus UVA light (PUVA).15 Localized areas of pigment loss also may be treated with epidermal allografts.16 Postinflammatory hypopigmentation usually resolves over time. Permanent postinflammatory pigment loss may require cosmetic cover-up.

RADIODERMATITIS Workers at risk are numerous and include all those involved with radioactive substances. Medical and industrial workers exposed to x-ray-emitting devices and isotopes, workers in uranium mining in the field of nuclear energy, and nuclear waste management personnel are particularly at risk.

Clinical course Cutaneous findings of acute, intense radiation damage include early blistering, with eventual loss of skin and skin appendages in the exposed area. Chronic radiation damage results in epidermal atrophy, enlarged blood vessels, and pigmentary changes (poikiloderma). The clinical appearance is that of a burn scar. Cutaneous carcinoma may develop in radiated skin many years later. Other effects include dryness and cracking of irradiated skin, alopecia, and chronic skin ulcers. If extensive areas of the body are exposed, the skin changes become a small part of the acute radiation syndrome, with its three phases of early nausea and prostration, followed in 1–2 days by a general feeling of wellbeing, and a third phase in about a week of fever, weight loss, gastrointestinal symptoms, and possible death from bone marrow suppression and infection (see Chapter 33.1).

shielding, based on the form of exposure, should prevent radiation damage and should be required for workers with radiation exposures.

Treatment and prognosis Treatment of acute radiation exposure is supportive. Because radiation exposure is irreversible and produces effects many years later, monitoring of exposed individuals for the development of cutaneous malignancies and premalignancies is needed; early diagnosis and treatment may alter the natural history of the condition. Routine use of sunscreens may prevent augmentation of x-ray-induced sequelae by radiation with the longer wavelengths of ultraviolet light.

PHYSICAL URTICARIA Urticaria, also known as hives, is a common condition that may affect up to 20% of the population at some time during the course of their lives. Urticaria may result from occupational or environmental exposures. Contact urticaria (see Chapter 29.1) has been increasingly recognized as an important and potentially serious entity. The physical urticarias may be exacerbated by environmental exposures.

Clinical course Physical urticarias have been divided into several subgroupings.17 Dermographism usually presents as hives, which are elicited by stroking or scratching. Pressure urticaria may be seen at localized points of skin pressure. Cholinergic urticaria presents as small wheals surrounded by wide areas of erythema in response to heat and sweating. Cold urticaria develops at sites of contact with cold and may be familial. Solar urticaria develops in response to light of varying wavelengths. All of these physical forms of urticaria may be accompanied by systemic complaints such as headache, wheezing, or syncope. Rare forms of physical urticarias include heat contact urticaria, aquagenic urticaria, vibratory angioedema, and exercise anaphylaxis.18,19

Pathogenesis Radiodermatitis commonly refers to alterations in the skin produced by ionizing radiation. Radioactive particles pass through and damage cell structures, including chromosomes, particularly in rapidly dividing tissue layers. Lethal damage to basal cells results in initiation of a classic inflammatory response. Death of melanosomes may result in subsequent pigmentary alteration. Non-lethal damage enhances the possibility of cell transformation and subsequent neoplasia (see Chapter 30.10).

Pathogenesis Urticaria can represent an IgE-mediated response to a variety of stimuli. Acute urticaria usually is due to exposure to various foods or medications. A specific cause of chronic urticaria, arbitrarily defined as urticaria lasting more than 6 weeks, is rarely discovered. Many of these cases may represent patients with autoreactive anti-IgE receptor antibodies.20 In evaluating any patient with urticaria, it is important to exclude physical urticarias as a possible cause to avoid unnecessary testing.

Diagnosis A history of radiation exposure and a compatible clinical picture are suggestive. A biopsy may be useful to confirm the diagnosis.

Prevention Radiation can be measured, and routine monitoring is needed for any worker with potential exposure. Protective

Diagnosis Clinical history and appearance usually are adequate to make a diagnosis of physical urticaria. Firm stroking of the skin leads to hives in patients with dermographism. Cholinergic urticaria may be induced by physical exercise or a hot shower. The diagnosis of cold urticaria may be made by using a 3–5-minute skin exposure to ice, a

Erythema Ab Igne 719 5–15-minute exposure to cold water, or a 5–10-minute exposure to cold air. Solar urticaria may be induced by phototesting the skin at the appropriate wavelengths, whereas pressure urticaria requires local application of weights.

Prevention The various forms of physical urticaria require specific preventive measures according to the nature of the eliciting stimulus. Job reassignment may be necessary.

Bergamot Bitter orange Carrot Celery Citron Cow parsley Dill

Giant hogweed Lemon Lime Masterwort Parsnip Persian lime Rue

Adapted from Fisher AA. Contact dermatitis, 3rd edn. Philadelphia: Lea and Febiger, 1986.

Table 29.2.6 Some common plants containing furocoumarins

Treatment and prognosis The treatment of physical urticaria primarily is preventive, although antihistamines and mast cell stabilizers also are used.21 Accurate diagnosis in the workplace setting is critical to avoid unnecessary evaluation. The prognosis for the various forms of physical urticarias is variable, although most patients improve over a period of months to years. A recent study showed that only 16% of patients with physical urticaria were symptom free one year after diagnosis, compared with 47% of those with other forms of chronic urticaria.22

PHOTODERMATOSES Photoeruptions related to occupational and environmental exposures are of several types. Excessive exposure to the sun is a risk for a number of outdoor occupations. Phototoxic reactions are an exaggerated sunburn response and occur in association with a variety of chemical exposures. Photoallergic eruptions require light and a photoallergen in order to fully develop. This last topic is discussed at greater length in Chapter 29.1. Persons at risk include farm workers, roofers, commercial fisherman, and other outdoor workers.

sis. Exposure to photoactive substances is further evidence in support of the diagnosis. A biopsy may be helpful. It is important to exclude other causes of photosensitivity (e.g., lupus erythematosus, medications) in the evaluation of a patient with a photodermatosis.

Prevention Sunscreens are helpful in the prevention of sunburn. It is important to use those with a sun protection factor (SPF) rating of #15 or better. These compounds are less effective in preventing damage from UVA, and therefore, physical sunblocks are more useful in these cases. Use of protective clothing when working with photoactive substances is important in the prevention of phototoxic dermatoses.

Treatment and prognosis Treatment of photoeruptions includes open-wet dressings and bland emollients. Rarely, a course of systemic steroids is necessary for severe cases. Postinflammatory hyperpigmentation resolves over a period of several months. Workers with clinical signs of chronic sun exposure are at risk for cutaneous malignancies and should be followed closely.

Clinical course Sunburn typically presents as a first-degree burn, characterized by erythema in sun-exposed sites followed in 1–2 days by desquamation. Phototoxic dermatoses represent an exaggerated sunburn response that occurs at sites that are exposed to both the sun’s rays and photoactive agents. Second-degree burns may result, characterized by intense erythema and bullae formation. The postinflammatory hyperpigmentation that results may persist for several weeks.

Pathogenesis Sunburn is primarily due to light in the UVB spectrum. Occupational phototoxic dermatoses are principally the result of exposure to light in the UVA spectrum in combination with furocoumarins (Table 29.2.6) and tar products.11,23 UVA is believed to be the wavelength of light primarily responsible for skin aging, although both UVA and UVB have been implicated in photocarcinogenesis.

Diagnosis A history of sun exposure, combined with a typical photodistribution, usually is adequate to make the diagno-

ERYTHEMA AB IGNE This condition may occur in workers exposed to furnaces, such as cooks, stokers, glass blowers, and kiln operators. In addition, long-term exposure to a heating pad, stove, or heater is a risk factor.

Clinical course The early changes are an asymptomatic reticulated pattern of the cutaneous blood vessels (livedo reticularis), which proceeds to reticulated pigmentation. Localized poikiloderma (epidermal atrophy, telangiectasia, and pigment alteration) develops later. The area usually is regional, corresponding to the site of repeated applications of heat. Both squamous cell and Merkel cell carcinomas have been reported to occur in the poikilodermatous area.24,25

Pathogenesis Radiant energy applied to the same area over time produces this change. Persistent vasodilation of the dermal-subcutaneous vasculature is the likely cause of the early changes (Fig. 29.2.6).

720 Other Dermatoses thermoregulation with accompanying heat exhaustion. If the condition progresses due to continued heat exposure, the lesions of miliaria profunda, which are indicative of loss of sweating mechanisms, may result in extreme susceptibility to systemic illness resulting from heat stress (see Chapter 34).

Pathogenesis Sweating and maceration cause plugging of the eccrine sweat duct with ductal keratin. Microbial organisms may invade the macerated keratin and cause further plugging of the duct.

Diagnosis Figure 29.2.6: Erythema ab igne due to prolonged use of a heating pad for a back injury.

Diagnosis is made by the clinical picture, symptoms, and the history of onset after excessive heat exposure and sweating. Excessive sweating also may exacerbate an underlying skin disease.

Diagnosis

Prevention

The local nature of the condition, along with a history of exposure to heat, is suggestive. A biopsy of the affected area may be useful to exclude other conditions associated with livedo reticularis.

Repeated heat exposures should be avoided in workers prone to miliaria. Hexachlorophene soap may be useful to decrease the skin’s bacterial population. Maceration of the skin should be avoided by frequent clothing changes when sweating is profuse.

Prevention Repeated exposure to intense radiant energy must be avoided. Education of workers at risk is the key to prevention.

Treatment Cessation of exposure may bring regression of early changes. Once permanent change has occurred, the area must be monitored for future development of skin carcinoma, similar to the monitoring of skin damaged by ionizing radiation.

MILIARIA Bakers, foundry workers, cooks, coke oven operators, and workers with similar exposure to excessive heat that causes sweating are at particular risk for developing this condition.

Clinical course This condition results from blockage of the sweat ducts. Clinical lesions are on a spectrum encompasssing clear vesicles if the blockage is in the superficial epidermis (miliaria crystallina), macules or papules if the blockage is in the lower epidermis (miliaria rubra), or flesh-colored to pale white papules if the obstruction is in the dermis (miliaria profunda). The trunk is the most commonly affected location, especially the chest, back, submammary, and axillary areas. The palms and groin also may be involved. Symptoms usually are absent with miliaria crystallina, while miliaria rubra and miliaria profunda may be pruritic or painful. Scaling results from the superficial vesicles, whereas deeper plugging, if extensive, may lead to inadequate body

Treatment and prognosis Removal of the worker from the hot working environment cures the disease, although the lesions and symptoms may persist for several days. A period of a week or more should elapse before re-exposure of the individual to the hot environment is attempted, particularly if the eruption is severe enough to cause a decrease in systemic heat tolerance. Some individuals are very susceptible to this condition and may need to avoid occupational heat stress altogether.

FROSTBITE, IMMERSION FOOT, AND CHILBLAINS These conditions caused by local tissue exposure to cold may affect a wide range of workers (see also Chapter 34). Persons at risk for frostbite include those who work in refrigerated environments, fire fighters, sailors, pipeline maintenance workers, and military personnel in frigid northern areas. Sewer and construction workers, as well as military personnel working in extremely cold weather, are at risk for immersion foot. Chilblains affect primarily northern European workers exposed to cold, damp conditions.

Clinical course Frostbite results from exposure to extreme cold, which initially causes blanching of exposed skin, especially on acral areas (Fig. 29.2.7). Dysesthesia accompanies this early cutaneous alteration. As exposure continues, the skin becomes frozen and appears white or waxy and becomes anesthetic. On rewarming, the extent of injury emerges over several days with an initial bluish appearance, then redness, edema, and blistering. Severe pain accompanies

Vibration-Induced White Finger Disease (Hand-Arm Vibration Syndrome) 721 chilblains from vasculitis, connective tissue disease, and cryoglobulinemia.

Treatment and prognosis

Figure 29.2.7: Frostbite of the distal fingers in a bridge worker. The yellow discoloration is due to povidone-iodine (Betadine).

the rewarming. Lines of necrosis develop, and recovery is often prolonged. Immersion foot is cold injury in conjunction with water.26 Continuous exposure for as few as 19 hours may cause an extremity to become pale bluish, edematous, and numb. The clinical course then progresses to erythema, further swelling, and extreme pain as the foot is warmed. This phase may last for weeks to months. Paresthesias, vasomotor instability, and joint stiffness may persist for years. Chilblains are characterized by red to bluish plaques accompanied by burning or itching, with an occasional vesicular component that may occur within 24 hours of exposure to a cold damp climate. Pressure eliminates the color of the plaque, which may progress to swelling and ulceration. These plaques may become chronic nodules, which tend to develop bullae and ulcerate, leading to fibrosis. They may disappear in warm weather and reappear in cold weather. Children frequently are affected.

Pathogenesis During intense cold exposure, there is channeling of blood from superficial vessels to conserve core body heat. Skin tissue can become cold enough for ice crystals to form in cells, leading to tissue death. Immersion foot results from constriction of cutaneous blood vessels by footwear and prolonged low temperatures of the skin not sufficient to cause freezing of the cells. Chilblains result from localized vascular injury due to exposure of the skin to cold and damp areas. Genetic factors that allow vasomotor instability in this type of environment are probably relevant.

Immediate and rapid rewarming of the affected area, usually by immersion in warm water, is recommended for frostbite and immersion foot. Warm drinks and warm food raise core body temperature, increasing cutaneous blood flow. Analgesics for pain and vigorous surveillance for and treatment of infection are required. For frostbite, the depth and extent of tissue necrosis dictate the long-term prognosis. Surgical debridement or amputation of gangrenous tissue may be required. Infection can greatly increase the damage from frostbite. Even when gangrene is absent, long-term effects may include vascular and neurologic alterations of the affected area such as episodic vasoconstriction, increased sweating, and paresthesias of the injured part. Long-term sequelae from immersion foot are similar to those of frostbite and may persist for years. Chilblains are treated symptomatically and do not require rapid rewarming. Vasodilating agents (e.g., calcium channel blockers) have proved effective when symptoms resist or recur.27 Moving to a warm climate is helpful in some instances, but some cases become progressive and chronic despite the use of therapeutic measures.

Prevention Cold injury may be prevented by using protective clothing, which prevents chilling of skin tissues. It is important that the clothing is not tight fitting, so as not to compromise cutaneous circulation. Education of workers regarding the early changes of these conditions and potential problems associated with prolonged cold exposure is essential.

VIBRATION-INDUCED WHITE FINGER DISEASE (HAND-ARM VIBRATION SYNDROME) Workers at risk for vibration-induced white finger disease, or hand-arm vibration syndrome (HAVS), include operators of vibrating tools, piston-operated compressed air tools, and tools that are turned by electric motors or airdriven turbines. Most recently, HAVS has been reported in workers operating high pressure hoses.28 This condition may be exacerbated by occupational activities in cold weather.

Diagnosis Characteristic signs and symptoms following extreme cold exposure are diagnostic of frostbite, although the extent of tissue injury may not be known for days to weeks. Chilblains and immersion foot also have characteristic histories of cold exposure preceding the cutaneous findings. However, the cold exposure is less severe with these conditions, the clinical appearance is different, and a wetwork environment is essential in the diagnosis of immersion foot. A skin biopsy may be needed to differentiate

Clinical course HAVS, characterized by blanching of the fingers when operating vibrating tools, is accompanied in its early stages by mild numbness and tingling. With continued operation, the fingers may develop loss of sensation and decreased dexterity. The blanching may be followed by cyanosis and hyperemia as in classic Raynaud’s phenomenon, although it tends to be asymmetric. Vascular changes are more prominent when vibrating tools are handled in

722 Other Dermatoses cold weather. The number of occurrences and the severity of symptoms may be exacerbated by continued work with vibrating tools. Although at first a single digit may be affected, the condition eventually may involve all the fingers on both hands, although the thumbs often are spared (see also Chapter 23.3). Cold exposure or emotional stimuli leading to pallor of the fingers followed by bluish cyanosis and subsequently erythematous hyperemia, known as Raynaud’s phenomenon, may be idiopathic or familial, but it also may be a cutaneous sign of an underlying connective tissue disease.

Pathogenesis Skin, vascular and nerve changes are seen in skin biopsies of patients with vibration-induced white finger disease.29 Vascular abnormalities in these patients include microangiopathic, vasospastic and thrombotic changes. HAVS has been associated with vibrating frequencies between 30 and 300 Hz. The effects of vibration are probably exacerbated by the pressure required to hold and guide a vibrating tool, which leads to vasoconstriction.

apparently contributing to development of disease.32 Exposure to organic solvents has also been associated with development of systemic sclerosis.33

Clinical course Occupational acro-osteolysis is the term used to define the triad of Raynaud’s phenomenon, sclerodermatous skin changes, and lytic bone lesions reported in workers exposed to vinyl chloride.34 Progressive systemic sclerosis associated with exposure to silica is difficult to distinguish from the idiopathic form of scleroderma. Prominent cutaneous features of this disorder include sclerodactyly, Raynaud’s phenomenon, and digital necrosis.

Diagnosis Patients presenting with Raynaud’s phenomenon without a history of vibration exposure should be questioned regarding exposure to vinyl chloride, silica, organic solvents, and epoxy resins. A thorough investigation for possible underlying connective tissue disease should be undertaken in every case.

Diagnosis

Prevention

A history of exposure to vibrating tools is necessary to make a diagnosis of HAVS. A thorough investigation for possible underlying connective tissue disease should be undertaken in every case.

Workers cleaning polymerization reactor tanks of vinyl chloride need complete skin and respiratory protection. Respiratory protection also is critical in those workers exposed to silica. All workers with Raynaud’s phenomenon, whether or not the condition is job related, should have protection of their hands from cold weather to avoid exacerbation of the disorder.

Prevention Insulation from vibrations, combined with tools designed to decrease the pressure needed for tool guidance, is useful. Operators of vibrating tools should be made aware of the risk of vibration injury to the hand and instructed to report the early changes. All workers with Raynaud’s phenomenon, whether or not the condition is job related, should have protection of their hands from cold weather to avoid exacerbation of the disorder.

Treatment and prognosis Treatment of Raynaud’s phenomenon involves both physical and pharmacologic therapies that are beyond the scope of this text. Avoidance of vibrating tools and protection from cold are important factors. The prognosis depends on the cause and extent of vascular injury. In general, early vibration changes will stabilize or revert if the trauma is discontinued.

OCCUPATIONAL ACRO-OSTEOLYSIS AND SCLERODERMA Cleaners of vinyl chloride polymerization reactor tanks have been reported to develop Raynaud’s phenomenon in association with osteolytic bone changes.30 This association, however, has more recently come under question.31 Likewise, workers exposed to silica dust have been reported to be at risk for developing Raynaud’s phenomenon and scleroderma, with the simultaneous use of vibrating tools

Treatment and prognosis Acro-osteolysis tends to stabilize after withdrawal from vinyl chloride monomer exposure. Scleroderma of any cause, however, tends to be progressive.

FOREIGN BODY REACTIONS Workers in construction, electronics, metal working, and mining are at greatest risk for foreign body reactions, but these reactions occur throughout the workplace.

Clinical course Foreign body reactions occur when the epidermal barrier is broken by various substances. The appearance of the reaction varies depending on whether it is acute or chronic. Acute reactions resemble irritant dermatitis. Chronic reactions typically are more papulonodular. Secondary bacterial infection may complicate the clinical picture. Common examples include reactions to metal filings and wood splinters. Fiberglass may cause an acute reaction that is extremely pruritic (see Chapter 29.1). Other specific causes of foreign body reactions in the occupational setting include beryllium and silica.35,36 An unusual occupational form of foreign body reaction is seen in the exposed skin of clam diggers as a result of exposure to avian schistosomes, so-called swimmer’s itch.37 Hairdressers may develop trichogranulomas following penetration of short hairs into the skin.38

Biologic Causes of Occupational Diseases 723

Pathogenesis A granulomatous response is the consequence of a chronic foreign body reaction. This is typically a non-allergic response, but in the case of beryllium, the granulomatous response is due to delayed hypersensitivity. Acute reactions are eczematous and represent a form of irritant contact dermatitis.

Diagnosis Biopsy may be helpful in arriving at a diagnosis of chronic foreign body reaction. In the case of systemic berylliosis with cutaneous granulomas, the lesions may be indistinguishable from sarcoidosis. In-vitro lymphocyte transformation testing with beryllium often is helpful (see Chapter 39.3).

Prevention Awareness of the risks of foreign body reaction and protective clothing are the two most important factors in prevention.

Treatment and prognosis Systemic berylliosis is only rarely a cause of cutaneous granulomas. Localized granulomas of any cause may be treated surgically. Topical therapies including open wet dressings and topical steroids are useful in the treatment of acute foreign body reactions. Fiberglass may be removed by using tape stripping of the skin.

BLISTERS AND CALLUSES Repetitive tasks often lead to marks on the skin of workers in various industries. The most common of these are blisters and calluses.

Clinical course Calluses appear over areas of trauma, especially on the hands and bony prominences. They are distinguished from warts by preservation of the skin markings throughout the lesion. They also may be associated with lichenification and fissures. Acute trauma to the skin, whether from friction or extremes of temperature, results in blisters. On thin skin, the overlying epidermis may be lost, resulting in an erosion.

Pathogenesis The epidermal response to chronic trauma is thickening, resulting in calluses and lichenification. In patients who are prone to various skin disorders (e.g., psoriasis) trauma can lead to the development of new skin lesions (Koebner phenomenon) (Fig. 29.2.8). Blisters are created when the epidermis separates from the dermis.

Diagnosis Collection of a patient history is essential. The Koebner phenomenon may complicate the diagnosis.

Prevention Because calluses and lichenification are an adaptive response to chronic trauma, prevention is not necessarily

Figure 29.2.8: Psoriasis of the palm in a train conductor demonstrating the Koebner phenomenon due to friction from a ticket punch. Patch testing was negative.

desired. However, thickened lesions may fissure and become secondarily infected. Protective clothing often is useful in these cases, as is skin lubrication.

Treatment and prognosis Calluses may be treated surgically, by paring, or chemically, by using keratolytics (e.g., salicylic acid, urea). Prevention of further trauma is necessary to allow healing. Blisters are best left intact to allow re-epithelialization, unless they are painful or secondarily infected. Erosions should be treated with an occlusive dressing because this approach enhances re-epithelialization.

BIOLOGIC CAUSES OF OCCUPATIONAL DISEASES Occupational exposures are responsible for a wide variety of infectious skin diseases.2 These can occur through direct skin contact, inoculation or inhalation. Up to 5% of all workers’ compensation claims for skin diseases are due to infection, but data are not available to correlate the various agents with given occupations (see also Chapter 22). It is important to recognize any potential case of infectious occupational skin disease as a sentinel health event.39,40 Action by the physician in consultation with public health authorities could prevent disease in other workers.

Bacterial diseases Table 29.2.7 details the most common causes of occupational skin disease due to bacteria. In general, workers at greatest risk are those who work with animals (Fig. 29.2.9) and those in the construction trades. Bacterial infections require identification of the pathogenic organism by a variety of techniques: culture, histology, and immunologic methods. Treatment is based on the definitive identification of a causative agent, and prognosis depends on early and effective treatment. Work-related

724 Other Dermatoses Disease

Causative agent

Workers at risk

Anthrax

Bacillus anthracis

Abattoir workers, dock workers, hunters, workers with animal skin or hair Military personnel, athletes, butchers Veterinarians, meat packers, abattoir workers

Furuncles

Staphylococcus aureus

Brucellosis

Brucella abortus, B. suis, B. melitensis

Erysipeloid

Erysipelothrix insidiosa

Mycobacterial infections, swimming pool granuloma Prosector’s wart

Mycobacterium marinum

Tularemia

Francisella tularensis

Lyme disease Impetigo

Borrelia burgdorferi Group A streptococci, S. aureus

Fishermen, butchers, workers with poultry, rabbits or pigs Gulf fishermen, butchers, fish tank workers Pathologists, morgue attendants, veterinarians, farmers Hunters, veterinarians, abattoir workers

Mycobacterium tuberculosis

Outdoor workers in endemic areas Construction workers, farmers

Clinical findings Ulcerated or bullous dark raised nodule; systemic signs of infection Red, tender fluctuant nodule; local symptoms unless severe Maculopapular or petechial rash or skin ulcer, severe systemic signs of infection Painful, erythematous plaques; lymphangitis Ulcerating papules and nodules along lympathic drainage Warty nodular lesion Ulcer with deep eschar; systemic symptoms accompanied by maculopapular rash with purpura Erythema migrans Crusted erosions

Table 29.2.7 Bacterial diseases

Causative agent Candidiasis Dermatophytes

Workers at risk

Clinical findings

Bartenders, dishwashers Veterinarians, agricultural workers

Chronic paronychia, intertrigo Tinea pedis, manum, corporis

Sporotrichosis

C. albicans Epidermophyton, Microsporum, and Trichophyton species Sporothrix schenckii

Agricultural and forestry workers

Mycetoma

Actinomycetes, Pseudoallescheria boydii

Agricultural workers

Chromomycosis T. versicolor

Fonsecaea pedrosoi Pityrosporum orbiculare

Agricultural workers Workers in hot, humid environments

Ulcerating papule with nodules following lymphatics Chronic, indurated swelling with sinus tracts, usually on hands or feet Verrucous plaques with scarring Scaling hypopigmented or hyperpigmented macules in a shawl-like distribution

Table 29.2.8 Fungal diseases

Fungal diseases

Figure 29.2.9: Infection with Mycobacterium marinum (fish tank granuloma), which demonstrates sporotrichoid spread in pet shop worker.

bacterial infections can be prevented by protective gear and awareness of the risks. Education of workers regarding the potential infectious agents in their work environment is an essential part of prevention.

A variety of superficial fungi are responsible for occupational disease. The most common of these are detailed in Table 29.2.8. In general, workers at greatest risk are those in the agricultural trades. Candida and dermatophyte infections are the most common superficial fungal infections seen in the occupational setting. Candida infection commonly presents as a chronic paronychia due to wet work (Fig. 29.2.10). Dermatophytosis may be acquired by exposure to species that are anthropophilic, zoophilic, or geophilic. An unusual variant of tinea pedis (one hand–two feet tinea) needs to be considered in the differential diagnosis of hand dermatitis (see Chapter 29.1). The specific diagnosis of Candida infection and dermatophytosis may be made by potassium hydroxide examination of scale and by fungal culture. Topical antifungal agents usually are adequate for treatment, although occasionally administration of oral antifungals (griseofulvin, ketoconazole, itraconazole, terbinafine) is necessary. As with other occupationally acquired infections, work-related fungal diseases may be prevented primarily through the use of protective gear and awareness of risks.

Biologic Causes of Occupational Diseases 725 Causative agent Herpetic whitlow Farmyard pox Butcher’s warts

Workers at risk

Herpes simplex type 1 > type 2 Orf–sheep, goats, milker’s nodules–cows (paravaccinia viruses) Human papillomavirus, type 7

Hospital workers Farm workers Meat handlers

Clinical findings Vesicles on an erythematous base Painless papulovesicle, evolves over 6 weeks Verrucous papules on the hands

Table 29.2.9 Viral diseases

Causative agent Leishmaniasis Swimmer’s itch Larva migrans

L. tropica, L. brasiliensis Avian schistosomes Ancylostoma braziliense

Workers at risk Tropical forest workers Skin divers, dock workers Workers in subtropical and tropical beaches

Clinical finding Mucocutaneous ulcers Urticarial papules on exposed skin Mobile, serpiginous plaque, especially on the feet

Table 29.2.10 Parasitic diseases

Figure 29.2.10: Chronic paronychia due to Candida albicans in a bartender, also known as bar rot. Figure 29.2.11: Herpetic whitlow due to herpes simplex virus type 1 in a healthcare worker.

Viral diseases The most significant occupationally acquired viral infection is herpes simplex infection of the finger (herpetic whitlow), which is usually seen in healthcare workers. Farm workers and meat handlers also are at risk for occupational viral diseases, as indicated in Table 29.2.9. Herpetic whitlow can be due to herpes simplex virus type 1 or type 2. It presents as a recurrent, often painful vesicular eruption on an erythematous base (Fig. 29.2.11). Lesions occur at the site of contact, generally on a finger. Untreated infections last for 1 to 2 weeks, though therapy with oral antivirals (acyclovir, valacyclovir, famciclovir) is helpful in shortening the course. Diagnosis is made by a Tzanck smear, showing multinucleated giant cells, or by viral culture. Prevention is through use of adequate barrier methods. Human immunodeficiency virus (HIV) may be transmitted in the occupational setting as well. The risk of percutaneous infection has led to the increased use of gloves, which, in turn, has led to an increased incidence of reactions to gloves in healthcare workers (see Chapter 29.1).

Parasitic diseases Parasites are unusual causes of occupational disease in temperate climates. However, workers in developing countries are at particular risk. Several of the parasitic diseases reported in the occupational setting are included in Table 29.2.10.

References 1. Bureau of Labor Statistics, US Department of Labor, 2001. 2. Lushniak BD. Occupational skin diseases. Prim Care 2000; 27:895–916. 3. Fowler JF. Occupational dermatology. Curr Prob Dermatol 1998; 10:211–46. 4. Adams BB, Chetty VB, Mutasim DF. Periorbital comedones and their relationship to pitch tar: a cross-sectional analysis and a review of the literature. J Am Acad Dermatol 2000; 42:624–7. 5. White GM. Acne therapy. Adv Dermatol 1999; 14:29–58. 6. Von Bettman S. Chlorakne ein besondere Form von professioneller Hauterkrankung. Dtsch Med Wochenschr 1901; 27:437. 7. Tindall JP. Chloracne + chloracnegens. J Am Acad Dermatol 1985; 13:539–58.

726 Other Dermatoses 8. Coenraads PJ, Olie K, Tang NJ. Blood lipid concentrations of dioxins and dibenzofurans causing chloracne. Br J Dermatol 1999; 141:694–7. 9. Zugerman C. Chloracne: clinical manifestations and etiology. Dermatol Clin 1990; 8:209–13. 10. Mosher DB, Fitzpatrick TB, Ortonne J-P, et al. Hypomelanoses and hypermelanoses. In: Freedberg IM, Eisen AZ, Wolff K, et al, eds. Fitzpatrick’s dermatology in general medicine. New York: McGraw-Hill, 1999;945–1017. 11. Seligman PJ, Mathias CG, O’Malley MA, et al. Phytophotodermatitis from celery among grocery store workers. Arch Dermatol 1987; 123:1478–82. 12. McNally WD. A depigmentation of the skin by rubber gloves. Ind Med 1939; 8:405–10. 13. Oliver EA, Schwartz L, Warren LH. Occupational leukoderma: preliminary report. JAMA 1939; 113:927–8. 14. Bolognia JL, Pawelek JM. Biology of hypopigmentation. J Am Acad Dermatol 1988; 19:217–55. 15. Iliev D, Elsner P. An unusual hypopigmentation in occupational dermatology: presentation of a case and review of the literature. Dermatology 1998; 196:248–50. 16. Orentreich N, Selmanowitz VJ. Autograft repigmentation of leukoderma. Arch Dermatol 1972; 105:734–6. 17. Mahmood T. Physical urticarias. Am Fam Phys 1994; 49:1411–4. 18. Shadick NA, Liang MH, Partridge AJ, et al. The natural history of exercise-induced anaphylaxis: survey results from a 10-year follow-up study. J Allergy Clin Immunol 1999; 104:123–7. 19. Luong KV, Nguyen LT. Aquagenic urticaria: report of a case and review of the literature. Ann Allergy Asthma Immunol 1998; 80:483–5. 20. Greaves MW, O’Donnell BF. Not all chronic urticaria is ‘idiopathic’! Exp Dermatol 1998; 7:11–13. 21. Lee EE, Maibach HI. Treatment of urticaria. An evidence-based evaluation of antihistamines. Am J Clin Dermatol 2001; 2:27–32. 22. Kozel MM, Mekkes JR, Bossuyt PM, et al. Natural course of physical and chronic urticaria and angioedema in 220 patients. J Am Acad Dermatol 2001; 45:387–91. 23. McKee RH, Maibach HI. The phototoxic and allergenic potential of EDS liquids. Contact Derm 1985; 13:72–9. 24. Hewitt JB, Sherif A, Kerr KM, et al. Merkel cell and squamous cell carcinomas arising in erythema ab igne. Br J Dermatol 1993; 128:591–2.

25. Iacocca MV, Abernethy JL, Stefanato CM, et al. Mixed Merkel cell carcinoma and squamous cell carcinoma of the skin. J Am Acad Dermatol 1998; 39:882–7. 26. Chow S, Westfried M, Lynfield Y. Immersion foot: an occupational disease. Cutis 1980; 25:662. 27. Rustin MH, Newton JA, Smith NP, et al. The treatment of chilblains with nifedipine: the results of a pilot study, a double-blind placebo-controlled randomized study and a long- term open trial. Br J Dermatol 1989; 120:267–75. 28. Cooke R, House R, Lawson IJ, et al. Hand-arm vibration syndrome from exposure to high-pressure hoses. Occup Med (Lond) 2001; 51:401–9. 29. Noel B. Pathophysiology and classification of the vibration white finger. Int Arch Occup Environ Health 2000; 73:150–5. 30. Dinman BD, Cook WA, Whitehouse WM, et al. Occupational acroosteolysis. I. An epidemiological study. Arch Environ Health 1971; 22:61–73. 31. McLaughlin JK, Lipworth L. A critical review of the epidemiologic literature on health effects of occupational exposure to vinyl chloride. J Epidemiol Biostat 1999; 4:253–75. 32. Haustein UF, Anderegg U. Silica induced scleroderma – clinical and experimental aspects. J Rheumatol 1998; 25:1917–26. 33. Aryal BK, Khuder SA, Schaub EA. Meta-analysis of systemic sclerosis and exposure to solvents. Am J Ind Med 2001; 40:271–4. 34. Markowitz SS, McDonald CJ, Fethiere W, et al. Occupational acroosteolysis. Arch Dermatol 1972; 106:219–23. 35. Rossman MD. Chronic beryllium disease: diagnosis and management. Environ Health Perspect 1996; 104(Suppl 5):945–7. 36. Mowry RG, Sams WM, Jr, Caulfield JB. Cutaneous silica granuloma. A rare entity or rarely diagnosed? Report of two cases with review of the literature. Arch Dermatol 1991; 127:692–4. 37. Osment LS. Update: seabather’s eruption and swimmer’s itch. Cutis 1976; 18:545–7. 38. Hogan DJ. Subungual trichogranuloma in a hairdresser. Cutis 1988; 42:105–6. 39. Newman LS. Occupational illness. N Engl J Med 1995; 333:1128–34. 40. Rutstein DD, Mullan RJ, Frazier TM, et al. Sentinel Health Events (occupational): a basis for physician recognition and public health surveillance. Am J Public Health 1983; 73:1054–62.

Chapter 32 Mechanical Stressors Thomas J Armstrong INTRODUCTION When work contributes significantly as a cause of musculoskeletal disorders (MSDs) or aggravates MSDs, regardless of their cause, the MSD is considered a work-related musculoskeletal disorder (WMSD). This chapter is intended to provide background to help healthcare providers assess work factors that may cause or aggravate MSDs. In particular it focuses on mechanical stressors, because there is significant literature supporting their contribution to WMSDs and to provide guidance for their assessment. It also focuses primarily on the upper limb, although many of the concepts are applicable to the back and other parts of the body. This chapter is not intended to make healthcare providers expert ergonomists or job analysts. Its goal is to help occupational physicians with: (1) walk-through inspections of worksites and (2) the clinical assessment of individual workers. In either case, the physician is likely to have limited time and resources. An understanding of these factors will enable the physician to work more closely with their working patients, managers, engineers, ergonomists, and occupational and physical therapists. Mechanical stresses result from transfer of force between muscles and external work objects, e.g., equipment controls, parts and materials, and movements of the body to reach, grasp and move work objects. Within certain limits the body adapts to mechanical stress, and some stress is desirable to sustain physical fitness; however, there are limitations for adaptation. These limits may be influenced by chronic diseases, previous injuries, and other personal factors. As a result, susceptibility to MSDs varies from person to person and the ‘dose–response’ relationship between mechanical stresses and musculoskeletal disorders is modeled probabilistically. There is an established and growing body of literature that describes the relationship between work activities, mechanical stress, and musculoskeletal disorders. This literature provides guidance for the assessment of physical job demands and their contribution to WMSDs. The literature on WMSDs has been thoroughly reviewed by the Occuptional Safety and Health administration for development of a standard13 that was subsequently withdrawn by the Congress for political reasons. Proposed standards have also been withdrawn in the state of Washington. The strong scientific basis for work-related musculoskeletal disorders has also been reviewed in two separate reports by the National Academy of Sciences in 19994 and 2001.5

WORK ACTIVITIES AND MECHANICAL STRESSORS Figure 32.1 illustrates the relationship between job/task, the environment, and the internal loads on tendons, muscles and adjacent structures. It draws on models put forward by Armstrong et al6 and the NRC5 and NRC & IOM.4 Job task demands specify quantity of work and performance expectations. To achieve this goal the worker must perform: (1) mental actions to obtain and process information and (2) physical actions to hold and use work objects (parts, materials, and controls). The physical actions involve the activation of muscles and the production of forces in the musculoskeletal system to produce the desired effects. Aside from accomplishing the desired work task, these actions are associated with physiologic and mechanical responses. Physiologic responses start with the release of ions to initiate contraction of muscles; they also include consumption of metabolic substrates and production of metabolites. Depletion of substrates, accumulation of metabolites, and ionic imbalance are all associated with transient fatigue response that is manifest by loss of strength, loss of motor control and discomfort. Localized muscle fatigue is a transient response that typically develops and recovers in periods of minutes or hours.7 Recovery begins with the cessation of work and, depending on its severity may require seconds to hours. As a practical matter, workers should not be experiencing effects of fatigue from one day to the next. Longerterm effects include tissue damage due to ischemia and free radical production. Tissue damage stimulates repair and adaptation of damaged tissue, which requires several weeks, assuming that the muscle is not re-injured. Biomechanical responses include tissue stress and strain. Mechanical stress refers to force divided by the area over which is it distributed, and strain is the resulting deformation of the material divided by the unstressed dimensions of the material. Stress may also be used in broad reference to factors that disturb an individual’s physical or mental state, and strain as a broad reference to the resulting disturbances. Tissue loads include the transmission of tensile forces from the muscles via tendons, and enable the body to maintain a mechanical equilibrium with the reaction forces associated with holding, using, and manipulating work objects. When the muscles pull on the tendons, tensile forces are produced in the direction of the long axis

838 Mechanical Stressors

Psychosocial factors

Task requirements • Work location • Work objects • Tools • Work standards

• Grip/pinch force • Contact stress • Posture Internal biomechanical, physiological and mental factors

Muscle force Tendon tension

• Tendon compression/friction • Bursa, ligament & nerve stretching, compression, friction

• Skin compression/friction • Bursa & nerve compression

Figure 32.1: A conceptual model illustrating how external loads impose internal loads on the muscles, tendons, and nerves that may in turn cause or aggravate MSDs.

Joint equilibrium

Flexor muscle

Tendon

Nerve Ligament

Finger force Figure 32.2: The wrist flexion or Phalen’s test is widely used for diagnosis of carpal tunnel syndrome. Flexion of the wrist causes the finger flexor tendons to compress the median nerve against the flexor retinaculum.8 The test may be accelerated by pinching at the same time, which produces tension on the finger flexor tendons and additional pressure on the nerve.9

of the tendon and cause it to stretch. Compression forces are produced between the tendons and adjacent anatomical surfaces that intrude into the space between the point of a tendon’s origin and its insertion, which occur when joints are flexed or extended.

This concept is exploited in the Phalen’s wrist flexion test8 (Fig. 32.2) in which flexion of the wrist causes the median nerve to be compressed against the flexor retinaculum. Numbness and tingling in the areas innervated by the median nerve is a positive sign of carpal tunnel syndrome.

c

Rest

Work

Rest

Work

Rest

Work

Rest

Work

Rest

Work

Rest

Response

b

Work

Response

a

Work

d

Rest

Work

Rest

Work

Rest

Response

Smith et al9 proposed that the test could be improved if patients forcibly pinched at the same time they flexed their wrist, because of increased tension on the finger flexor tendons and pressure on the median nerve. In addition to its diagnostic value, the Phalen’s test demonstrates how certain postures of the body impose secondary loads on tendons and nerves and can contribute to WMSDs. Flexion or extension of a joint also causes the shape of the tissues around the joint structure to change and can produce increased fluid pressure. These pressure changes have been measured around the median nerve in the wrist, and shown to be sufficient to interfere with microcirculation. Tissues demonstrate viscoelastic properties; that is, they undergo both elastic and viscous deformations. Elastic deformations are essentially instantaneous with the application of force. Similarly, recovery is also instantaneous with the removal of the load. Viscous deformation continues to increase after the load is applied. Similarly, recovery requires some period of time following removal of the force. As a result, tendons may continue to stretch if there is insufficient recovery time between successive exertions. Touch and proprioception involve tissue deformation, but tissue deformation also produces pain. As with muscle fatigue, muscle deformation and associated pain will recover with the cessation of work. Elkus and Basmajian10 refer to intramuscular fatigue as the performance changes that occur as a result of changes inside the muscle, and extramuscular fatigue that occurs as a result of changes outside muscles. Although tissue deformations are unlikely to result in acute tissue failure, e.g., tendon rupture, they may be sufficient to interfere with circulation and other physiologic functions. They may be sufficient to create micro tissue damage. In the short term, micro damage affects tissue and capillary permeability. This can lead to swelling following the cessation of work, and explains why many persons experience nocturnal numbness in their hands. Micro damage also stimulates degradation, repair, and adaptation of damaged tissues. This is a likely explanation for non-specific synovitis that results in proliferation of connective tissue and is commonly found inside the carpal tunnel. Thickening of the synovial membranes inside the carpal tunnel is a likely explanation of increased inter-carpal canal pressures in carpal tunnel patients8,11 and secondary carpal tunnel syndrome. It can be seen that there are multiple primary and secondary mechanisms of musculoskeletal disorders, but that they share a common link to the patterns of exertions. Figure 32.3 depicts a general response function to varying exertion intensities and frequency. Figure 32.3a depicts a case where there is sufficient time in between exertions for complete recovery. Figure 32.3b depicts a case in which the response is increased either as a result of a primary factor like loading of the muscle tendon unit, or as a result of co-factors (e.g., posture or vibration). In this case the recovery time is insufficient, so that the response is increased with each successive exertion. Additional recovery time is required to arrest the growing response. Figure 32.3c depicts a case where there is adequate recovery time, but an irregular work element results in an extreme response that is then aggravated with each successive response, shown in Figure 32.3d.

Response

Work Activities and Mechanical Stressors 839

Work

Rest

Work

Rest

Work

Rest

Figure 32.3: Four scenarios for repetitive work. In Case (a), the recovery periods are sufficient for complete recovery of biomechanical and physiologic disturbances that occur during the work periods. The work exposures to mechanical stresses have been increased in Case (b), such that the recovery periods are insufficient and the disturbances are cumulative. In Case (c), the recovery periods have been increased to compensate for the increased work levels in Case (b). In Case (d), the recovery periods are generally adequate until the worker is exposed to unusually high stresses as a result of an irregular work activity. Subsequent actions continue to aggravate the initial disturbance so that the worker is unable to achieve full recovery.

As shown in Figure 32.1, task demands are a primary determinant of the physical stressors that are placed on the worker’s body and ultimately on the expression of symptoms. The load may be influenced by psychologic and sociologic factors.4,5,12 Some of these other factors are not explicitly part of the work requirements, but pertain to the workplace (e.g., manager and coworker). Other factors do not pertain to the job at all, but influence how the worker responds to mechanical stressors, e.g., family concerns, personal attitudes and beliefs, or coping and cognitive skills. These factors may be broadly referred to as psychosocial factors. There is a growing body of literature regarding psychosocial factors, but there is not yet broad consensus on how to evaluate and manage these factors. Management of psychosocial factors is complicated by the personal nature of many of the issues, which may be considered confidential. The concept of biologic and mechanical response to physical exertion provides the foundation for an expo-

840 Mechanical Stressors

Prevalence or incidence

All work-related cases

Work-caused cases

Work-aggravated or precipitated cases

Non-work related cases Work intensity Figure 32.4: General dose–response curve for musculoskeletal disorders and exposure to various physical stressors. Some cases occur irrespective of work exposures. Other cases are due to aggravation of existing conditions and still others are primarily due to work exposures.

sure–response relationship. There is a growing body of epidemiological data that shows increased risk of MSDs with increasing intensity of handwork. Although there is not yet agreement on the best way to measure exposure or risk of MSDs, this is an important concept for those concerned with treatment. Figure 32.4 shows a general exposure– response relationship. Available literature demonstrates that: (1) there is a background risk of MSDs, even at the lowest level of exposure, and (2) the risk increases with increasing exposure. There is disagreement about the location of the inflection point where the risk of WMSDs begins to increase with increasing exposure. Part of the confusion can be attributed to different exposure and outcome metrics among investigators. It may be necessary for the healthcare providers and employers to aim for exposure levels that are below those that will cause an MSD in the absence of predisposing personal conditions. A practitioner may elect to intervene with the onset of symptoms and in the absence of objective findings. Knowledge that the worker is experiencing persistent pain, and is exposed to a physical demanding job, will generally be more important to a practitioner than physiologic parameters such as intercarpal canal pressure, particularly since such physiologic data are not readily available. This approach is reflected in the current OSHA reporting regulations. Although there is a ‘temporary’ moratorium on the reporting of musculoskeletal disorders, the rules provide general guidelines for assessing the ‘work-relatedness’ of worker injuries and illnesses. According to OSHA13 section 1904.5a: ‘Work-relatedness is presumed for injuries and illnesses resulting from events or exposures occurring in the work environment. . .’ The regulations include specific exemptions such as: it is clear the condition resulted from activities to exposures outside of the workplace, the worker was participating in a ‘voluntary’ company fitness program or it was due to motor vehicle accidents while commuting to work, to communicable diseases unrelated to work or to mental illness. Section 1904.5(b)(5)

states: ‘An injury or illness is a pre-existing condition if it resulted solely from a non-work-related event or exposure that occurred outside the work environment’. Section 1904.5(b) describes the meaning of significantly aggravated: • One or more days away from work, or days of restricted work, or days of job transfer that otherwise would not have occurred but for the occupational event or exposure. • Medical treatment in a case where no medical treatment was needed for the injury or illness before the workplace event or exposure, or a change in medical treatment was necessitated by the workplace event or exposure. There is a high burden of proof on the employer and healthcare provider to show that a case is not work related. If there is a question about the contribution of work, Section 1904.5b says: ‘You must evaluate the employee’s work duties and environment to decide whether or not one or more events or exposures in the work environment either caused or contributed to the resulting condition or significantly aggravated a pre-existing condition’. OSHA not withstanding, many employers will want information about jobs performed by injured workers so that they can investigate factors that might affect other workers or, that would pose a barrier to employment by affected workers.

JOB ANALYSIS As a practical matter, the time that a healthcare provider has to assess a worker’s exposure to mechanical stressors may be limited. This makes it important that the provider understand what a job analysis entails to effectively review the detailed analysis of others, to understand job factors that produce mechanical stress and to prescribe meaningful work restrictions. It is important that the job analysis be systematic so that all possible stresses are identified. Several structured approaches have been proposed for assessing mechanical stressors. Selected systems are summarized in Table 32.1 along with the factors they include. These systems provide structured approaches for the analysis of jobs; however, systems developed in one setting are not necessarily applicable to others. For example, Keyserling et al14 describe a checklist developed for automotive manufacturing. The checklist used cycle time of less than 30 seconds as the threshold for high repetition work. Although a 30-second threshold works well for some types of manufacturing work, it does not work well for others, e.g., keyboard data entry. In some cases, healthcare providers may find that all of the information required to apply a given analysis system is not available, and that the action limit or threshold cannot be fully assessed. Finally, even though a threshold may protect most workers, it is likely that some fraction will be affected, and that a healthcare provider will be asked to manage that fraction. An alternative to the structured systems listed in Table 32.1 is an open-ended structured approach in which each factor is considered, in so far as necessary or possible.15 The open-ended analyses require the user to recognize risk factors and exercise judgment about their severity and importance. They provide the user with the flexibility to increase the breadth and depth of the analysis as necessary

Upper limb

Upper limb, trunk

Upper limb & backs Upper limbs

Whole body

Hand, wrist, forearm

ACGIH TLV for Monotask Hand Work Generic18

Washington State generic office & mnfc22

Open ended15 Generic OCRA17 generic, mnfc examples

OWAS, whole body posture assessment, generic43,44 Wrist Velocity36 generic, test on mnfc wkers

No

Included in repetition

No

Included in repetition

No

Ex/min (continuous scale)

No

No

% time spent in posture

% (continuous scale)

Angular velocity & acceleration of wrist (continuous scale) No

No

No

No

No

No

1, 2, 3-5, 6-8 hr

Yes, >2hr/day, >4hr/day

Applies to 4 or more hours per day

4hr/day; intense keying >2hr/day

None to maximum, continuous scale

80%

No

No

No

Work duration

No

Static/ dynamic

Included in repetition

Dynamics

Table 32.1 Summary of selected job analysis systems for assessing upper limb physical stressors

Wrist flexions or extension/ min

Angular velocity & acceleration of wrist (continuous scale)

Hand, wrist, forearm

Strain index, generic, tested on pork, poultry and light manufacturing24

Snook et al45 generic, lab test only

No

Whole body

REBA generic, no test data42

No

No

Yes; static >1 minute or repeated >4 times/min add one point to score

Upper limb and torso

RULA generic work, tested on office work41

Activity score increased if static >1 min, repeated, >4 times/min or very large movements Efforts/min 30s; Hi 1/min & >2hr/day

Defers to expert judgment

No

No

No

Selected factors yes/no

Contact stress

No

No

No

No

No

Defers to expert judgment

No

No

No

Selected factors yes/no

Low temp

Vibration

No

No

No

No

12 m/s2 > 2.5 hr/day

Separate guideline

No

No

No

Yes; three categories

Job Analysis 841

842 Mechanical Stressors for a given situation. The open-ended analysis can be supplemented as needed with some of the standardized procedures listed in Table 32.1. The following section presents a generic open-ended approach. The job analysis is divided into two steps: (1) documentation and (2) assessment of mechanical stressors:

1. Objective/purpose The reasons the job is performed. These can be stated as simply as possible in list form.

Documentation

3. Production standards The quantity of work that must be completed in a given amount of time. Standards may be stated as the number of parts per hour, the time per part, the line speed, or a work duration.

The first step of the job analysis is documentation. The documentation is descriptive information about the job,6,15-17 which is an aid to understanding what the worker does and determining his or her exposure to physical stressors. For example, repetitiveness may be related to a work standard that specifies a certain number of parts be assembled in a given amount of time, or that a certain amount of data be entered into a computer in a given amount of time. Posture stresses may result from the location of a work object or the shape of a tool. An understanding of what the worker does – even at a minimal level – will help the healthcare provider work more effectively with management to identify the causes of MSDs and to facilitate return to work. The following is a general outline of the necessary documentation information. Table 32.2 provides a listing of the basic information that should be considered for documentation. How the documentation is performed will depend on the level of concern and available resources. Some organizations maintain a library of job information. Examples include job descriptions used by personnel for hiring and placement, work methods and standards data used by management for work allocation and production planning, and job safety analyses used by safety managers to identify unsafe work conditions. Often these data are not available in a form that healthcare providers can use. Few healthcare providers have the luxury of visiting the worksite to observe or videotape jobs, to make direct measurements, or to interview workers and supervisors. They are most likely to perform the analysis when they see the worker/patient in their clinic, which may be offsite. As a result it may be necessary for someone to review the available documentation, observe and/or videotape the job, and to interview supervisors and workers. In the absence of this support, healthcare providers often have to obtain the needed information from workers at the same time they are evaluating or treating them. Table 32.3 provides a list of questions for obtaining descriptive job information from worker. These questions are intended as starting points. Additional questions may be required to achieve the desired level of detail. It will be helpful if the healthcare provider visits the worksites on a regular basis so that they have a background perspective for interpreting the worker’s comments. Healthcare providers should understand that workplaces are highly dynamic, and work activities may vary from day to day or even hour to hour depending on the production demands, absenteeism and worker turnover. The time required for completing documentation increases with job complexity. Complexity refers to the number of steps required to complete a job activity and how

2. Tasks Groups or cycles of work elements that have a common objective. Tasks need not be performed in the same sequence, and one task may interrupt another. All jobs are composed of at least one task.

4. Work objects Objects, such as parts and materials, on which work is performed. Work objects have attributes of size, shape, weight, hardness/softness, surface friction, location, orientation and temperature. 5. Tools and equipment Objects or devices held or manipulated with the hands to do something to the work object, such as, insert fasteners, trim parts, or control. Tools and equipment have many of the same properties as work objects. Additionally they may require force to engage, activate or resist (reaction forces). 6. Workstation The equipment that is used to hold or support work objects, equipment, or the worker, such as, the benches, chairs, and assembly lines. Workstations have important location, size and shape attributes. 7. Methods A step-by-step description of the elements required to perform the task, such as, get part, inspect part, insert part, and drive screw. Steps can be listed for the right and left hand, but one list is sufficient if both hands do the same steps. 8. Environment Where the work is performed and its physical characteristics, such as, inside, outside, and air temperature. Table 32.2 Factors to include in job documentation. The level of detail may be adjusted according to the user’s needs and resources 1. Does your job have an official name? Does it have a nickname (how do you refer to your job)? 2. In your own words, what is the purpose of your job? 3. How much are you expected to produce in a given minute, hour or shift – give averages and ranges? 4. Do you repeat the same work ‘tasks’ (e.g., get and assemble parts) over and over in the same way or do you perform different tasks (e.g., perform various maintenance, construction or office tasks as needed) in different ways from one time to the next? Please list the ‘tasks’ that you do. Indicate if they occur in the same or random order. Indicate approximately how many times per day and how long each is performed. 5. For each task list the things that you use: machines, hand tools, parts, and personal protective equipment (e.g., gloves). Also indicate if and how you interact with coworkers and clients. 6. Where is each task performed? Is it hot, cold or just right? Do your hands ever get cold? * Detail may be added or removed depending on the needs and available resources

Table 32.3 Suggested questions for interview regarding work documentation* (can be modified as needed)

much those steps vary from one time to the next. At one extreme are simple mono-task jobs in which a few steps are completed in the same way in a few seconds, time after

Job Analysis 843 time. Examples of these jobs include transcription on a computer or feeding a machine in a production manufacturing setting. At the other extreme are complex multitask jobs with high inter-task and inter-cycle variability. Examples of complex jobs might include maintenance jobs in which the workers must diagnose and fix equipment, or software under a variety of conditions, or utility workers who rotate from job to job. Even short cycle time mono-task jobs may require significant time to ascertain all of the material, process and worker variations. Long cycle time jobs, multi-task jobs and jobs with irregular schedules may require considerably more time. While it often is not possible to do a rigorous analysis, it is important that the analyst recognize the uncertainties associated with their assessment. Some employers videotape jobs and send them with the patient to the healthcare provider. Videotapes can be very helpful if the healthcare provider has time to review them; however, videotapes can be misleading. Video recordings are only a sample of what the worker does in a much larger work shift, with potential variation from day to day. For example, a brief interruption in the supply of parts may leave the worker briefly idle and create the impression that there is a lot of recovery time built into the job. Similarly, an irregular part may require extra attention and make it look like the worker has to hurry to keep up with the process. Unless the person making the videotape understands the job and takes care to record sequences that are representative, the recording may provide an incorrect impression. Video recordings are a two-dimensional representation of a three-dimensional system. As a result, there will inevitably be distortions and it may not be clear if the worker is reaching towards or away from the camera. It may be necessary to record multiple views to see all of the hand actions. The job documentation provides a framework for considering what the worker does and lays a foundation for assessing the physical stressors. Even if a thorough documentation is not possible, knowledge of these issues will help the healthcare provider with their assessment of physical stressors and interact with workers and supervisors to more effectively manage cases. Careful documentation provides a framework for assessing mechanical stressors.

Assessment of mechanical stressors Physical stressors are identified and quantified from: 1. Interviews of workers and supervisors 2. Observations (direct or using videotape) 3. Computations based on job specific attributes, and 4. Instrumentation. This chapter focuses primarily on interviews and observations because they are most readily available to healthcare providers. Computations are only introduced to help the healthcare provider understand the basis for the observational methods.

Repeated and static exertions (hand activity level) Repeated and sustained exertions refer to the time qualities of the exertions of the hand or other body parts. An exer-

tion can be defined as an event during which the physiologic or biomechanical demand is placed on the musculoskeletal system at rate greater than at which recovery occurs. Exertions are depicted in Figure 32.3 as increasing response, and recovery is depicted as decreasing response. The rate of response and recovery will be related to force, posture, contact, vibration and other work factors, as well as personal factors. An exertion often corresponds to a work element or step, such as getting the screw, putting it on the bit, and driving the screw, but one work element and exertion may run into the next without intervening recovery periods. The result may be fewer but longer discrete exertions with reduced recovery time. Important time qualities include exertion frequency, exertion time, and recovery time. As a practical matter, few jobs entail repetition of a single exertion all day, such as shown in Figure 32.3, although some keyboard data entry, transcription, certain food processing, and packing jobs may approximate this. Most jobs entail one or more sequences of exertions. Exertion frequencies, durations and duty cycles are reported as average values and/or as ranges (minimum and maximum). The ACGIH18 refers to repetition as hand activity level (HAL) and spells out the steps for calculating HAL for a detailed methods analysis. It also provides a scale for rating HAL from zero to ten, based on observations of the job or a video recording of the job, including six verbal anchor points based on work by Latko et al19,20 (see Table 32.4a). The ACGIH also includes a recommended exposure limit for various combinations of repetition and force. Ebersole and Armstrong21 examined a procedure for applying the Latko scale in which two trained analysts (a) Verbal anchor points used to rate repeated and sustained exertions of the hand* 0 Hands idle most of the time, no regular exertion 2 Consistent, conspicuous long pauses; or very slow motions 4 Slow steady motion/exertion, frequent brief pauses 6 Steady motion/ exertion; infrequent pauses 8 Rapid steady motion/ exertion; infrequent pauses 10 Rapid steady motion/ continuous exertion, difficulty keeping up (b) The UAW-GM Risk Factor Checklist † Low: Hands/wrists pause to wait for equipment and/or parts Medium: Hands/wrists are in steady motion/pace. Infrequent opportunity for pauses High: Hands/wrists are in constant motion to keep up. No opportunity for pauses. (c) The State of Washington Workplace Ergonomic Standard ‡ The ‘caution zone’ is for an exposure to the following activity, occurring for more than one day per week and one week per year. 1. Repeating the same motion with the neck, shoulders, elbows, wrists, or hands (excluding keying activities) with little or no variation every few seconds more than 2 hours total per day 2. Performing intensive keying more than 4 hours total per day * Hand activity level from 0 to 10.18,19 † Keyserling et al 199314 has been modified to include three ranges of repetition based on Latko et al.19 ‡ Includes a checklist for assessing repetition.22

Table 32.4 Repeated and static exertions (hand activity level)

844 Mechanical Stressors observed and rated jobs in real time, and after discussion re-rated the jobs by consensus. Six raters working in random combinations of two rated a total of 410 automobile assembly jobs. Ninety-one percent of the HAL ratings were within one point on a 0–10 scale. The UAW-GM Risk Factor Checklist initially characterized repetition as low (work cycle time greater than 30 seconds) and high (work cycle time less than 30 seconds).15 This checklist has subsequently been modified to include three levels of repetition: low, medium and high (see Table 32.4b). Medium and high risk are interpreted as having ‘potential risk to some workers’, further analysis is required for high exposures. The State of Washington Workplace Ergonomic Standard includes a checklist for assessing job risk of musculoskeletal injuries.22 The criteria for assessing repetition are included in Table 32.4c. Their criteria ‘Repeating the same motion with the . . . wrists, or hands (excluding keying activities) with little or no variation every few seconds more than 2 hours total per day’ could correspond to the middle and upper parts of the Latko19 and ACGIH scales.18 ‘Performing intensive keying more than 4 hours total per day’ corresponds to the upper limit of Latko and ACGIH scales.

Worker self-assessments. Table 32.5 includes a list for questions that can be asked of the worker to assess their perception of physical stressors. Question 1 pertains to repetition and asks the worker if they can work at his or her own pace. It also addresses if the worker has trouble keeping up and would indicate if the worker is near the upper end of the ACGIH Hand Activity Scale.18 It may be 1. Is there adequate time for your hands to rest between work cycles, between tasks or during breaks, or do you have to use your hands continuously to keep up? (note: your hands may be resting while you are performing tasks such as reading documents, walking to parts, etc.) 2. Are you able to repeatedly exert the force required to get, put, or assemble parts, or to hold and use tools throughout the work shift without excessive fatigue? What, if any, parts of your job seem to require excessive force? How frequently do they occur? 3. Does your job require you to assume any unusual or uncomfortable postures? As appropriate demonstrate extreme: a. wrist flexion and extension, b. wrist deviation, c. forearm rotation, d. elbow flexion, e. reaching overhead, f. reaching forward, g. reaching side to side, h. reaching back, i. neck rotation, j. neck flexion/extension, k. bending, l. twisting and m. squatting. What work activities require these postures? How frequently? 4. Do you strike, press, or rest your body against any hard or sharp objects (e.g., pound parts together, rest wrists on sharp edge of desk or keyboard, lean on elbow? Where? Frequency? 5. Does anything you hold or touch vibrate or shake? What? Where? Frequency? 6. Do your hands get cold? Where? Frequency? * For greater quantification, workers can also be asked to rate their exposures on a scale of 0 to 10 and to indicate exposure duration and frequency. Assessments may be performed for each task and the overall job depending on job complexity, needs and resources.

Table 32.5 Suggested questions for interview regarding exposure to physical stressors*

instructive to compare worker self-assessments of repetitions with those of a third party. Significant differences may indicate where the workers capacity is excessively challenged by the job demands.

Forceful exertion Forces are exerted to hold or move the body or to hold or move work objects. Force is an important determinant of stresses on muscles, tendons and adjacent connective and nerve tissues. Force also is an important factor in muscle metabolic rates (Fig. 32.1). Tissue loads and metabolic rates may be influenced by posture. Recovery occurs when the hands are able to relax. The lowest force and greatest recovery occurs when workers are able to sit with their hands on their lap. A more likely scenario is sitting or standing waiting for a machine to cycle or respond; if the forearms are not supported, muscle activity to support the shoulder girdle may prevent recovery in the neck–shoulder region of the body. The mechanical relationship between external and internal loads is affected by the position of the body. As a result, greater muscle and tendon loads are required to work in some positions. For example, grip strength is typically four or five times greater than pinch strength. Force is also affected by work pace. Rapid or sudden motions will impose inertial loads and require higher muscle forces than for slow steady motions. Force can be expressed on an absolute scale of newtons (kilograms or pounds). It also can be expressed on a relative scale of 0 to 10, where 10 corresponds to the greatest effort possible or 100%. Using a relative scale of 0–10 makes it easier to compare the magnitude of force with other factors. If repetition is rated as 8 and force as 2, this would mean that it is a highly repetitive, but low force job. Physicians might focus more on the possible contribution of repetition than force to patient’s MSD, and an engineer would focus more on steps to reduce repetition than force. A relative force value can be calculated as the required force divided by the strength of an individual or population percentile and multiplying by ten: Relative force (0–10) = Absolute force / Strength *10 A 0–10 scale was adopted by ACGIH so that it corresponded to the Hand Activity Level scale. Because strength varies from group to group, the relative force required for a given job will vary as well. For example, if a tool requires 350N of grip force to cycle, then the required force would be 10 for an average female, but only 5 for an average male. As a practical matter, the relative force will probably be one or two points greater for persons with high strength compared with low strength, because most people exert more than the minimum required force unless they are near their maximum. A list of factors affecting hand force is included in Table 32.6. To understand the relationship between work requirements and force requirements, two examples are considered (Fig. 32.5a). In the first example, the worker is holding an object such as a toolbox or a suitcase. Flexing the fingers

Job Analysis 845

Pinch force, Fp Friction from adjacent files

(a)

Figure 32.5: (a) Holding a tool box requires the worker to flex his fingers against the weight of the tool box. The force requirements of holding the tool box will be equal to the gravity and inertial forces on the tool box. (b) To pull the files from the file drawer the worker must pinch the file hard enough to overcome the weight, friction and inertial forces on the file.

Weight of work object or tool Center of gravity versus center of grip Friction between hand and work object/tool Size/shape of work object/tool Glove fit/stiffness/friction Force to activate controls for machines/tools Force required to join parts and engage tools Reaction forces of tools as they start and stop Mechanical assists that are used to support and transfer materials and tools Tool maintenance Quality control Table 32.6 Factors affecting hand force

against the handle supports the object. The force exerted by the fingers on the work object must be exactly equal to force exerted on the toolbox by gravity. In this case the force requirements to hold the toolbox will be equal to its weight. The relative workload required to hold the toolbox can be calculated by dividing the task force requirements by the strength of the worker. Finger force = Weight of work object The force required to pick up the toolbox will be greater than the force required to hold it because of the inertial

Weight

(b)

forces. As workers approach their maximum strength they will generally slow down. Thus a strong worker may end up exerting as much force as a weak worker because they lift the object more quickly. The second example shown in Figure 32.5b involves pulling and holding a file. In this case the fingers must press against the sides of the file to produce enough friction to overcome the force of gravity and frictional forces from adjacent files. A mathematical expression that describes the relationship between the force exerted by the hand, the weight of an object, and the slipperiness or coefficient of friction of the skin can be written: Finger force ≥ File Weight / 2 x Friction Coefficient This equation is written as an inequality because the worker can exert more than the minimum required force to hold the file. Friction is related to the object’s material properties and moistness of the skin. Coefficients of friction for porous materials, such as paper and leather, are sensitive to moisture and range from 0.25 to 0.4 for dry skin and 0.4 to 0.7 for moist skin.23 Coefficients of friction for non-porous materials such as aluminum, vinyl, and sandpaper are insensitive to moisture and range from 0.4 to 0.6.

846 Mechanical Stressors The simplest case corresponds to holding the file at a stationary location in which the friction forces from adjacent files and inertial forces are zero. The required pinch force can therefore be expressed simply as: Finger force = File Weight / 2 x Friction Coefficient As an example, if a file folder weighs 15N and the coefficient of friction for moist skin and paper is 0.5, the minimum pinch force required to overcome gravity is calculated as 15N; if the coefficient of friction is 0.25 for dry skin, the pinch force is 30N. The required pinch force to hold the folder if the skin is moist is approximately equal to the weight of the folder. Handling of paper folders results in the drying of the skin and decreasing the coefficient of friction. Thus, the required pinch force for the dry skin is nearly twice the weight of the file. Other factors affecting force include the balance of the work object in the hand, the use mechanical assistance, glove fit, stiffness and friction, part quality, and fit and tool maintenance (Table 32.6). Many of these factors are addressed in the UAW-GM Risk Factor Checklist.14 Other systems such as the Strain Index,24 the ACGIH,18 and the State of Washington22 leave it up to the user’s professional judgment or the use of other methods. The relative force can be calculated by dividing the force required by the strength of the hand in the position required to grip the work object. The file is held in pinch posture, so the pinch strength data from Table 32.7 are used to estimate strength. Approximately 99% of the males and 94% of the females with moist skin would be able to hold the folder and have a relative force less than 10. Only 89% of the males and 48% of the females with dry skin would be able to hold the folder and have a relative force less than 10. The relative force would range from 3 to 5 for average males and females with moist skin, and from 6 to 10 for average males and females with dry skin. Relative forces above 10 are not possible, which means persons with pinch strength less than that of a 5th percentile female would not be able to perform the task. Jobs that require ‘pinching an unsupported object(s) weighing 2 or more pounds per hand, or pinching with a force of 4 or more pounds per hand, more than 2 hours total per day (comparable to pinching half a ream of paper) or ‘gripping 10 or more pounds’ would be designated by the State of Washington criteria as a caution zone job.22 This simple example shows how hand force is related to weight and friction. Hand force is related to weight and friction of the work object for other postures in addition to pinch. The relationship between finger force and muscle Posture 46

Power grip Power grip47 Power grip48 Pulp 3 pinch49

Population

5%tile

Steel mill applicants US Adults Office workers US Adults

383.7 333.8 327.8 23.5

force varies significantly from one hand posture to another (Fig. 32.1). Generally, the lowest muscle forces for a given external load are found when the fingers can be closed around the object, such as holding a hammer or a suitcase. It can be seen from Table 32.7 that the hand is roughly 10 times stronger in power grip than in pinch. The finger flexor muscles have to work harder in pinch than in power grip to produce equivalent external forces. It is possible to exert more than the required force. The pulps of the fingers are richly endowed with nerve endings, which provide feedback to the central nervous system when the object is slipping. The system does not provide perfect control, and the amount of force exerted will vary from person to person. As a practical matter, submaximal exertions will be greater than the minimum required. Armstrong et al25 reported that subjects apply an average of three times the required force for keyboard jobs, at the low end of the scale. Westling and Johansson26 refer to this excess force as a safety margin. Frederick and Armstrong27 showed that the safety margin decreases with force as task strength requirements approach the maximum strength of the subject. Thus while strong subjects in theory do not have to exert as much force for a given task as weaker subjects, in practice stronger subjects tend to exert proportionally more effort than weaker subjects. As a result the relative workload difference between strong and weak subjects may vary from predicted. It has been suggested that the amount of force may be affected by psychological stressors, and provides a mechanism by which psychosocial stresses contribute to musculoskeletal disorders.28 The sensitivity of the fingers and ability to control force may be altered by factors such as low temperatures and vibration, which alter the sensitivity of the fingers. It should be apparent that the force required to perform a job typically varies from one work element to the next. It is possible to have a high maximum peak force while having a very low average force. The average force can be calculated as the sum of the average force times the time for each element, divided by the sum of the times. This is the same procedure that is used to calculate time-weighted averages for evaluating chemical TLVs. There is a debate about acceptable levels of average force. Physical endurance goes down precipitously when the average force increases above 15–20% (1.5–2.0 on a relative scale of 0–10), but workers may also experience adverse effects, such as acute and chronic muscle pain, as a result of prolonged and sustained exertions at lower levels. Some investigators argue that complete relaxation is necessary for a muscle to recover and avoid adverse effects of fatigue (see Bystrom and Fransson-Hall7 and NRC4,5 for a summary of this literature).

Male Average 503 487 452 48

(72.5) (92.5) (75.5) (15.3)

95%tile

5%tile

Female Average

95%tile

622.3 639.2 576.2 73.8

237.3 186.4

311 (44.8) 308 (73.9) Not reported 30 (9.4)

384.7 429.6

15.0

* Values may vary greatly from one population to another. See text for discussion.

Table 32.7 Average, fifth percentile and ninety-fifth percentile grip and pinch strength data for adult males and females*

45.9

Job Analysis 847 Observational methods do not have enough precision and accuracy to reliably rate absolute force levels between 1 and 2 on a scale of 0–10. However, it is often possible to determine which of two or more workers is exerting the most force within the low force range.29 The UAW-GM Risk Factor Checklist and the State of Washington ergonomic standard checklist use weights in pinch and grip postures as criteria for forceful exertions.

Worker self-assessments. In Table 32.5, force is assessed by asking the worker if they perceive their job as fatiguing. While this is basically a yes/no question, it does take into consideration other factors, e.g., repetition and posture. Also, workers will add qualifications, e.g., ‘on days when we run a specific part’. The modified Borg31 scale is widely used to obtain worker assessments on a scale from 0 to 10, which correspond to the relative force ratings described above. Workers are shown the scale and simply asked to rate the effort required to perform their job using the verbal anchor points shown in Table 32.8. By dividing the Borg score by 10 and multiplying by the worker’s strength, the absolute force can be estimated. As a practical matter, we have found that a simple 10 cm horizontal line with 0 at one end and 10 at the other works as well as the Borg scale. Workers may be asked to rate the overall job or they may be asked to rate specific tasks or work elements, e.g., the insertion of a connector or use of a hand tool, etc.30 In some cases it is instructive to obtain independent ratings from multiple workers. The Strain Index (Table 32.1) specifies five levels of force achieved on the Borg scale.24

of a work surface holding a keyboard. In this case, the area of contact is 2 cm × 0.5 cm, again assuming a resting weight of 20 N, the contact stress is calculated as 200 kPa. (For comparison a 75 kg person with a 10 × 10 cm heel would produce about 75 kPa pressure if they were to step on your foot with their full weight.) These are average stress concentrations. When the load is distributed over large fleshy areas by a soft surface, these are reasonable approximations of the actual contact stresses. When the forces are exerted against bony areas of the body, such as the elbow or on hard surfaces, then the force is not uniformly distributed and there will be areas of much higher and much lower stress. The level of insult and discomfort increases with the magnitude of contact stress. Some parts of the body are more tolerant to contact stresses than others. For example, a contact force may be tolerated longer if it is distributed in the thenar eminences than over the median nerve at the wrist and palm. Nerve response to even brief percussion over the median nerve is brisk, explaining its use as a diagnostic test (Tinel’s sign, Fig. 32.6). The steps for analyzing contact force are as follows. Contact stresses can be rated on a scale of 0–10, as was described for repetition and force. One of the difficulties with rating contact stress is that the upper limit (i.e., scale designation of ‘10’) is not always clearly defined. Also, some parts of the body are more tolerant to contact stresses

Mechanical stresses from localized contact A localized mechanical stress is defined as a force distributed over some part of the body. For example, resting an arm on the edge of a workstation causes the weight of the arm to be distributed over the area of contact between the arm and armrest; or gripping a tool causes the force required to activate the tool to be distributed between the handle of the tool and the hand. Localized mechanical stresses can be quantified by dividing the contact force by the area of contact. For example, if an armrest, 2 cm × 5 cm, is in complete contact with the fleshy area of the forearm and if the forearm is resting on the armrest with a force of 20 N, the average contact stress is 20 kPa. As a second example, consider an arm resting against the edge

0 0.5 1 2 3 4 5 6 7 8 9 10

Nothing at all Very, very easy Very easy Easy Moderately hard Somewhat hard Hard Very hard Very, very hard

Table 32.8 Ten-point Borg scale for worker assessment of task force requirements

Figure 32.6: The Tinel’s test in which percussion over the base of the palm demonstrates how contact stress at certain locations on the surface of the body can produce stress on underlying tissues.

848 Mechanical Stressors than others.32 Even though a 10-point scale is used, it may be only possible to discriminate between high and low values or low, medium and high values. Jobs that involve high contact stresses from pounding or exertions of connectors may still have low average values because the duration of the exertions is short as compared to the overall work cycle. High average values are sometimes observed where a worker rests his hand or forearms on hard or sharp surfaces. The UAW-GM Risk Factor Checklist uses yes/no questions about contact with the fingers, palm, elbow and armpits with hard or sharp objects to identify contact stress hazards.14 The Washington State workplace ergonomic standard criteria for a ‘caution zone job’ is: ‘(9) Using the hand (heel/base of palm) or knee as a hammer more than 10 times per hour more than 2 hours total per day’.22

Worker self assessments. In Table 32.5, contact stress is assessed by asking the worker if they strike, press or rest their body against any hard or sharp objects. The worker’s hands often provide evidence of repeated contact via calluses, which can be observed in the clinic. In work settings, tape and packing materials on work equipment and on the worker’s hands often provide evidence of contact stresses. Workers’ bodies often show evidence of impact in the form of abrasions and calluses at specific locations contact with equipment (Fig. 32.7).

Posture stresses All postures are stressful if they are maintained long enough. For example, lying in bed or sitting in an easy chair eventually becomes uncomfortable. Conversely, conspicuously stressful postures, such as stretching the arms above the head, may feel good for a brief period. Thus the question becomes: what makes one posture more stressful than another? First, some postures require more

Figure 32.7: It is often possible to see evidence of contact stress on the surface of the body. For example, use of needle nose pliers can produce a major callus near the base of the palm.

effort than others. For example, the body must work harder to produce the same force with an outstretched arm or with a pinch grip than with the arms close to the body or with the hand in a power grip posture. Second, some postures result in greater internal stresses than others. The most common sites of these stresses are in tendons, nerves and blood vessels around joints. For example, exertion of the hand with a flexed wrist causes the flexor tendons to press against the median nerve (see Fig. 32.2). Precipitation of numbness and tingling in the median dermatome of the hand by this maneuver is considered a sign of carpal tunnel syndrome (Phalen’s test). Similarly, extreme extension of the wrist causes increased carpal canal pressure and stretching of the median nerve. Ulnar deviation of the wrist causes the extrinsic abductor and extensor tendons of the thumb to be stretched around the radial styloid process. Discomfort produced by forced deviation of the wrist with the fingers wrapped around the thumb is regarded as a positive sign of de Quervain’s disease (Finkelstein’s test). Stressful upper limb postures for repeated or sustained exertions include elevation of the elbows, reaching behind the torso, extreme elbow flexion, extreme forearm rotation, wrist deviation, flexion or hyperextension, and pinching. Joint angles can be measured in degrees and both peak and average values should be considered. Video methods have been proposed for assessing posture at fixed time intervals.33,34 Corlett et al35 proposed a graphical scheme for rating each major joint based on its degree of rotation. Armstrong et al33 proposed an ordinal scheme for rating postures in which each major joint of the upper limb was divided into three or more zones. The Strain Index uses a system in which wrist posture is divided into five zones: very good (neutral), good, fair, bad and very bad (maximum deviation).24 The UAW-GM risk factor checklist specifies acceptable ranges of postures for each joint. Postures that exceed those ranges are given a star if it occurs less than one-third of the time or a star if it occurs more than one-third of the time. Postures or joint positions can be rated from 0 to 10 where 0 is neutral and 10 is maximum deviation – as was described for repetition, force and contact stress. In most cases it is possible to achieve an extreme joint position in two or more directions. For example, the wrist can be flexed or extended or deviated in a radial or ulnar direction. The elbow can be flexed and the forearm can be rotated. The shoulder can be flexed or extended, it can be abducted or adducted, or it can be rotated. Rating can be performed for the most extreme position of a given joint or each degree of motion can be rated individually. Posture ratings should include both peak and average values. Instrumental methods are generally beyond the scope of this discussion; however, the proliferation of commercially available equipment increases the likelihood that it will be encountered by an occupational health physician. Goniometers that can be attached externally to joints and connected to data loggers are commercially available. As was shown in the introduction to this chapter, there are important reasons why wrist postures and movements may contribute to specific MSDs. Indeed, Marras and

Plant Visit/Walk-Through 849 Schonmarklin36 have presented evidence that wrist velocity and acceleration measurements are associated with risk of cumulative trauma disorders. While these results are promising, further studies are needed to examine the relationship between joint postures and specific disorders before this equipment is widely used. Also, goniometers only measure posture and movement; they do not measure other factors such as force and contact stress. Presently this equipment is best used to answer narrowly defined questions about posture and as a research tool.

Worker self-assessments. In Table 32.5, posture stress is assessed by asking the worker if their job requires them to assume any unusual or uncomfortable postures. It is suggested that the physician demonstrate wrist flexion/ extension and radial/ulnar deviation, extreme forearm rotation, extreme elbow flexion, extreme reaching, extreme neck motion, and, if appropriate, bending. Pictures can be used for greater standardization. The worker should then be asked where and how frequently the extreme postures are encountered. Workers also can be asked to rate posture comfort or discomfort for the overall job or specific tasks.30

grinder wheel, or as reciprocating, in which the tool moves something back and forth, such as a manual hammer, powered chipping hammer, or a rivet gun. Free running (not in contact with another object) industrial quality rotary tools that are well maintained should not exceed physiologic vibration limits; however, this may not be true for reciprocating tools. Some tools encompass both rotary and reciprocating parts. Examples include tools with internal combustion engines, such as lawn mowers and chain saws. Although these tools turn, they have reciprocating components such as pistons that may produce high vibration levels. Controls for machines and vehicles are another source of vibration exposure that should be considered. The State of Washington workplace ergonomics standard specifies maximum exposure duration limits for specified types of tools.22 For example, the limit for impact wrenches, chain saws and percussive tools is 30 minutes per day, while the limit for grinders and sanders is 2 hours per day.

Worker self-assessments. In Table 32.5 the worker is asked if anything he or she holds or touches vibrates or shakes. Subsequent questions would investigate how frequently and how long if the worker answers affirmatively.

Low temperature Strength, dexterity, and sensitivity are profoundly affected by skin temperatures below 18 degrees C. Most people have experienced the effect of cold fingers when working outside on a cold day. People normally exert just enough or slightly more force than is required to keep objects from slipping out of their hand, but cooling the skin results in reduced sensory function and excess force. Sources of cold exposure include ambient air, work materials, and the exhaust air from tools.

Worker self-assessments. In Table 32.5 the worker is asked if his or her hands get cold. If the answer is ‘yes’, subsequent questions would focus on the frequency and duration as well as the cause, e.g., is it due to low ambient temperatures or due to localized cooling from handling cold materials?

Vibration Hand–arm vibration (HAV) has been shown to be associated with a number of soft tissue disorders of the upper limb, secondary Raynaud’s phenomenon being the most notable.37 It is important to remember that HAV occurs in combination with other mechanical stressors. Workers undergo vibration exposure by holding something that vibrates. Thus, vibration exposure inevitably occurs in combination with repeated or sustained exertions. Additionally, these exertions may involve high forces, localized mechanical stresses, posture stresses, and exposure to low temperatures. Although quantitative vibration measurements require specialized instrumentation and trained technicians, it is possible to estimate the intensity of exposure from information about the job and the tools. There are several important considerations in estimating the intensity of HAV from tools. The first consideration is the action of the tool. Tool action can be classified as rotary, in which the tool turns something, such as a threaded fastener or a

PLANT VISIT/WALK-THROUGH Many physicians find that walk-through inspections of workplaces provide important background information for evaluating workers they see in clinic. They also afford an excellent opportunity to learn about the attitudes of workers and employers. Table 32.9a includes a list of useful background information. Information about what the plant does, processes, equipment, materials and numbers of workers provides an overview of the operations and potential exposures. Often this information can be obtained in advance, and can be used to plan the visit. Table 32.9b is intended for collection of data about physical stress exposures for individual jobs and tasks at the time of the visit. It is important to recognize that exposures may change from job to job, hour to hour, day to day and season to season, and that the walk-through only

Company name Contact information Location Department/work area (list) For each department/work area: Description of service or product List of production and service areas Description of: Products/services Tools/equipment Materials General layout Estimate of number of workers on each shift Daily or seasonal variations Source of information Last update Table 32.9a General worksite information that may be collected in advance of plant walk-through

850 Mechanical Stressors Analyst: Job/task: Department: Location Production std: Work objects: Tools: Workstation: Brief job description (list steps): Stressor Repetition‡ Force Posture Wrist F/E Wrist U/R Forearm Elbow Shoulder Neck Back Contact Fingers Palm/wrist Forearm Elbow

Peak (0–10)†

Date:

Avg (0–10)†

Comments

* Additional details can be added as desired. See Table 32.2 and text for details. † Other factors: 0 – Least exposure imaginable 10 – Greatest exposure imaginable ‡ Repetition: 0 – Hands idle most of the time, no regular exertion 2 – Consistent, conspicuous long pauses; or very slow motions 4 – Slow steady motion/exertion, frequent brief pauses 6 – Steady motion/ exertion; infrequent pauses 8 – Rapid steady motion/ exertion; infrequent pauses 10 – Rapid steady motion/ continuous exertion, difficulty keeping up

Table 32.9b Simple data collection forms for walk-through job surveys*

provides a sampling of some of the jobs at one point in time and may not be representative of the entire operation over the entire year or even the entire day. This can also be seen in a grocery store between the middle of the day and in the early evening or days immediately preceding holidays versus after holidays. Table 32.9a specifically asks for information on daily and seasonal variations. The walkthrough should be designed to provide an overview, while allowing additional time to examine jobs with high exposures. Medical visit logs and OSHA records sometimes can be used to identify problem areas, but may be subject to statistical limitations due to small numbers, or variations in reporting. An important source of information is from people at the plant – both workers and managers. It is helpful if both a management and worker representative accompany the physician on the walk-through. This is reinforced by a personal experience by the author during a walk-through inspection of a pork processing plant with a labor and a management representative. The management representative stated that he did not understand why people were having problems, because he himself had worked on the line for a number of years. The labor representative countered that they were previously running only 350 animals an hour through the plant, and now they were running 600. Also, the presence of the labor and management

representatives is usually reassuring to both workers and supervisors. A plant walk-through also provides an opportunity to involve others, e.g., other healthcare providers, engineers, and ergonomists who may be called upon to provide addition job information, develop job accommodations and treat patients. Table 32.9b is intended for assessing exposures on individual jobs or job tasks, providing structure for the survey; it is not a definitive survey tool. The healthcare provider may want to request a detailed analysis using one or more of the tools listed in Table 32.1 as the need arises. The walk-through survey should be documented in writing. It is likely that the healthcare provider will see workers from many plants and a written record will enable them to refresh their memory when they see a patient from that site. The written record also can be shared with the employer. This will help the healthcare provider and the employer reach a common understanding about the exposures. It also will give the employer feedback about exposures at that site so they can determine the feasibility of interventions. The following examples illustrate how plant walkthrough data can be used to supplement worker assessments in the clinic.

Office seating By visiting an office worksite, the healthcare provider discovers to what degree jobs are clerical and involve keyboard and mouse work, filing, phoning, copying and meetings. It also might be discovered that some of the jobs entail repetitive handling of heavy files, while others involve prolonged interaction with the computer to compose documents with both text and graphical components. It may be determined that workers are using some new and some old furniture with varying degrees of adjustability. By inquiring about a patient’s job title, job responsibilities and work equipment, the healthcare provider may be able to quickly assess the worker’s exposure to mechanical stressors. Further discussion could focus on details unique to the case at hand. The following scenario illustrates the use of this job information. A male worker from the office with persistent forearm, neck and shoulder pain visits a healthcare provider in his off-site clinic. It is quickly determined that the worker is an accounts manager who uses the computer and mouse to prepare financial reports and propose budgets. It is learned that the patient has a healthy lifestyle, with no evidence of chronic diseases, traumatic injuries, or excess recreation that might explain his symptoms. Finally it is learned that the worker had a recent deadline that required intense use of the computer and mouse and that he was working extended shifts and skipping breaks. It is also determined that the worker is 1.9 m tall (approximately 95th percentile), does not have an adjustable work surface for his keyboard, mouse, and monitor, and that his chair was difficult to adjust. By demonstrating the job, it appears that the worker has been spending 75% of his work shift, up to 12 hours a day, facing a computer monitor (non-adjustable)

Plant Visit/Walk-Through 851 with his neck flexed to highly flexed and his hands on the keyboard or mouse. The remaining time is spent handling documents and on the phone (often held between the neck and shoulder). A work restriction in this case might specify how much time the worker is permitted to key each day, and frequency of breaks. It might specify the use of a headset or speakerphone. An evaluation of the workstation to investigate modifications and to make sure the worker understands how to adjust the workstation and chair should also be initiated. Restrictions should be specified for a fixed period time, at which point the worker and restrictions would be reassessed.

Kitchen appliance manufacturer A second example involves a manufacturer of small kitchen appliances. It is learned that the jobs include: 200 general utility workers who may be assigned to any job on the assembly line, 10 materials handlers who operate fork lifts and load and unload stock, 20 maintenance workers who set up and service equipment, six custodial workers who perform housekeeping activities, and 30 administrative/clerical workers who work in the office. From discussions with production supervisors and observations of the assembly line, it is determined that the assembly jobs have a 1-minute cycle time. The workers do not appear to be rushing, but there are no definable rest periods. Using Table 32.4 it is estimated that the repetition varies from 4 to 7 on the scale of 10. Most of the parts handled by assemblers are small, with weights below 1 pound. While the overall force requirements are low, some jobs require nearly all of the worker’s strength to install harnesses and join electrical connectors. The small size of the connectors forces assemblers to use their finger tips in awkward locations. It is also observed that about one-third of the workers use small in-line and pistol-shaped power tools. Tool use varied from one fastener per cycle to six per cycle. While most of the workers do not appear to have any trouble holding and controlling tools, many appear to jerk their wrist as the tool starts and stops. Using the methods described above, peak forces for assemblers were rated from 2 to 10 on a scale of 10. Average forces are generally below 2 on a 10 scale, but some jobs are of concern because the force never reaches zero, enabling the hands to rest. These cases occur where workers hold tools in their hand continuously – even while performing non-tool related work elements. As a result, these workers may never fully recover during work cycles (see Fig. 32.3). Contact stress on the hand – particularly on the fingertips – results from grasping and handling of parts and tools. In some cases workers were observed joining parts by pressing and pounding on parts with the base of their palms. It was estimated peak finger and palm contact stresses are in the upper third of a 10-point scale (7–9 for some jobs), although the averages are all in the lower third of the scale 1–3. Postures are observed to vary widely from one workstation to the next, as workers obtain and install parts and use

tools. Some notable examples include shoulder stresses associated with overhead reaching to get suspended tools, reaching for parts in tote pans, and installing parts in the moving assemblies. Some involve wrist flexion and deviation, and extreme forearm rotation to grasp and position parts and tools. Peak shoulder, forearm, and wrist postures are 7–9 on the 10-point scales as described above for repetition, force and contact stress. For most workers the average values are below 3, but for some workers the average shoulder stress is above 5. These workers were sitting down and had to keep their elbows up to reach their work. The only source of vibration exposure is the power tools. It was observed that only about one-third of the workers use power tools. The most fasteners installed with a tool are six, which require less than 6 seconds of the 1-minute cycle time. Air-powered screw drivers generally have low vibration unless they use slip clutches to regulate torque.38 The major stress associated with the hand tools is posture and force. The assembly area is at room temperature. The only exposure to low temperature is from occasional batches of parts that are brought in from an unheated warehouse in the cold season. Similar information would be compiled for other jobs at this worksite, e.g., materials handling, maintenance, custodial activities. These jobs will be much harder to characterize because they involve multiple and often irregular work tasks. Multiple visits or assistance from ergonomists or industrial hygienists may be required to get information about all jobs. Medical visit records and OSHA logs may be used to determine which jobs have a history of injuries and to set priorities for evaluation. One of the first work questions for a patient with MSD symptoms from this plant would pertain to which area of the plant or which job he or she was employed in and for how long. How far back the exposures are traced would depend on the history and duration of symptoms. If the symptoms are new, it may be necessary to go back only a few months. If they are longstanding, it may be necessary to go back several years. Workplace rotations are becoming more and more common, which can increase the time required to get a full work history. At the same time, work rotations may provide an opportunity to compare jobs to see if there is anything about one position relative to another that makes symptoms worse. For example, it may be that the worker’s hand symptoms are worse when performing assembly jobs that require power tools. From the plant visit, the physician knows that hand tools are only used for a small part of any one work cycle, and that the tools do not weigh much or require high torque levels. By asking the worker to demonstrate how the tools are held and how they are used, it is learned that they are held in the hand continuously, even while other assembly operations are being performed, and that worker has to reach over the assembly and point the tool back towards himself to drive two screws. By watching the worker simulate this in the clinic, it is determined that the job requires elevation of the shoulder (7 on a scale of

852 Mechanical Stressors 10), inward rotation of the forearm (8 on a scale of 10) and extreme wrist flexion (9 on a scale of 10). Rather than restrict the worker from the use of all power tools, the restriction is crafted to emphasize avoidance of continuous holding of tools or exertions that require flexion of the wrist or extreme forearm rotation. Further qualifications could be added depending on the severity of the case. The restrictions could be decreased over time. For example, the first week could specify that the worker should not perform any job that provides less than 20 seconds recovery time for each minute and no wrist flexion or forearm rotation more than two-thirds of their range of motion. It might still say ‘no continuous holding of tools’. The following week might reduce the recovery time to 10 seconds, no wrist flexion, and no extreme forearm rotation, and still forbid continuous holding of tools. The physician would have to either instruct the employee on how to interpret wrist flexion and forearm rotation or meet with appropriate plant representatives. In many cases plants will designate someone to coordinate work restrictions between management, engineers, and healthcare providers. These two scenarios illustrate the assessment of work exposures in both plant walk-through and clinical settings. Quality job information is essential for assessing the contribution of work to MSDs and for developing effective work restrictions. It is assumed that the work assessment also considers and prescribes for non-work factors, e.g., existing medical conditions and outside activities that can also cause MSDs. Even in cases where an MSD is due to non-work related factors, work exposures still need to be considered to avoid aggravating those conditions and facilitating return to work.

ANALYSIS OF JOBS FROM THE CLINIC In the absence of first-hand information or summary information from the plant, the healthcare provider will have to rely on his or her interview of the worker. The information included in Tables 32.2 and 32.3 should serve as general guidelines. The healthcare provider will have to explore details as dictated by the needs of the worker and permitted by time. Recognizing the limitations inherent in a clinical assessment of a job, many physicians will issue very conservative work restrictions, e.g., no hand tools, no gripping greater than 10 pounds, nor repetitive work. These may pose a challenge for the plant to accommodate. Physicians who issue such restrictions may want to add some qualifying language to indicate that the restriction was issued in the absence of a formal job analysis and that such an analysis should be performed to clarify the work restriction. Depending on the time or the service arrangement with the employer, the physician might want to arrange to visit the plant or for a trained job analyst to visit the plant and provide more detailed job analysis. Worker assessments are sometimes criticized for not being objective. Studies by Homan and Armstrong39 found that

workers in office settings tended to over-estimate their use of keyboards and mice by as much as 200% when compared with second party observations and electronic surveillance. The discrepancy between self-reports and external observations can be attributed to a difference in perspective and the analyst’s failure to consider the worker’s perspective. Most of the people were next to the computer and in some cases looking at the monitor. From their perspective they were using the keyboard. If the question is how much time they are using the keyboard, they should be asked that directly. They also should be asked what other activities they perform and how much time they devote to those tasks. Discrepancies can then be noted and discussed. It is important that the person asking the questions be perceived as neutral, and not express approval or disapproval.40

CAUSALITY As mentioned above, present and proposed OSHA regulations presume that injuries are work related and place a high burden on employers to prove otherwise. Healthcare providers frequently are called upon to decide if specific MSD cases are related to a specific job or, in some cases, a specific activity or tool. Determining causality goes beyond the scope of this chapter; however, information about exposure to mechanical stressors is an important consideration in determining causality. Most musculoskeletal disorders can be caused by one or more factors. The available data may be insufficient to assign exact probabilities to any one or combination of factors with a high degree of certainty; however, the data may effectively address ‘more probable than not’ criteria for causation. It is possible to determine if and how much a given worker was exposed to a mechanical stressor or group of stressors. This information must be weighed against knowledge of other physical and mental factors that may affect a worker’s risk. This issue is likely to remain within the realm of expert medical opinion and be subject to much controversy over workers’ compensation claims for the foreseeable future. This controversy must not dissuade healthcare providers from assessing mechanical stressors and treating both the cause and the consequences of musculoskeletal disorders.

SUMMARY Musculoskeletal disorders may be caused or aggravated by certain types of work activities. Recognized risk factors include: repetitive or prolonged force, contact stresses, postures, vibration, and low temperatures. The healthcare provider should consider, along with health and personal information, information about the workers’ exposure to mechanical stresses. In most cases, healthcare providers do not have time to visit the worksite each time a worker with symptoms reports to them. Thus they often have to obtain information about the job from the worker. It is helpful if the healthcare provider conducts regular walk-through inspections of workplaces so that they have a background for interviewing the worker and interpreting their responses. It is helpful if the healthcare provider under-

Summary 853 stands how work contributes to musculoskeletal disorders, and uses a structured approach in obtaining and interpreting clinical and workplace data.

References 1. Occupational Safety and Health Administration. Occupational safety and health standards: ergonomics program final rule. Fed Reg 2000; 65(220):68261–870. 2. Keyserling WM. Workplace risk factors and occupational musculoskeletal disorders. Part 1: A review of biomechanical and psychophysical research on risk factors associated with low-back pain. AIHAJ 2000; 61:39–50. 3. Keyserling WM. Workplace risk factors and occupational musculoskeletal disorders. Part 2: A review of biomechanical and psychophysical research on risk factors associated with upper extremity disorders. AIHAJ 2000; 61:231–43. 4. National Research Council and the Institute of Medicine. Musculoskeletal disorders and the workplace: low back and upper extremities. Washington, DC: National Academy Press, 2001. 5. National Research Council Committee on Human Factors. Work-related musculoskeletal disorders report, workshop summary, and workshop papers. Washington, DC: National Academy Press, 1999. 6. Armstrong TJ, Buckle P, Fine LJ, et al. A conceptual model for work-related neck and upper-limb musculoskeletal disorders. Scand J Work Environ Health 1993; 19:73–84. 7. Bystrom S, Fransson-Hall C. Acceptability of intermittent handgrip contractions based on physiologic response. Hum Factors 1994; 36:158–71. 8. Phalen GS. The carpal-tunnel syndrome. Clinical evaluation of 598 hands. Clin Orthop 1972; 83:29–40. 9. Smith EM, Sonstegard DA, Anderson WH Jr. Carpal tunnel syndrome: contribution of flexor tendons. Arch Phys Med Rehabil 1977; 58:379–85. 10. Elkus R, Basmajian JV. Endurance in hanging by the hands. Why do people hanging by their hands let go? Am J Phys Med 1973; 52:124–7. 11. Armstrong TJ, Castelli WA, Evans FG, et al. Some histological changes in carpal tunnel contents and their biomechanical implications. J Occup Med 1984; 26:197–201. 12. Feuerstein M. Biobehavioral mechanisms of work-related upper extremity disorders: a new agenda for research and practice. Am J Ind Med 2002; 41:293–7. 13. Occupational Safety and Health Administration. Regulations (Standards – 29 CFR) Determination of work-relatedness – 1904.5. Cincinatti: OSHA, 2002. 14. Keyserling WM, Stetson DS, Silverstein BA, et al. A checklist for evaluating ergonomic risk factors associated with upper extremity cumulative trauma disorders. Ergonomics 1993; 36:807–31. 15. Keyserling WM, Armstrong T, et al. Ergonomic job analysis: a structured approach for identifying risk factors associated with overexertion injuries and disorders. Appl Occup Environ Hyg 1991; 6:353–63. 16. Barnes RM. Motion and time study: design and measurement of work. New York: Wiley, 1980;407. 17. Colombini D. An observational method for classifying exposure to repetitive movements of the upper limbs. Ergonomics 1998; 41:1261–89. 18. ACGIH. Hand activity level. 2002 threshold limit values for chemical substances and physical agents and biologic exposure limits. Cincinnati: American Conference of Governmental Industrial Hygienists, 2002;109–19. 19. Latko WA, Armstrong TJ, Foulke JA, et al. Development and evaluation of an observational method for assessing repetition in hand tasks. Am Ind Hyg Assoc J 1997; 58: 278–85.

20. Latko WA, Armstrong TJ, Franzblau A, et al. Cross-sectional study of the relationship between repetitive work and the prevalence of upper limb musculoskeletal disorders. Am J Ind Med 1999; 36:248–59. 21. Ebersole M, Armstrong TJ. Inter-rater reliability for hand activity level (HAL) and force metrics. Baltimore: Human Factors and Ergonomics Society, 2002. 22. Washington State. WAC 296-62-051 Ergonomics. 2002. http://www.lni.wa.gov/wisha/ergo/ 23. Buchholz B, Frederick LJ, Armstrong TJ. An investigation of human palmar skin friction and the effects of materials, pinch force and moisture. Ergonomics 1988; 31:317–25. 24. Moore JS, Garg A. The strain index: a proposed method to analyze jobs for risk of distal upper extremity disorders. Am Ind Hyg Assoc J 1995; 56:443–58. 25. Armstrong TJ, Foulke JA, Martin BJ, et al. Investigation of applied forces in alphanumeric keyboard work. Am Ind Hyg Assoc J 1994; 55:30–5. 26. Westling G, Johansson RS. Factors influencing the force control during precision grip. Exp Brain Res 1984; 53:277–84. 27. Frederick LJ, Armstrong TJ. Effect of friction and load on pinch force in a hand transfer task. Ergonomics 1995; 38:2447–54. 28. Bongers PM, de Winter CR, Kompier MA, et al. Psychosocial factors at work and musculoskeletal disease. Scand J Work Environ Health 1993; 19: 297–312. 29. Pascarelli EF, Kella JJ. Soft-tissue injuries related to use of the computer keyboard. A clinical study of 53 severely injured persons. J Occup Med 1993; 35:522–32. 30. Armstrong T, Punnett L, et al. Subjective worker assessments of hand tools used in automobile assembly. Am Ind Hyg Assoc J 1989; 50:639. 31. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 1982; 14:377–81. 32. Fransson-Hall C, Kilbom A. Sensitivity of the hand to surface pressure. Appl Ergon 1993; 24:181–9. 33. Armstrong TJ, Foulke JA, Joseph BS, et al. Investigation of cumulative trauma disorders in a poultry processing plant. Am Ind Hyg Assoc J 1982; 43:103–16. 34. Fransson-Hall C, Gloria R, et al. A portable ergonomic obser-vation method (PEO) for computerized on-line recording of postures and manual handling. Appl Ergon 1995; 26:93–100. 35. Corlett E, Madeley S, et al. Posture targeting: a technique for recording working postures. Ergonomics 1979; 22:357–66. 36. Marras WS, Schoenmarklin RW. Wrist motions in industry. Ergonomics 1993; 36:341–51. 37. ACGIH. Hand-arm (segmental) vibration. 2002 threshold limit values for chemical substances and physical agents and biologic exposure indices. Cincinnati: ACGIH Worldwide, 2002;120–3. 38. Radwin RG, Armstrong TJ. Assessment of hand vibration exposure on an assembly line. Am Ind Hyg Assoc J 1985; 46:211–9. 39. Homan, MM, Armstrong T. Evaluation of three methodologies for assessing work activity during computer use. Am Ind Hyg Assoc J 2003; 64:48–55. 40. McCormick E. Task analysis. Handbook of industrial engineering. New York: John Wiley, 1982;2.4.4–2.4.6. 41. McAtamney L, Corlett E. RULA: a survey method for the investigation of work-related upper limb disorders. Appl Ergon 1993; 24:91–9. 42. Hignett S, McAtamney L. Rapid entire body assessment (REBA). Appl Ergon 2000; 31:201–5. 43. Karhu O, Kansi P, et al. Correcting working postures in industry: a practical method for analysis. Appl Ergon 1977; 8:199–201. 44. Karhu O, Harkonen R, et al. Observing working postures in industry: examples of OWAS application. Appl Ergon 1981; 12:13–17. 45. Snook SH, Vaillancourt DR, Ciriello VM, et al. Psychophysical studies of repetitive wrist flexion and extension. Ergonomics 1995; 38:1488–507.

854 Mechanical Stressors 46. Schmidt RT, Toews JV. Grip strength as measured by the Jamar dynamometer. Arch Phys Med Rehab 1970; 51:321–7. 47. Imrhan, SN, Loo CH. Trends in finger pinch strength in children, adults, and the elderly. Hum Factors 1989; 31:689–701. 48. Josty IC, Tyler MP, et al. Grip and pinch strength variations in different types of workers. Br J Hand Surg 1997; 22:266–9.

49. Dempsey PG, Ayoub MM. The influence of gender, grasp type, pinch width and wrist position on sustained pinch strength. Int J Ind Ergon 1996; 17:259–73.

Chapter 31 Occupational Injuries Bruce H Alexander, Frederick P Rivara BACKGROUND Injuries are a leading cause of work-related morbidity and mortality among persons in the labor force, and are responsible for substantial financial loss and decreased productivity. The US Department of Labor estimated that from 1997 to 2001 an average of 6036 fatal work-related injuries occurred annually in the US, with 5915 being reported in 2001.1 These numbers exclude losses due to the terrorist attacks of September 11, 2001, which accounted for an additional 2886 work-related fatalities. In the same time period, 4.9 million non-fatal injuries were reported by the Bureau of Labor Statistics.2 The long-term and economic burden of work-related injuries is substantial. The National Safety Council (NSC) estimates that 3.9 million disabling occupational injuries occurred in the United States in 2001.3 The total estimated costs for these injuries were $132.1 billion, which included $24.6 billion in medical costs alone.3 Work-related injuries also result in enormous losses in worker productivity. Injuries sustained in 2001 caused an estimated 85 million lost workdays, and another 45 million lost workdays were attributed to injuries from previous years. The cumulative economic effect of these lost workdays totaled $69.2 billion in lost wages and productivity. On an individual basis, approximately $970 of every worker’s annual productivity is absorbed by the costs of occupational injury. The true number and costs of workrelated injuries are unknown, and the NSC estimates are subject to debate. However, there is little debate on the importance of work-related injuries and the need to improve injury prevention measures.

THE SCIENCE OF INJURY CONTROL Over the last three decades, the field of injury control has developed a solid scientific basis by which a substantial reduction in the morbidity and mortality from trauma can be expected. Injuries are viewed as preventable problems, not as ‘accidents’ that occur without pattern, predictability or reason. Viewing injuries as ‘accidents’ fosters a sense of fatalism, in which a certain number of deaths and permanently disabling injuries are believed to be inevitable. One of the basic premises of this approach to trauma is conceptually separating the injury (i.e., the trauma to the individual) from the events leading to the injury. A motor vehicle collision is separate from the injuries that may or may not occur in the collision; the injuries are not

an inevitable outcome of such an event. Even if the collision cannot be prevented, the chance of injury may be substantially decreased through the use of seat restraints and vehicle design to protect the occupants. Likewise, a slip from a roof does not need to lead to a fall and injury, if the person is restrained with a safety harness. Once injuries occur, the outcome (recovery, temporary or permanent disability, or death) is also influenced by the availability of adequate pre-hospital care at the scene, rapid transport to an appropriately equipped trauma center, and the provision of rehabilitation care to return the individual back to his or her level of functioning before the injury. The emphasis is on injury control, a more comprehensive concept than the more traditional approach, which encompasses only injury prevention. The paradigm of injury control incorporates etiologic research, interventions, and outcomes (Fig. 31.1).4 Etiologic research explores not only the causes of injury and the factors that reduce the frequency and severity of injury, but also how the body responds to energy transfer during an injury event and the healing process. The intervention phase of injury control focuses on modifying risk factors to prevent an injury from occurring and ameliorating the negative effects of injury events and any injuries that occur. The final aspect of injury control is to change the outcomes of hazardous situations by reducing the rate and severity of injury, decreasing the costs of injury and their impact on productivity, improving trauma care, and the overall quality of life of injured workers. A useful tool for examining injury problems and determining potential avenues for intervention is the matrix developed by Haddon (Table 31.1).5 This takes the classic epidemiologic paradigm of agent, host, and environment and applies it to the pre-event, event, and post-event factors associated with the occurrence of injury. The matrix also separates the physical environment from the sociocultural environment of laws, expectations, and norms that affect injuries and prevention strategies. A large body of research conducted over the last two decades has established that injury prevention strategies are likely to be most successful if they operate automatically, without needing repeated behavioral change on the part of the individual.6 Thus, modification of the environment or of the product and vehicle to reduce the hazards associated with their use are likely to have the greatest impact on injuries. Examples of these successes are airbags, anti-lock brakes, interstate highways, rollover protection devices on tractors, and bulletproof partitions in taxicabs. Legislation and regulation have also proved to be effective

826 Occupational Injuries Figure 31.1: Conceptual model for occupational injury control. (Adapted from: Rivara FP. Introduction: the scientific basis for injury control. Epidemiol Rev 2003; 25:20–3 by permission of Oxford University Press.)

Etiologic research

Epidemiology, behavioral sciences

Biomechanics

Basic science

Causes of injury Risk and protective factors

Human tolerance for damage

Pathophysiology of trauma and healing

Interventions

Equipment and environment design

Prevention programs

Trials and evaluations of interventions

Outcomes

Injury rate and severity

Costs and productivity

Quality of care

Factors influencing injury SocioIndividual Agent/ Physical economic behavior vehicle environment environment Pre-event Phases of Event injury Post-event *Adapted from: Haddon W. Advances in the epidemiology of injuries as a basis for public policy. Public Health Rep 1980; 95:411–21.

Table 31.1 Haddon matrix*

adjuncts to decreasing the toll from injury. Mandatory seat belt use and motorcycle helmet laws have resulted in a substantial reduction of fatalities on the nation’s highways. Gun control laws may help to curb the epidemic of homicides that occur both at home and at work. The effectiveness of educational programs to change behavior is difficult to establish. Training workers in the proper use of equipment is a key safety measure in many industries. The fact that injuries are often associated with unsafe behavior attests to the difficulties inherent in this approach. Since its inception in 1971, OSHA has required engineering changes and workplace controls as the primary means of reducing hazards in the workplace.

Health-related quality of life

The severity of an injury will vary depending on the mechanism and the underlying causal force. When evaluating the effectiveness of control programs or prioritizing prevention efforts, an understanding of the severity of injuries is necessary. Several algorithms for scoring the severity of injuries have been developed, although they have traditionally been directed toward predicting the survival and functional outcome of victims of very severe injuries. Some essential injury severity scoring systems include: the Abbreviated Injury Score (AIS), developed by the Association for the Advancement of Automotive Medicine,7 that is primarily based on the diagnostic classification and location of injury, and the Injury Severity Score (ISS),8,9 which is directly calculated from the most severe AIS scores of each body region and was developed as a method to characterize multiple injuries. Limitations of these scoring systems have led to the proposal of potential alternatives, including the New Injury Severity Score (NISS)10 and the Anatomic Profile Score (APS),11 that are intended to more accurately describe severity due to multiple injuries. More complex injury severity rating systems, which include physiologic parameters, have been specifically developed for evaluating patients in trauma centers.12–14 The above scoring systems, and several other proposed algorithms, are directed at the most severe injuries and often require substantial clinical information. A majority of the work-related injury burden may result in a range of

Sources of Data for Work-Related Injuries 827 Indicators • Death • Permanent total disability : loss of functions necessary for work or routine life skills. • Permanent partial disability : limits work or normal activities. • Time-loss injuries: injuries that meet a minimum amount of time away from work, allowing the injured worker to collect worker compensation wage replacement. The criteria vary by state. • Return to work: return to work or return to normal tasks. This may be influenced by the occupation and usual job tasks. • Re-assignment of work tasks/Restriction of work activities: injury requires change in normal work duties. May or may not accompany reduction of income. Care received • Hospitalization: most severe injuries receiving medical treatment. Duration of hospitalization, procedures, and cost may correlate with severity. Usually only available through worker compensation insurance records. • Emergency room: urgent care needed. • Physician office: professional medical care needed. May or may not be urgent. • First aid: these injuries are not generally reportable unless they result in a restriction or re-assignment of work activities. Table 31.2 Potential indicators of severity for work-related injuries based on administrative data. Adapted from Haddon W. Advances in the epidemiology of injuries as a basis for public policy. Public Health Reports 1980; 95:411–421, by permission of Oxford University Press

permanent or temporary disability and medical care, but the relative severity of most of these injuries is not distinguishable using the existing scoring systems. Moreover, most of the available data for studying occupational injuries reside in administrative or surveillance databases, which may not contain the information necessary to calculate these injury severity scores. Alternative methods of evaluating injury severity for occupational injuries are based on the outcome of the injury and the immediate care received (Table 31.2). The most severe injuries are fatal, followed by those injuries resulting in either total or partial permanent disability. The level of medical treatment required, ranging from first aid to hospitalization, and other administrative variables, including time lost from work and re-assignment of usual tasks, can be used to classify injuries by severity. These methods, however, are not well validated and may be a function of occupation as well as the nature of the injury. An injured foot may not interfere with the work of a typist, but would prevent a construction worker from numerous activities.

SOURCES OF DATA FOR WORK-RELATED INJURIES Data sources that characterize the injury problem and prioritize prevention measures are a fundamental element of injury prevention programs. Characterizing an injury problem and evaluating prevention efforts require effective means for injury surveillance. Data resources available for general injury surveillance and research include vital records, hospital discharge data, surveillance of emergency rooms, and trauma registries.15,16 The extent to which

these traditional sources of injury data can capture workrelated injuries is quite variable. Several sources of data on occupational injuries in the United States are available; however, each source uses different methods of collecting information. The usefulness of these data requires an understanding of how the data are collected and their attendant limitations. The sources presented below represent the types of data routinely collected in the United States, but are by no means the only sources of data on occupational injuries.

Fatal work-related injuries In 2002 an estimated four fatal work-related injuries occurred in the United States for every 100,000 persons employed. Historically, the number of work-related fatalities has been difficult to determine. In the United States, the Bureau of Labor Statistics and the National Institutes of Occupational Safety and Health (NIOSH) previously had systems that used different methods. The BLS primarily used employer reporting, while the NIOSH National Traumatic Occupational Fatality (NTOF) database relied solely on death certificate information collected by the National Center for Health Statistics. Other sources of information on fatal injuries, including OSHA reports, medical examiners, and workers’ compensation claims, collected detailed information about fatal injury events, but did not capture most injuries. The available surveillance data were also used by other organizations, specifically the National Safety Council (NSC), to derive estimates of work-related fatalities. The different approaches to fatality surveillance yielded markedly variable results and the lack of consistency represented a significant limitation. A review by Stout and Bell of the various reporting methods concluded that death certificates had the highest capture rate but only identified, on average, 81% of occupational fatalities, followed by medical examiner records (61%), workers’ compensation claim data (57%), and lastly, OSHA records that identified only 32% of occupational fatalities.17 A problem common to all systems was identifying the work-relatedness of a fatal injury. For example, identifying work-related fatalities from death certificates relies on the recorder to check the designated ‘injury at work’ box, a fact that is not always apparent to healthcare professionals treating the injury, or coroners and medical examiners. Incidents that happen on an employer’s premises while the individual is conducting their job-related tasks are fairly readily identified. However, when the person is self-employed or the location of the incident is not at the work site (e.g., a fatal motor vehicle crash for a person in sales), the ability of most single systems to capture such incidents is limited. Consequently, a segment of the work-related fatalities is under-reported and it is unlikely these are distributed evenly by industry, occupation or cause. Recognizing the limitations of existing systems, the BLS created the Census of Fatal Occupational Injuries (CFOI) in 1992.2 This system uses multiple sources of information to identify cases of fatal work-related injuries, including death certificates, workers’ compensation claims, newspapers,

828 Occupational Injuries and investigations by OSHA or other state agencies. Each state is contracted to collect available data within the state and provide the BLS with the information. As with any surveillance system, the CFOI will not capture all fatal work-related injuries; however, the use of multiple data sources improves the probability of capture and will likely provide more information on the characteristics of the injured worker and cause of the injury. Surveillance systems like CFOI are designed to enumerate cases, but are limited in their ability to identify contributing causes to the fatality. These depend on investigations of fatal injury incidents conducted by federal and state regulatory agencies. While these incident investigations are conducted too infrequently to qualify as classic surveillance systems, they provide a wealth of qualitative information on specific injury events and are examples of sentinel surveillance tools that can identify immediate problems. OSHA has developed the Integrated Management Information System (IMIS) based on OSHA inspection records. OSHA is responsible for workplace safety and health inspections in the 29 states and territories that do not have a state occupational health and safety plan. The other 23 states and two territories with state plans submit inspection reports to OSHA for the IMIS. Other than routine inspections, a large number of OSHA inspections are performed in response to fatalities, which are required to be reported to OSHA within 48 hours of occurrence. OSHA does not investigate all fatalities but focuses on those that appear to be related to workplace safety. Inspections also are conducted in response to worker complaints and as a follow-up of citations from routine inspections. NIOSH has implemented the state-based Fatality Assessment and Control Evaluation Program18 that conducts detailed investigations of specific types of fatal work-related injury incidents. Collection of these specific data can alert employers and workers to particular problems. NIOSH publishes a series of requests for assistance in preventing work-related injury and illness that is directed at identified problem areas,19 for example trench cave-ins, falls through skylights, and work-related homicides.

Sources of injury morbidity data In contrast to fatal injury events that represent the most severe work-related injuries, non-fatal work-related injuries may range from a minor injury that results only in a brief alteration of usual duties to one that results in permanent disability. Though less severe, non-fatal injuries occur over 1000 times more frequently than fatal injuries. The actual number of non-fatal work-related injuries is difficult to determine. In addition to the problems of designating work-relatedness, the mechanisms for surveillance may vary from state to state, with different definitions and rules for reporting. However, several sources of data are available to characterize the extent of the problem. Some of these data sources are amenable to etiologic research, while others are primarily useful as surveillance tools. In the United States, national data are primarily available from the Bureau of Labor Statistics (BLS). The BLS compiles

data on business and labor activities in the United States, including information on occupational injuries and illnesses occurring in the labor force. The primary source of data is the annual Survey of Occupational Injuries and Illnesses that is based on employer reports from approximately 180,000 private industry establishments. Additional information on mining and transportation activities is provided by the Department of Labor’s Mine Safety and Health Administration (MSHA) and the Department of Transportation respectively. Selected establishments submit summarized data from the OSHA 200 forms to the BLS, indicating the number of occupational injuries and illnesses that occurred each year. Employers are required to report all injuries that are fatal and injuries that require medical care beyond first aid, result in loss of consciousness, lost work days or restriction of work, or transfer to another position. The average number of workers and total hours worked that year are collected, enabling rates to be calculated. Each establishment is classified by Standardized Industrial Code (SIC) according to its major enterprise. Annual survey data can be used to estimate the total number and rates of morbidity, mortality, and lost workdays due to occupational injury for each industry. Although this procedure identifies high-risk industries, the data lack information concerning the cause and type of the injury and specific high-risk groups identified by job classification or demographic characteristics. At the state level, worker compensation claims data provide a reasonable estimate of the burden of non-fatal work-related injuries. The worker compensation insurance laws are state specific, with each state having a mechanism to characterize injury burden, therefore the data are not necessarily comparable between states. Most states require the reporting of fatal injuries and compensable injuries that result in time-loss, generally involving 3 days lost. In some states, such as Washington, the state is the sole-source provider of worker compensation insurance; thus all establishments that do not self-insure or are not covered by a federal program have claims processed by the state. In a majority of states the worker compensation insurance is issued by a third party insurer; thus collecting more than summary statistics is problematic. Nevertheless, worker compensation insurance claims can be a useful tool in monitoring non-fatal work-related injuries and identifying priority problem areas to target.

Other data sources Several other sources of injury data that are not specific for occupational injuries are available and can be useful in identifying work-related injuries. The National Electronic Injury Surveillance System (NEISS), co-ordinated by the Consumer Product Safety Commission, surveys a sample of hospital emergency rooms throughout the country.20 The NCHS collects fatality data, from which information on occupational fatalities can be extrapolated. The NCHS also collects injury morbidity data through the National Health Interview survey. The NSC does not collect original data on occupational injuries; however, it compiles data

Characteristics of Occupational Injuries 829 firearms. Men account for 92% of the occupational fatalities with an annual rate that is 10-fold that of women (6.9 vs 0.69 per 100,000 persons employed); however, homicides account for 31% of fatalities in women, but only 9% in men (Fig. 31.2). The rates of occupational fatalities are similar for younger age groups, but begin to dramatically increase after age 65 (Fig. 31.3). Whether this increase is due to a higher rate of more severe injuries in this age group or a greater case-fatality rate associated with the elderly is not clear. In most traumatic injuries the elderly do have a substantially higher case-fatality rate than younger individuals.

from other sources listed above and provides estimates of the occurrence and cost of occupational injuries.

CHARACTERISTICS OF OCCUPATIONAL INJURIES Understanding the general distribution of occupational injuries by type of industry, occupation, and characteristics of the worker is necessary for establishing any injury prevention and control activities. These characteristics comprise the first layer of factors influencing the injury event, using the Haddon approach to injury control, and are distinct with increasing severity of injury.

Industrial division Construction Transportation Agriculture Services Manufacturing Government Retail trade Wholesale trade Mining Finance Selected occupations Truck drivers Farm occupations Sales occupations Construction laborers Other laborers Groundskeepers Police and detectives Electricians Carpenters Pilots and navigators Total

Occupational fatalities While the completeness of fatality surveillance data can be debated, the CFOI data effectively summarize the major factors related to fatal work-related injuries. In 2002 there were 5524 identified fatalities in the United States. The industry-wide rate was four fatalities for every 100,000 employed persons; however, this varied markedly by industry, occupation, and other characteristics. The highest rates were seen in the agriculture, mining, construction, and transportation industries (Table 31.3). The rates for agriculture and mining were more than five times the industrywide average. Similar patterns have been identified by occupation type where construction laborers, farm occupations, and truck drivers had the highest rates of fatalities. The most frequent types of fatal injury incidents were transportation related, falls, and homicides. Fully 43% (2381) of the 2002 fatalities were transportationrelated incidents, of which 1372 occurred on highways. Accordingly, truck driving was an occupation with one of the highest fatal injury rates (25/100,000). Fatal fall injuries were predominantly falls from one level to another level, and 77% of the homicides were related to

%

Rate*

1121 910 789 680 563 554 487 205 121 87

20.3 16.5 14.3 12.3 10.2 10.0 8.8 3.7 2.2 1.6

12.2 11.3 22.7 1.7 3.1 2.7 2.1 4 23.5 1

808 519 347 302 181 146 140 116 108 90 5524

14.6 9.4 6.3 5.5 3.3 2.6 2.5 2.1 2.0 1.6 100

25.0 28.0 2.1 27.7 14.2 15.0 11.6 13.5 6.9 69.8 4.0

*Fatalities per 100,000 persons employed.

Table 31.3. Number and rate of fatal occupational injuries in the United States in 2002 as reported by the Bureau of Labor Statistics Census of Fatal Occupational Injuries

Figure 31.2: Gender distribution of fatal occupational injuries in the United States, 2002. (From the Bureau of Labor Statistics Census of Fatal Occupational Injuries.)

Injury event Fires and explosions Exposures to harmful substances Falls Contact with objects Women Men

Assaults and violence Transportation 0

n

10

20

30

40

Percent of total fatalities

50

60

830 Occupational Injuries The hazards identified in surveillance of fatal and nonfatal injuries will vary, even when the non-fatal injuries are severe. A comparison of fatal injuries and severe non-fatal injuries in Washington State revealed different patterns. For instance, roofers are at high risk of fatal fall injuries, based on national data, but burns are a significant cause of non-fatal injury. Similarly, in orchard workers fatalities were primarily due to motor-vehicle incidents, but the majority of hospitalized injuries were due to falls.21

14.0

Fatalities/100,000

12.0 10.0 8.0 6.0 4.0 2.0 0 16–19 20–24 25–34 35–44 45–54 55–64 Age group

65+

Figure 31.3: Rate of fatal occupational injuries by age group in the United States, 2002. (From the Bureau of Labor Statistics Census of Fatal Occupational Injuries.)

Identifying the industries and occupations with the highest occupational fatality rates provides some direction for injury prevention programs; however, it is necessary to identify the specific mechanism of injury to prevent further fatalities. The mechanism of injury will vary within an industry and occupational group. Among motor vehicle operators, highway fatalities account for 69% of fatal occupational injuries in truck drivers, while homicides account for only 2% of the fatalities. By contrast, 59% of fatalities among taxicab drivers are due to violence, whereas 37% of the deaths are roadway related. A clear difference between taxicab drivers and truck drivers is the degree to which they interact with the public. Similar differences are observed in the construction trades where the rate of fatal injuries among construction laborers is four times higher than for carpenters (27.7 vs 6.9/100,000); however, carpenters are much more likely to die from falls. These differences reflect the types of hazards inherent in the specific job activities.

Non-fatal occupational injuries Unlike fatalities, non-fatal injuries lack a clear definition; in addition to determining the actual work-relatedness of an injury, it is important to understand which injuries are reportable. The BLS definition of a traumatic injury is an instantaneous event resulting in medical treatment beyond first aid or loss of consciousness, lost workdays, restriction of work or being transferred to another job. According to the BLS Annual Survey, the construction industry has the highest reported rates of injury, followed by agriculture (including forestry and fishing), manufacturing, and mining.2 This ranking is essentially the same for the rate of lost workdays, a surrogate measure of severity. Mining, however, which had the fifth highest rate of injury, had the greatest number of lost workdays. Comparing the rankings of fatal and non-fatal injuries along with lost workdays identifies the high-risk industrial groups. Mining, transportation, construction, agriculture, and manufacturing are high-risk industries compared with services, retail and wholesale trade, and finance and real estate, which have relatively low injury rates.

TRENDS IN INJURY RATES Rates of occupational injuries have not been stable over time, and trends vary by industry and occupation. The incidence of disabling injuries in manufacturing industries in the United States declined from the 1920s until approximately 1958, after which it appeared to rise. The incidence of injuries in the construction industry decreased from the 1940s until the mid-1950s; from approximately 1975, there has been a marked upward trend. Similar trends have been seen for injuries in wholesale and retail trade.22 Conversely, the rate of fatal injuries in the mining industry dropped drastically over the last century as regulatory measures improved the safety records of mines.23 A general downward trend has been observed for both fatal and non-fatal injuries in the last two decades; however, the rates vary by industry and mechanism of injury. Fatality data indicate that the rate of motor vehiclerelated deaths dropped in the transportation industry, but there was little improvement in the public administration, construction, and agriculture sectors. Within these long-term trends, there is cyclic fluctuation that has been observed to be associated with changes in the strength of the economy. For example, a 10% increase in the rate of new hires is associated with a 10.4% increase in injury rates, providing evidence that new workers have higher injury rates than their more experienced counterparts. Increases in productivity, measured as output per hour, are also associated with increases in injury rates.24 The extent to which this increase is due to laxity in enforcement of safety practices in order to meet increased demand is uncertain. Changes in occupational injury rates can serve as a measure of the effectiveness of prevention programs as well as pointing to areas in need of more effective interventions. However, the value of such data depends heavily on their quality and validity, as described earlier.

OCCUPATIONAL INJURY TYPES AND SETTINGS Occupational injuries are characterized largely by the type of occupation or the type of injury. In either case there will be considerable heterogeneity within subsets of the population under study. In this section the characteristics of intentional injuries and agriculture-related injuries will be presented to demonstrate the scope of issues associated with specific injuries and their respective control interventioning.

Occupational Injury Types and Settings 831

Assaults and homicides The classic perceptions of work-related injuries and fatalities evoke images of catastrophic events related to machinery, falls or explosions. The role of intentional injury in the workplace has been somewhat overlooked as a significant cause of morbidity and mortality. Homicide ranks as one of the leading causes of fatal occupational injury in the United States, accounting for the deaths of an estimated 609 workers in 2002 or 11% of all work-related fatalities.1 While the rate of work-related homicide is two to three times greater for men than women, homicides are the leading cause of work-related death in women, accounting for approximately one-third of occupational deaths. Homicide is the second leading cause of death for workers aged 25–44, behind highway-related deaths, yet the rate of work-related homicide increases in workers over age 55. The majority of the homicides are committed by firearms. Assault-related injuries and fatalities are, to a large degree, specific to occupation. Surveillance data from 2002 indicate that 50–80% of the work-related fatalities in the technical, sales, and administrative support jobs are homicides. A variety of surveillance systems have identified occupations at higher risk of homicide and assault; a common theme is the close interaction between the worker and the general public, often in isolation, including taxicab drivers, gas station and convenience store clerks, liquor store clerks, police officers, and security guards.25,26 The occurrence of these fatal events is often associated with another crime, primarily robbery, and often in establishments with a prior history of such events.27,28 In 1982 Baker and colleagues called for uniform standards to protect workers from homicides in the workplace;29 however, none have been enacted to date. Possible prevention strategies include making cash registers visible from the street; the use of a drop safe at night for bills larger than $5; bulletproof barriers between driver and passenger in taxis; enclosed cashier booths in gasoline stations; protective vests for law enforcement personnel. Other strategies include the provision of training programs in conflict resolution and non-violent response; these have been implemented in some retail settings. Non-fatal work-related assault is also a significant morbidity concern. As with homicides, the non-fatal workrelated injuries occur disproportionately in occupations that have routine contact with the general public. The risks are higher for healthcare workers, social service workers, public transportation workers, and safety and corrections personnel.27,30–33 Sexual assault in the workplace constitutes an important proportion of rape cases, but has received little attention in the occupational health and safety literature. The National Crime Victimization Survey has been used to characterize the victims of work-related physical assaults and robbery,34 while occupational information is not collected by the FBI, nor reported in Uniform Crime Reports. Violence and assault within the healthcare occupations have been evaluated for potential risk factors. Certain jobs that involve contact with high-risk patients, particularly

patients with mental health conditions, are at highest risk.35–37 Low staff-to-patient ratios may increase risk. The administrative recognition of risk as an inherent problem related to the job activities provides an opportunity to target intervention, including assault prevention training or presence of security guards. The success of interventions in workplace violence, however, has not been well documented, with several studies proving inconclusive.38,39

Agricultural injuries Agriculture is one of the most dangerous industries in the United States. The annual death rate is 22.7 deaths per 100,000 workers compared with 4.0 deaths per 100,000 for the general private sector workforce.1 Over the course of a typical 40–50-year career, farmers accrue a substantial lifetime risk of occupationally related mortality. Data on agricultural injuries include injuries to individuals involved with agricultural production, agricultural services, forestry, and fishing. Specifically, for agricultural crop production, livestock production and agricultural services, the rates of fatal injuries in 2002 were approximately 37.0, 12.5, and 13.8 per 100,000 workers, respectively.1 The mortality for farmers 65 years of age and older appears to be two to three times that of younger workers. Although agricultural machinery including tractors accounts for around 20% of minor farm injuries, they are responsible for more than half of all severe injuries that result in permanent disability, and about 40% or more of the fatalities. In contrast, injuries from animals are very rarely fatal but account for a sizeable proportion of mild injuries. Tractor-related injuries are the most common type of fatal agricultural trauma, accounting for 30–50% of all fatal injuries on the farm. Approximately one-half to threefourths of these fatal incidents involve tractor rollovers. In the NTOF database, 69% of agricultural machinery deaths involve tractors, of which 52% were rollovers.40,41 One way to prevent these deaths is by using roll bars or crush-proof cabs, commonly referred to as rollover protection structures (ROPS), which were developed as early as 1959. In 1976, OSHA imposed standards requiring ROPS on all tractors used by employees, but this standard does not apply to the many tractor operators who are self-employed or farm family members. Furthermore, special legislation limited OSHA to inspect only farms employing more than 10 non-family members, effectively eliminating 89% of farms in the United States. A population-based study of tractor-related injuries in Minnesota indicated that only 42% of tractors were equipped with a ROPS,42 which is representative of use in the US generally.43 In contrast, mandatory standards for ROPS have been markedly effective in nearly eliminating tractor fatalities in Sweden. Since 1959, all new tractors in Sweden have had to be equipped with ROPS and since 1983 all workers, including employers and family members, had to be protected with ROPS on tractors.44 A comparison of the number of fatalities, normalized to the number of tractors, suggests a 24-fold greater risk of tractor rollover fatalities in the US compared to Sweden.43

832 Occupational Injuries Other important sources of injury related to agricultural machinery include power take-off (PTO) devices and augers.45 PTOs are rapidly rotating shafts on the tractor to power other machinery and account for 10–15% of fatal injuries and an even higher percentage of severe, disabling, but non-fatal traumatic injury. The injuries consist of fractures, severe soft tissue trauma, dislocations, nerve injuries and amputations. Although shielding is provided by the manufacturer, the shielding is often removed to make access to the PTO devices easier. Typically, a worker in proximity to an unshielded PTO shaft has an article of clothing pulled into the machine. As with ROPS, OSHA shielding regulations for PTOs apply only to inspected farms, thus most family farms and those with fewer than 11 employees are not covered. As with ROPS, reliance on voluntary compliance by manufacturers and farmers appears to be ineffective. Grain augers are a source of serious and potentially devastating injuries. Invented by Archimedes to move water up a hill, modern augers and screw conveyors are widely used to transport wood, grain, meat, and other materials on the farm. The screw rotation of augers is 800 to 900 RPM with a pitch of 1:1. An auger with a 6-inch diameter moves material up the tube at 400 feet per minute or 7 feet per second. A finger that is caught in a tube will thus ascend a number of feet before the individual has time to react. The most common injuries are amputations, usually of fingers. Many of the injuries are at multiple levels because the spiral blades strike the limb several times as it is drawn up the tube. One half of those injured suffer significant permanent impairment. The distribution of both fatal and non-fatal injuries is largely determined by the type of agriculture. In the Midwest, farm animals have been associated with up to 40% of the non-fatal farm-related injuries.46 The risk of a farm animal-related injury is obviously related to the amount of contact an individual has with the animals, but is also influenced by the type of work done. For example, injuries among dairy cattle handlers will be affected by the amount of time doing specific tasks such as milking and hoof trimming.47 Among California migrant workers, the tasks associated with injury included manual harvest, operating machinery, and cannery work.48

CHILDREN Children make up a significant portion of the agricultural workforce and hence are exposed at an early age and often tragically to its hazards. Children may work for their parents; many more work alongside their migrant families as seasonal laborers. Like adults, children suffer injuries when engaged in agricultural work and the injuries are often severe. Children employed in agriculture production in California had elevated risks for death by machinery (OR = 81), animals (OR = 10), electricity (OR = 5), and nontraffic motor vehicles (OR = 3).49 Minors under the age of 15 employed on farms in Washington State were overrepresented in the number of claims filed by all minors;50 furthermore, 26% of claims filed by minors employed in

agriculture were for serious injuries, compared with 16% for all other occupations. The rates of injury are higher for older adolescents, with males age 15–17 having the highest rates of injury.51 While children employed in agriculture are clearly at substantial risk of injury, the extent of this problem has not been well characterized. Estimates vary on the number of fatal injuries that occur among children and adolescents on farms each year, but range from 100 to approximately 300 children, with another 23,000 non-fatal injuries. Farm machinery accounts for onethird of deaths and nearly one-half of deaths of children under the age of 10 years. Firearms account for one in 10 deaths of children on farms, whereas farm animals are responsible for less than 2% of farm deaths among children, despite the relatively frequent exposure to this hazard. There are very different attitudes and legislation for child farm labor compared with other employment of children in the workplace. Under the Fair Labor Standards Act, children as young as 12 and 13 years of age may be employed on any farm with the consent of their parents. These federal regulations are in marked contrast to those applying to non-agricultural occupations. For most other occupations, the basic minimum age is 16 years. Adolescents are especially prohibited from employment in manufacturing industries, which have a lower rate of fatalities than agriculture.

SUMMARY In summary, occupational injuries are a complex problem requiring an array of preventive strategies. The first steps necessary for attacking the problem are (1) a change in orientation to one that views these injuries as events that are intrinsically preventable and (2) the use of a scientific approach to development of interventions. An important requirement for the success of this approach will be the development of adequate data systems in order to better understand the magnitude of the problem among specific populations, industries, and jobs at risk.

References 1. United States Department of Labor Bureau of Labor Statistics. National Census of Fatal Occupational Injuries in 2002. Washington, DC: Bureau of Labor Statistics, 2003. 2. United States Department of Labor Bureau of Labor Statistics. Workplace injuries and illnesses in 2001. Washington, DC: Bureau of Labor Statistics, 2002. 3. National Safety Council. Injury facts. Itasca, IL: National Safety Council, 2002. 4. Rivara FP. Introduction: the scientific basis for injury control. Epidemiol Rev 2003; 25:20–3. 5. Haddon W. Advances in the epidemiology of injuries as a basis for public policy. Public Health Rep 1980; 95:411–21. 6. Bonnie RJ, Fulco CE, Liverman CT, eds. Committee on Injury Prevention and Control, Institute of Medicine. Reducing the burden in injury: advancing prevention and treatment. Washington, DC: National Academy Press, 1999. 7. Association for the Advancement of Automotive Medicine Committee on Injury Scaling. The Abbreviated Injury Scale – 1990. Des Plains, IL: Association for the Advancement of Automotive Medicine, 1990.

Summary 833 8. Baker SP, O’Neill B, Haddon W, Long WB. The severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 1974; 14:187–96. 9. Baker SP, O’Neill B. The injury severity score: an update. J Trauma 1976; 16:882–5. 10. Osler T, Baker SP, Long W. A modification to the Injury Severity Score that both improves accuracy and simplifies scoring. J Trauma 1997; 43:922–6. 11. Copes WS, Champion HR, Sacco WJ, et al. Progress in characterizing anatomic injury. J Trauma 1990; 3:1200–7. 12. Champion HR, Copes WS, Sacco WJ, et al. Improved predictions from A Severity Characterization of Trauma (ASCOT) over Trauma and Injury Severity Score (TRISS): results of an independent evaluation. J Trauma 1996; 40:40–49. 13. Champion HR, Copes WS, Sacco WJ. The major trauma outcome study: establishing norms for trauma care. J Trauma 1994; 36:499–503. 14. Champion HR, Sacco WJ, Copes WS, Gann DS, Gennarelli TA, Flanagan ME. A revision of the trauma score. J Trauma 1989; 29:623–9. 15. Horan JM, Mallonee S. Injury surveillance. Epidemiol Rev 2003; 25:24–42. 16. Pollack ES, Keimig DG, eds. Counting injuries and illnesses in the workplace: proposals for a better system. Washington, DC: National Academy Press, 1987. 17. Stout N, Bell C. Effectiveness of source documents for identifying fatal occupational injuries: a synthesis of studies. Am J Public Health 1991; 81:725–8. 18. NIOSH. Fatality Assessment and Control Evaluation (FACE) Program, 2003. Atlanta, GA: National Institute for Occupational Safety and Health, 2003. 19. NIOSH. NIOSH alerts on traumatic occupational injury topics. Atlanta, GA: National Institute for Occupational Safety and Health, 2003. 20. US Consumer Product Safety Commission. National Electronic Injury Surveillance System. Bethesda, MD: US Consumer Product Safety Commission, 2003. 21. Alexander BH, Franklin GM, Fulton-Kehoe D. Comparison of fatal and severe nonfatal traumatic work-related injuries in Washington state. Am J Ind Med 1999; 36:317–25. 22. Robinson JC. The rising long term trend in occupational injury rates. Am J Public Health 1988; 78:276–81. 23. Stout NA, Linn HI. Occupational injury prevention research: progress and priorities. Injury Prev 2002; 8(Suppl 4):IV9–14. 24. Robinson JC, Schor GM. Business-cycle influences on workrelated disability in construction manufacturing. Milbank Quart 1989; 67:92–113. 25. Kraus JF, Blander B, McArthur DL. Incidence, risk factors and prevention strategies for work-related assault injuries: a review of what is known, what needs to be known, and countermeasures for intervention. Annu Rev Public Health 1995; 16:355–79. 26. Peek-Asa C, Erickson R, Kraus JF. Traumatic occupational fatalities in the retail industry, United States 1992–1996. Am J Ind Med 1999; 35:186–91. 27. Erickson RJ. Retail employees as a group at risk for violence. Occup Med 1996; 11:269–76. 28. Schaffer KB, Casteel C, Kraus JF. A case-site/control-site study of workplace violent injury. J Occup Environ Med 2002; 44:1018–26. 29. Baker SP, Samkoff S, Fisher RS, Van Buren CB. Fatal occupational injuries. JAMA 1982; 248:692–7.

30. Kraus JF, McArthur DL. Epidemiology of violent injury in the workplace. Occup Med 1996; 11:201–17. 31. Nelson NA, Kaufman JD. Fatal and nonfatal injuries related to violence in Washington workplaces, 1992. Am J Ind Med 1996; 30:438–46. 32. Peek-Asa C, Howard J, Vargas L, Kraus JF. Incidence of nonfatal workplace assault injuries determined from employer’s reports in California. J Occup Environ Med 1997; 39:44-50. 33. McGovern P, Kochevar L, Lohman W, et al. The cost of workrelated physical assaults in Minnesota. Health Serv Res 2000; 35:663–86. 34. Klein PJ, Gerberich SG, Gibson RW, et al. Risk factors for work-related violent victimization. Epidemiology 1997; 8:408–13. 35. Lee SS, Gerberich SG, Waller LA, Anderson A, McGovern P. Work-related assault injuries among nurses. Epidemiology 1999; 10:685–91. 36. Bensley L, Nelson N, Kaufman J, Silverstein B, Kalat J, Shields JW. Injuries due to assaults on psychiatric hospital employees in Washington State. Am J Ind Med 1997; 31:92–9. 37. Simonowitz JA. Healthcare workers and workplace violence. Occup Med 1996; 11:277–91. 38. Runyan CW, Zakocs RC, Zwerling C. Administrative and behavioral interventions for workplace violence prevention. Am J Prev Med 2000; 18:116–27. 39. Casteel C, Peek-Asa C. Effectiveness of crime prevention through environmental design (CPTED) in reducing robberies. Am J Prev Med 2000; 18:99–115. 40. Myers JR, Hard DL. Work-related fatalities in the agricultural production and services sectors, 1980–1989. Am J Ind Med 1995; 27:51–63. 41. Myers JR. National surveillance of occupational fatalities in agriculture. Am J Ind Med 1990; 18:163–8. 42. Lee TY, Gerberich SG, Gibson RW, Carr WP, Shutske J, Renier CM. A population-based study of tractor-related injuries: Regional Rural Injury Study-I (RRIS-I). J Occup Environ Med 1996; 38:782–93. 43. Reynolds SJ, Groves W. Effectiveness of roll-over protective structures in reducing farm tractor fatalities. Am J Prev Med 2000; 18:63–9. 44. Springfeldt B, Thorsen J, Lee BC. Sweden’s thirty-year experience with tractor rollovers. J Agric Health Safety 1998; 4:173–80. 45. Gerberich SG, Gibson RW, French LR, et al. Machinery-related injuries: regional rural injury study – I (RRIS – I). Accident Analysis Prev 1998; 30:793–804. 46. Gerberich SG, Gibson RW, French LR, et al. Injuries among children and youth in farm households: regional rural injury study – I. Injury Prev 2001; 7:117–22. 47. Boyle D, Gerberich SG, Gibson RW, et al. Injury from dairy cattle activities. Epidemiology 1997; 8:37–41. 48. McCurdy SA, Samuels SJ, Carroll DJ, Beaumont JJ, Morrin LA. Agricultural injury in California migrant Hispanic farm workers. Am J Ind Med 2003; 44:225–35. 49. Schenker MB, Lopez R, Wintemute G. Farm-related fatalities among children in California, 1980–1989. Am J Public Health 1995; 85:89–92. 50. Heyer NJ, Franklin GM. Work-related traumatic brain injury in Washington State, 1988–1990. Am J Public Health 1994; 84:1106–9. 51. Rivara FP. Fatal and non-fatal farm injuries to children and adolescents in the United States, 1990–3. Injury Prev 1997; 3:190–4.

Chapter 30 Cancer 30.1 Malignancies of the Respiratory Tract and Pleura Mark B Russi

LUNG CANCER Epidemiology Lung cancer, a virtually unknown disease 100 years ago, now kills more American men and women than any other tumor. Although cigarette smoking represents the most important risk factor for lung cancer, occupational and environmental exposures, both singly and in conjunction with tobacco, are responsible for a substantial burden of preventable disease. This chapter addresses current estimates of the proportion of lung cancer attributable to occupation, provides a framework for the occupational medicine practitioner to make etiologic judgments, and discusses specific exposures associated with lung cancer. Estimates during the 1960s and 1970s regarding the percentage of lung cancer attributable to occupation varied from 1% to greater than 60%. More recent studies suggest a narrower range of risk. Doll and Peto in the early 1980s proposed that 15% of male lung cancer cases and 5% of female lung cancer cases in the US resulted from occupational exposures.1 The combined data from five American case–control studies spanning a wide range of industrial activities suggested that between 3% and 17% of lung cancers were due to well-recognized occupational carcinogens.2 Another multicenter case–control study suggested that 9% of lung cancers were due to work in occupations with excess lung cancer risk.3 A recent review has suggested that occupational exposures are responsible for 9% of lung cancer among men and 2% among women.4 Much of the burden of lung cancer morbidity among working populations has been attributed to heavy exposures in the past. There may be numerous complex interactions between the risk factors for lung cancer. Studies illustrating the combined effect of cigarette smoking with asbestos, arsenic, and radon exposure provide illustrative examples. When interactions exist between occupational or environmental exposure and any other risk factor, such as cigarette smoking, the percentage of lung cancer associated with the exposure reveals only what percentage of lung cancer could be eliminated through elimination of the exposure. The percentage itself does not reveal whether another exposure or several other exposures must be present to induce disease, only what percentage of disease would be eliminated if the exposure did not exist. Assigning the proportion of lung cancer attributable to specific occupational or environmental exposures has been challenging due to the difficulty of both estimating accurate exposure levels in the workplace and extending the results

of studies in highly exposed groups to the population at large. Estimates of risk may vary considerably due to wide ranges of exposure among different occupational groups to agents of importance, such as asbestos. Further, reconstructing an accurate exposure history for cohorts of workers over several decades is difficult. Once data exist for a highly exposed population, extension to the general population requires accurate exposure estimates for all occupations, as well as domestic and environmental exposure settings. An alternative method extends job category-based risk estimates for specific occupations to the total number of workers employed in such occupations based on census data. While such studies provide a simpler extrapolation method, they do not consider exposures outside the workplace, and job category-based risk estimates themselves may be subject to a number of biases. Better data regarding interacting risk factors, occupational carcinogen doses, and exposures of the population at large may lead in the future to more precise estimates of the percentage of lung cancer due to occupational or environmental exposures.

Clinical evaluation In individuals exposed to one or more carcinogens, a determination of etiology may be difficult. Although a few occupational lung carcinogens predispose more persons to one histologic type or location than others, lung cancers caused by occupational exposures do not differ from cancer induced by other causes in a way that allows pathologic, anatomic, or clinical distinction. From a public health standpoint, recognition of the cause of an individual worker’s lung cancer may serve to identify excess risk at the group level, an indication for action to reduce exposure. Although individual etiologic diagnosis does not have an impact on selection of therapy, the occupational and environmental medicine physician may be called on to decide whether or not an individual’s cancer is work related, often for purposes of workers’ compensation. A legal solution to this issue, as seen in the case of specific exposures in certain countries and specific occupations in the US, has been to establish a legal presumption that the cancer was caused by workplace exposures if an agreed-upon degree of exposure in the specified settings can be proved. However, such approaches currently are limited, and include asbestos-associated cancers, and some radiation-associated lung cancers in certain settings. Most individual cases arise in industries or exposure groups not formally recognized to have presumptive exposure under the workers’ compensation laws.

728 Malignancies of the Respiratory Tract and Pleura An alternative approach bases individualized etiologic diagnosis of occupational lung cancer on the probability that a given cancer is occupational. This does not imply medical certainty but only that the exposure most likely caused or contributed to the development of the cancer. In the case of lung cancer due to uranium mining, various formulas have been developed to rate factors influencing the assessment of probability, including dose (average amount of exposure multiplied by years of exposure), amount of cigarette smoking, age, latency (years since first exposure), and histologic type of lung cancer. Similar formulas have been attempted for asbestos-related cancer. Typically, however, the clinician must use quantitative or semiquantitative reasoning without a simple or legally sanctioned formula. The first step is to determine whether or not the patient was exposed to a factor causally related to lung cancer or that he or she worked in an industry with known increased risk. A complete occupational and environmental history, including a thorough review of all potential workplace carcinogens, is a prerequisite. Although workers may have histories of lifelong employment in well-established risky industries such as shipbuilding, insulating, or smelting, more often than not, a complicated history of numerous potential work exposures must be pieced together carefully. Assessment of dose is the next step. Most epidemiologic studies of occupational lung cancer have used surrogates for dose, such as job description, or attempted rough reconstructions based on worker interviews. Quantitative information is rarely available, and dosages are often listed as only low, medium, or high. In assessing an individual’s exposure, one may access certain information by interview and review of industrial hygiene data, but historical estimates of exposure for the industry or occupation in question must often suffice. Often, dose cannot be determined because the suspect exposure is not fully characterized. Work in the vicinity of coke ovens provides an example in which the specific carcinogen is unknown but the association between occupation and lung cancer is strong. Dose in such cases must be estimated by the individual’s proximity to the coke oven and length of employment. Latency must also be taken into consideration because many cancers do not appear until 30 to 40 years after exposure has occurred. Latency is dependent on whether a substance acts early in the carcinogenic process or at a later stage. A simplified initiator and promoter model of carcinogenesis may apply to cigarette smoking and asbestos exposure, but not to some other etiologies, including radiation. Asbestos probably acts as a tumor promoter, albeit with a long latency. Arsenic may be an example of a lung carcinogen that acts late in the transformation process. Finally, a thorough smoking history – including passive smoke exposure – must be gathered from any patient with lung cancer. The individual’s smoking history deserves special attention in cases in which the relative risk associated with the occupational or environmental exposure of interest is small. Domestic radon exposure should also be considered, though most individuals with radon-induced lung cancer will have been exposed to relatively average dose levels.

Substance Asbestos Arsenic Chloromethyl ethers Chromium Crystalline silica Mustard gas Nickel Polyaromatic hydrocarbons Radon Environmental tobacco smoke

Major exposure settings Insulation workers, shipyard workers Smelting of copper, zinc, lead; arsenical pesticide processing Chemical production workers Chromate production, pigment manufacture, electroplating Mining, sandblasting, stonework, foundry work Mustard gas production workers Nickel mining, refining, plating Coke oven workers, rubber workers, aluminum reduction workers, roofers Uranium mining, hard rock mining, widespread domestic exposure Environments with active smokers

Table 30.1.1 Known lung carcinogens

Substance Beryllium Cadmium Man-made vitreous fibers (MMVF) Ambient air pollution

Exposure settings Beryllium production/processing Smelting, battery production Insulating, rock slag, wool production Indoor and outdoor environments

Table 30.1.2 Probable lung carcinogens

In the end, rational judgment as to the work-relatedness of a case of lung cancer demands careful consideration of a patient’s exposure history and a working knowledge of the epidemiology of occupational or environmental carcinogens. The following section addresses known and probable agents, summarized in Tables 30.1.1 and 30.1.2, respectively.

KNOWN LUNG CARCINOGENS Asbestos (see Chapter 46) The earliest recognition of asbestos as a possible lung carcinogen occurred in a case report by Kenneth Lynch and Atmar Smith in 1935.5 Subsequent work by Merewether in 19496 and Gloyne in 19517 revealed an increased frequency of lung cancer at necropsy in patients with asbestosis when compared with patients with silicosis. A proven or highly probable causal relationship between asbestos and lung cancer was agreed upon in 1953 at the International Symposium on the Epidemiology of Lung Cancer (Council of the International Organizations of Medical Sciences). In 1955, a careful study by Sir Richard Doll based on necropsy data from 105 asbestos textile workers revealed a 10-fold increased rate of lung cancer in workers employed over 20 years in the trade, firmly establishing asbestos as a major lung carcinogen.8 Further support was lent by Selikoff’s study of 632 insulation workers in 1964, which revealed a sevenfold increase in the rate of lung and pleural tumors over that of the general population.9 Numerous studies have since confirmed the carcinogenic potential of asbestos.10–14 As epidemiologic and basic scientific research of asbestos exposure has progressed, a number of secondary

Known Lung Carcinogens 729 issues have arisen.15 The interaction between cigarette smoking and asbestos exposure in producing lung cancer remains to be precisely quantified. The question of whether lung cancer arises from asbestos exposure itself or merely from lung fibrosis induced by asbestos is also a topic of some debate. Markedly different dose–response relationships in different trades and industries are still unexplained, and controversy exists regarding the carcinogenic potential of various asbestos fiber types. Finally, the concept of a threshold effect and its implications for the risk of very low-level asbestos exposure remain unresolved. Although various models have been proposed for the interaction of cigarette smoking and asbestos exposure, the bulk of epidemiologic evidence implicates asbestos as a carcinogen by itself, whose effect is augmented by cigarette smoking. A synergistic relationship between the two carcinogens is commonly accepted, and a recent review of 23 studies addressing smoking and asbestos exposure lends support to a multiplicative interaction.16 Asbestos exposure appears to multiply lung cancer risk by a similar factor in smokers and non-smokers. Because smoking impedes clearance of inhaled fibers, smokers may suffer increased penetration of fibers into airway walls.17,18 Carcinogenic polycyclic aromatic hydrocarbons (PAHs) in cigarette smoke may also adsorb onto asbestos fibers, enhancing their transport into cells.19 The question of whether asbestos itself or asbestosis is the true risk factor for lung cancer has implications for present surveillance practices as well as compensation issues. If the development of asbestosis is not a necessary intervening step in carcinogenesis, the current practice of close radiologic monitoring of asbestotics also may be appropriate for significantly exposed persons without asbestosis. Clearly, most lung cancers arising in persons with significant asbestos exposure occur in the setting of radiologically documentable asbestosis, and cohort studies of asbestotics have revealed some of the highest rates of lung cancer mortality.20–22 Correlations between asbestosis and lung cancer also have been shown to be stronger than those between cumulative asbestos exposure and lung cancer.23 The finding in some studies that asbestos-related cancers frequent the lower lobes where fibrosis is maximal, as well as the occurrence in some series of a disproportionate number of adenocarcinomas, a cancer that may arise in areas of scar formation, is cited as evidence that fibrosis is a prerequisite for asbestos-related lung cancer. On the other hand, lung cancers do appear in heavily exposed individuals without clinical, radiologic, or pathologic evidence of asbestosis;24–26 many studies reveal no predominance of adenocarcinoma among asbestos-related lung cancer cases; some studies reveal no lower lobe predominance of asbestos-related tumors;27,28 and in-vitro experimental models have demonstrated a tumor-promoting effect of asbestos.15 Because a clear understanding of the pathophysiologic mechanisms underlying both asbestosis and cancer remains elusive at present, resolution of these issues likely will come only with continued study of hostrelated and exogenous mechanisms of disease, as well as epidemiologic investigation of heavily exposed subgroups without evidence of asbestosis.

Cohort mortality studies conducted to date have revealed that the dose response for asbestos and lung cancer may be 50-fold higher in occupations involving insulation products and chrysotile textiles than it is in the lower-risk occupations of chrysotile mining and friction products work. Although in some cases the discrepancy has been attributed to inadequate historical dose estimates, many claim that examining the fiber types and sizes most commonly used in these industries may yield a more complete explanation of the phenomenon. Asbestos minerals are divided into two classes: (1) the amphiboles, including amosite, crocidolite, anthophylite, actinolite, and tremolite; and (2) the serpentine class, of which the only member is chrysotile. Ninety percent of all asbestos used in the US is chrysotile. Although some of the lowest rates of lung cancer have been reported in miners, cement manufacturers, and friction products workers, whose sole exposures are to chrysotile, some of the highest rates have been found in the asbestos textile industry, where chrysotile is used exclusively. And although autopsy studies have revealed low levels of chrysotile asbestos in heavily exposed persons, suggesting greater clearance of chrysotile from the lung relative to other fibers and possibly reduced overall toxicity, animal studies of asbestos-related lung cancer reveal chrysotile to be a potent carcinogen.29 One explanation for the paradoxically high rates of lung cancer in chrysotile asbestos textile workers is that lowlevel contamination of the textile with other forms of asbestos such as tremolite is responsible. Although tremolite fibers are seen at autopsy in the lungs of chrysotile workers, this explanation would imply an inordinately high toxicity of tremolite given the relatively low exposure doses. Studies of populations exposed solely to chrysotile also reveal elevated lung cancer risk.30 Another proposed resolution to the controversy derives from fiber length measurements. It may be that carcinogenicity is more directly a function of exposure to very long asbestos fibers (exceeding 10 to 20 μm) than merely to all fibers greater than 5 μm.31 Because chrysotile textile contains higher proportions of very long fibers than does the chrysotile to which miners and other lower-risk groups are exposed, lung cancer rates may be elevated on that basis alone. As more detailed environmental exposure data, including mineralogic analysis and fiber length distributions, are combined with careful reconstructions of historical exposure settings, much of the present paradox may clarify itself.15 A final issue that has arisen out of asbestos research is the degree to which very low-level exposures pose risk to the general population. The manner in which this issue is resolved carries important implications for the allocation of public health and economic resources for asbestos abatement in public buildings. In favor of a threshold level, below which cancer risk is not elevated above that of the general population, stand the majority of cohort mortality studies, animal studies of low-dose laboratory exposures, negative autopsy-derived asbestos fiber counts in lung cancers drawn from the general population, and the lack of elevated cancer rates among a well-studied population of non-miner residents of Quebec asbestos mining regions with higher than normal background asbestos exposure.32,33

730 Malignancies of the Respiratory Tract and Pluera Arguments against the existence of a safe level of asbestos exposure include the linearity of dose response to asbestos at higher levels in both cohort mortality studies and animal inhalation studies, as well as the results of two cohort mortality studies in which exposure to low levels for brief periods has been associated with increased lung cancer risk.15 Some have suggested that the assumption of linear dose response with no threshold is conservative and appropriate for public and individual health practice. Nonetheless, given the quality of present data, decisions to allocate resources to asbestos abatement in public buildings with negligible potential for exposure should be made with serious consideration of the relative reduction in risk compared with that obtainable through alleviation of other, possibly more pressing, public health hazards. Removal of asbestos already in place also raises the possibility of increased rather than decreased total exposure under conditions of substandard abatement practices.

Arsenic (see Chapter 39.2) The first cases of arsenic-induced lung cancer were reported by Saupe in 1930.34 Subsequent studies have revealed lung cancer risk elevations associated with arsenic exposure in the settings of copper smelting,35–38 tin mining,39 and the manufacture of arsenical pesticides.40 A recent study reported elevation of lung cancer risk among lead smelter workers also exposed to arsenic.41 Arsenic is considered a late-stage promoter of cancer and may act by interfering with DNA repair and by modulating cell growth signaling pathways.42 Arsenic-induced lung cancer was found to have a predilection for the upper lobes in one study. Another study of copper smelter workers suggested a higher percentage of adenocarcinomas when compared with controls. A clear dose–response relationship has emerged from studies of arsenic-exposed workers, and the risk of lung cancer may be increased by as much as ninefold in heavily exposed workers. There appears to be synergism between arsenic exposure and smoking in the induction of lung cancer.43

Chloromethyl ethers Bis-chloromethyl ether (BCME) and chloromethyl methyl ether (CMME) were first suspected as lung carcinogens in 1962, when cases of small-cell carcinoma of the lung were noted in a chemical manufacturing plant in Philadelphia. Subsequent institution of a screening program documented additional cases, and in 1973, a report associating lung cancer with CMME was published by Figueroa and colleagues.44 The chloromethyl ethers are associated primarily with the ion exchange resin industry. They also may form when chloride reacts with formaldehyde under appropriate conditions in textile manufacturing plants. BCME is the more potent of the two carcinogens, and both are highly reactive as alkylating agents in vivo. They are associated primarily with the induction of small-cell lung carcinoma. Numerous studies have documented the association of chloromethyl ethers with lung cancer, including an industry-

wide cohort study sponsored by the National Institute for Occupational Safety and Health (NIOSH).45 The relative risk of lung cancer may be quite elevated, and long-term exposure does not appear to be an absolute prerequisite for disease. Standardized mortality ratios (SMR) from 2.8 to 5.0 have been reported in exposed populations, with relative risks as high as 7.0 to 18.0 within the highest exposure categories.46 Risk appears to peak at less than 20 years following exposure, and several authors have documented a decline in risk following cessation of exposure.

Chromium (see Chapter 39.5) In 1935, two cases of lung cancer that had occurred more than 20 years previously were reported in a German chromium manufacturing facility. A subsequent report by Machle in 1948 detailed respiratory cancers occurring in the chromate-producing industry in the US.47 Chromium has since proved to be a human carcinogen, with increased incidences of lung cancer in the chrome plating, chrome alloy, chromate pigment, chromate production, and ferrochromium industries. Highest risks of lung cancer are in the chromate production industry, and may persist at 20 years following cessation of exposure.48 Risks in the pigment industry are lower and generally associated only with zinc chromate exposure. Although chromium exposure in electroplating facilities is a risk factor for lung cancer,49,50 exposures to other lung carcinogens such as nickel may confound this relationship. Risk is not as well established in the ferrochromium industry, in which study results are conflicting, and concomitant exposures to PAHs and asbestos may play a more important role. Similarly, elevated rates of lung cancer among chromium-exposed welders in some studies may be confounded by nickel and asbestos exposures.51–54 Conflicting results exist regarding the role of chromium as a carcinogen in the leather tanning industry. Several studies suggest that industries with improved controls over chromium exposures are not associated with elevated lung cancer risks, though statistical power to detect marginal risk elevations has been limited by cohort sizes.55–59 Chromium exposure appears to be a risk factor for all lung cancer cell types, and hexavalent chromium is listed by the International Agency for Research on Cancer (IARC) as a known human carcinogen.60 The evidence implicating trivalent chromium as a human carcinogen is less compelling. While some studies suggest elevated risk among trivalent chromium-exposed workers,61 concomitant exposures to hexavalent chromium have made dose reconstructions challenging. A recent study combined historical air sampling data with analysis of settled dust samples to estimate trivalent and hexavalent exposures. Cumulative hexavalent exposure was associated with an increased lung cancer risk, while cumulative trivalent exposure was not.62

Crystalline silica (see Chapter 46) Crystalline silica exposures occur in mining, sandblasting, construction, and foundry work. In 1996, the IARC upgraded its classification of crystalline silica from a prob-

Known Lung Carcinogens 731 able human carcinogen (Group 2A) to a definite human carcinogen (Group 1).63 The strongest evidence implicating crystalline silica as a human lung carcinogen derives from studies of individuals with silicosis. Steenland and Stayner have reported a lung cancer summary relative risk from 19 cohort and case–control studies of silicotics as 2.3 (95% confidence interval [CI], 2.2–2.6).64 Both smoking and non-smoking silicotics have been shown to have elevated lung cancer risks. The same authors reported a lung cancer summary relative risk based on 16 large studies of silica-exposed workers as 1.3 (95% CI, 1.2–1.4).64 More recent studies have largely corroborated the combined relative risks reported by Steenland and Stayner. A recent mortality study of silicotic patients in Sardinia revealed a SMR for lung cancer of 1.37 (95% CI, 0.98–1.91), though the correlation with severity of silicosis based on the International Labour Organization (ILO) category of the chest radiograph was weak.65 A study of 4626 industrial sand workers found a lung cancer SMR of 1.60 (95% CI, 1.31–1.93) and revealed a significant dose response based on average exposure.66 An important recent study pooled 10 cohorts comprising 65,980 workers and 1072 lung cancer deaths and produced quantitative exposure estimates comparable across the cohorts. The study revealed that the log of cumulative silica exposure, with a 15-year lag, was a strong predictor of lung cancer (P = 0.0001).67

Mustard gas The first association between mustard gas (bis [βchloroethyl] sulfide) exposure during World War I and lung cancer was made by Case and Lea in 1955.68 Subsequent studies of Japanese workers involved in the production of mustard gas during World War II revealed substantial lung cancer risk elevations.69 British production workers during World War II exhibited much less pronounced elevations of risk.70 Latency has been estimated at 20 years, and there appears to be a higher rate of squamous cell carcinomas among those exposed.

Nickel (see Chapter 39.6) The first mention in the literature of respiratory cancers associated with nickel was a report of nasal cancer cases in a nickel refinery in 1933 by Bridge.71 Subsequent studies of workers in nickel mining, nickel refining, and nickel subsulfide roasting facilities have revealed high rates of lung cancer and nasal cancers. Nickel carbonyl, nickel subsulfide, nickel oxides, nickel acetate, nickel hydroxide, nickel fluoride, soluble nickel dusts, and nickel itself are animal carcinogens, but the specific forms of nickel carcinogenic to man remain the subject of ongoing investigation. Results of a collaborative investigation of nine cohorts of nickel workers and one case–control study attribute most respiratory cancers to oxidic and sulfitic nickel.72 Soluble forms of nickel appear to induce cancers at lower doses than do less soluble forms. Nickel-associated lung cancers have been noted more frequently in the large airways.

Exposure to metallic nickel has not been associated with increased lung cancer risk in humans.73 Studies of workers exposed to nickel alloys and pure nickel dust have not revealed an elevated rate of lung cancer, and elevated rates of lung cancer among nickel-exposed welders may be attributable to chromium or asbestos exposures.52–54 Similarly, confounding by chromium may be responsible for elevations of lung cancer risk among nickel/chromium plating workers.49

Polycyclic aromatic hydrocarbons (see Chapter 44) The link between human cancer and the PAHs was first forged by Sir Percivall Pott when he reported the development of scrotal cancer among English chimney sweeps in 1775.74 In the early part of the 20th century, many investigators suggested that increasing rates of lung cancer in the general population might be due to the use of tar and tar products. The first formal study of PAHs and lung cancer was carried out at a coal carbonization facility in Japan in 1936. Of the malignant neoplasms occurring in men at this facility, 80% (12 of 15 cases) were lung cancer – this at a time when lung cancer accounted for only 3% of all cancers in Japan. PAHs occur as large mixes of substances formed through the incomplete combustion of coal tar, pitch, coke, and oil. Each compound is made up of at least three aromatic rings and contains only carbon and hydrogen. Occupational exposure to the PAHs of coal tar occurs in coal gasification facilities, gas and coke works, iron and steel foundries, aluminum reduction plants, tar distillation facilities, shale oil extraction operations, wood impregnation facilities, and in the roofing and transportation industries. Other sources of PAHs include carbon black (a pyrolysis product of petroleum or natural gas), petroleum distillates, soot, and diesel exhaust. Increased risk of lung cancer has been found in coke oven workers,75,76 tar distillation workers,77 roofers,78 chimney sweeps,79 truckers,80,81 and persons employed in coal gasification facilities82 and aluminum reduction plants.83–85 Coke oven workers have been well studied, and a dose response based on proximity to the coke oven is firmly established. Risk among tar distillation workers is less well documented, and not all studies have revealed correlations between exposure and disease. Elevations of lung cancer risk among roofers and asphalt workers are likely due to coal tar and bitumen fume exposure. A long-term cohort study among Swedish chimney sweeps reported a doubling of lung cancer risk, and several smaller studies have revealed similar findings. Excess lung cancer risk has been reported among truckers and attributed to diesel exhaust, which contains variable amounts of PAHs. Estimation of dose in many such studies has been challenging, and dose responses have not been consistent.86,87 Coal gasification has been long associated with increased risk of lung cancer. Among aluminum reduction workers, those exposed to Soederberg electrolysis typically experience both higher PAH exposure and greater risk of lung cancer than those

732 Malignancies of the Respiratory Tract and Pluera employed in plants using the prebake process. Although many studies of aluminum reduction workers have revealed no excesses of lung cancer nor clear dose–response relationships,88,89 a nested case–control study of workers from two Soederberg plants showed a dose–response relationship based on years of exposure as well as measured benzene soluble fraction and benzo[a]pyrene.85 Exposure to PAHs appears to elevate the risk for all cell types of lung cancer. Exposure to PAHs is also widespread in the environment, deriving from tobacco smoke, fire fumes, ambient air pollution, and cooked food. Several studies have demonstrated increased lung cancer mortality associated with urban air pollution, but it is uncertain to what degree such excesses can be attributed to PAH contamination. There is evidence implicating environmental PAH as a lung cancer risk factor from studies in China of women heavily exposed to cooking and/or heating fumes.90–92

Radon (see Chapter 33.1) In 1597, Georgius Agricola wrote of a characteristic wasting disease observed in central European miners.93 Nearly 300 years later, Harting and Hesse identified ‘mountain illness’ as lung cancer and associated it with underground mining.94 Ongoing studies in the early 20th century supported the link between radon and lung cancer at the mines of Schneeberg in Germany and Joachimsthal in Czechoslovakia. By the late 1950s, a study of uranium miners on the Colorado plateau demonstrated elevated rates of lung cancer in heavily exposed workers, and numerous subsequent studies have confirmed that risk in other mining populations. With the discovery of high radon levels in homes built on the Reading Prong geologic formation in Pennsylvania in the 1980s, considerable attention turned to the hazard of low-level radon exposure in the general population. Radon 222 is an inert gas released in the radioactive decay of uranium 238. It has a half-life of 3.8 days and decays with alpha-particle emission through the shortlived radon daughters of polonium 218, lead 214, bismuth 214, and polonium 214 to lead 210. Polonium 218 and polonium 214 are themselves alpha-particle emitters. The radioactive particles may attach to aerosols or inhaled particles, gain access to cells lining the airways, and irradiate those cells at close range, causing chromosomal damage. Radon is an established carcinogen only in the respiratory tract. Units of radiation most commonly used for occupational and domestic exposures are the working level (WL), the working level month (WLM), and the concentrations of picocuries per liter (pCi/L) or becquerels per cubic meter (Bq/m3). One WL is any combination of radon progeny in 1 liter of air that ultimately releases 130,000 MeV of alpha energy during decay. Exposure to 1 WL for 170 hours equals 1 WLM. A radon concentration in the domestic setting of 1 pCi/L (2.2 radioactive decays per minute per liter of air) is approximately equal to 0.005 WL. One pCi/L is approximately equal to 37.5 Bq/m3. A typical domestic indoor radon exposure of 0.8 pCi/L is roughly equivalent

to 0.2 WLM/year, which translates into an approximately 15 WLM exposure over a lifetime. The present occupational standard for exposure in mines is 4 WLM per year. A high lung cancer mortality rate has been recognized among underground miners. Most studies have focused on uranium and iron miners, but elevated rates of lung cancer in miners of tin, zinc and lead, magnetite, fluorspar, metal ores, and niobium have been revealed as well. A pooled analysis of data from 11 mining cohorts determined that 40% of lung cancer among miners was due to radon.95 The study, which comprised over 65,000 men and more than 2700 lung cancer deaths, found that the association between estimated radon exposure and risk of lung cancer remained linear throughout the dose range examined. Overall excess relative risk was 0.0049 per WLM. Risk tended to decline with increasing age and time since last exposure. Consistent with other studies,96 it increased with duration of exposure, suggesting that for a given cumulative dose, longer duration exposures to lower radon levels are more harmful than short-term exposures to higher levels. The interaction of smoking with radon exposure has been addressed in a number of studies. Analyses of nonsmoking miners as well as a case–control study in Navajo men in whom smoking rates were very low have confirmed radon’s status as an independent carcinogen.97,98 The interaction of smoking and radon exposure contributes to the difficulty of assigning radon-related relative risks for the various tumor histologies. Early reports suggested relatively high rates of small-cell carcinoma, but more recent studies reveal a distribution of cell types that does not differ from that of the general population. In the combined analysis of 11 mining cohorts, it was estimated that radon accounted for 70% of lung cancer among non-smoking miners and 39% among miners who smoked. An extrapolation based on these data suggested that 30% of lung cancer deaths among non-smokers in the general population may be due to radon, while radon may account for as much as 11% of lung cancer deaths among smokers.95 With the investigation of indoor air quality in the 1970s, it became apparent that radon was invariably present in homes and other indoor spaces. Little attention was directed at the problem, however, until the mid 1980s, when a home located on the Reading Prong in Pennsylvania was found to have extremely high radon levels. Subsequent investigation of other dwellings located on the same geologic formation revealed radon concentrations in excess of 110 pCi/L. The lifetime dose for an individual living in such a dwelling would exceed 1500 WLM and would lie well within the dose range of heavily exposed miners. Subsequent surveys have revealed other areas with unusually high radon levels, including a region in the Tyrolean Alps, where indoor radon levels exceeding 6000 pCi/L have been measured.99 Surveys suggest that average domestic exposures in the US range from 0.8 to1.5 pCi/L. Primary sources of domestic radon are rock, soil, building material, and drinking water. Domestic radon levels are determined by the ease with which air can pass from soil or rock into the dwelling and the adequacy of ventilation; older dwellings are some-

Known Lung Carcinogens 733 what more prone to elevated levels. Most domestic radon measurements are made under conditions intended to maximize instrument readings – namely, basement sampling with windows closed. Such levels may differ by as much as fivefold from the average radon exposure in lived-in areas over the course of a year. Estimates that one in three homes have unacceptably high radon levels (greater than 4 pCi/L) are based largely on basement sampling and may not accurately reflect actual domestic exposures. Three principal study types have been used to evaluate the lung cancer risk of domestic radon exposure: ecologic surveys, case–control studies utilizing geologic and construction characteristics to estimate dose, and case–control studies utilizing measured radon levels. Ecologic surveys carried out in the US, Canada, Europe, and China have related estimates of prevailing radon exposure to lung cancer trends within specific geographic areas. Some such studies have shown higher rates of lung cancer in regions with greater average radon exposures, while others have not. Case– control studies, in which domestic radon levels have been estimated based on the geologic and construction characteristics of houses and neighborhoods, have been carried out in Sweden, the US, and Canada. Study results are mixed and suffer from small numbers of cases and controls. Surrogate exposure indices also do not consistently parallel measured levels of radon exposure. One study revealed that only 5% of the variance among domestic radon levels could be accounted for by geologic variables, and that geologic and housing-related variables combined accounted for only 15%–18% of the variance.100 The most reliable data regarding an association between domestic radon exposure and lung cancer come from a series of case–control studies in which actual radon measurements have been made in dwellings inhabited by subjects. More than 20 such studies have been undertaken. A meta-analysis of eight studies, each of which examined at least 200 lung cancer cases, has revealed a statistically significant summary odds ratio of 1.14 (95% CI, 1.0–1.3) associated with exposures of 4 pCi/L.101 The meta-analysis, which included more than 4200 lung cancer cases and over 6600 controls, concluded that actual risk in the general population from radon exposure is not likely to be higher than mining study-derived estimates. Several studies published subsequent to the meta-analysis have produced corroborating results.102–104 Domestic radon studies of nonsmokers have revealed similar risk elevations.105,106 Based on the accumulated evidence, it is reasonable to estimate that 15,000 lung cancer deaths per year in the US population are attributable to radon exposure. Because most individuals are exposed to low levels of radon, it has been estimated that mitigation of all dwellings to below 4 pCi/L would result only in a 2%–4% drop in lung cancer mortality.95 Due to the mobility of the US population, only a small percentage of radon-related lung cancers will occur in individuals currently living in dwellings with radon levels above 4 pCi/L. Still, radon levels in certain dwellings yield cumulative lifetime exposures comparable to those of miners. Although the use of radon mitigation systems in

such dwellings is clearly desirable, disagreement exists both as to how such high-risk dwellings should be identified, and whether or not radon mitigation should be instituted in dwellings with levels only marginally above the Environmental Protection Agency (EPA) action level of 4 pCi/L. Unfortunately, there is little that can be done to reduce background low-level lung cancer risk within the large segment of the population with low or average lifetime radon exposure.

Environmental tobacco smoke Environmental tobacco smoke (ETS) contains over 5000 chemicals, of which 43 are listed as known human or animal carcinogens by the IARC.107,108 ETS consists of sidestream smoke from the burning tip of a cigarette, as well as mainstream smoke inhaled and exhaled by the smoker. Most vapor phase contaminants and more than half of the air-borne particulates derive from the sidestream component.109 A meta-analysis carried out by the US EPA examined more than 30 studies, generating country-specific summary relative risks. Most studies enrolled female nonsmokers, with exposure tagged to the spouse’s smoking habits. The summary relative risk for the US based on eight studies was 1.22 (90% CI, 1.04–1.42). To date there have been more than 75 epidemiologic studies which have examined ETS and lung cancer, and 20 meta-analyses.110 One meta-analysis of 37 published studies involving more than 4600 cases of lung cancer reported a 26% (95% CI, 7%–47%) adjusted excess lung cancer risk among nonsmokers who lived with a smoker.111 Dose response based on both number of cigarettes smoked by the spouse and duration of exposure was statistically significant. A more recent meta-analysis examining 35 case–control studies and five cohort studies generated a pooled relative risk associated with exposure to husbands’ smoking of 1.19 (95% CI, 1.10–1.29) for case–control studies and 1.29 (95% CI, 1.04–1.62) for cohort studies. Because workplace exposure to ETS continues to be prevalent in many industries, ETS is also an important occupational risk factor for lung cancer. Based on the pooled results of five ETS studies, it has been estimated that non-smoking women exposed to smoking in the workplace have a 25% increased risk (95% CI, 8%–41%) over their baseline lung cancer risk.112 Studies which have directly assessed workplace exposures to ETS generate risk estimates similar to those from studies of domestic exposures to spousal smoking.113 Compared with mean nicotine concentrations from 1 to 3 μg/m3 in the homes of smokers, mean concentrations in workplaces which permit smoking range from 2 to 6 μg/m3 in offices, from 3 to 8 μg/m3 in restaurants, from 1 to 6 μg/m3 in the workplaces of blue-collar workers, and from 10 to 40 μg/m3 in bars.114 According to the NHANES III study, nearly 40% of US workers report being exposed to ETS in the workplace.115 Accumulated epidemiologic evidence strongly implicates ETS as a significant lung cancer risk factor. The US

734 Malignancies of the Respiratory Tract and Pluera EPA has estimated that ETS accounts for approximately 3000 lung cancer deaths per year in non-smokers.109

PROBABLE LUNG CARCINOGENS Exposures considered as probable lung carcinogens are summarized in Table 30.1.2.

Beryllium (see Chapter 39.3) The primary occupational groups exposed to beryllium are miners, beryllium production workers, workers in the production and fabrication of beryllium alloys, and workers in the electronics industries. Data supporting the association between beryllium exposure and lung cancer are derived from cohorts of workers at beryllium production facilities in Ohio and Pennsylvania and from the US Beryllium Case Registry, a database which has tracked clinical outcomes among individuals with acute and chronic beryllium disease. Follow-up of men and women enrolled in the Beryllium Case Registry has revealed a doubling of lung cancer risk based on 28 observed lung cancer deaths,116 while a cohort study of over 9000 workers employed in seven beryllium processing facilities reported an SMR of 1.26 (95% CI, 1.12–1.42).117 Significantly elevated SMRs were found at two of the seven plants, and the highest rates of lung cancer were seen among employees with a history of acute beryllium disease at a Lorain, Ohio, facility. A recent nested case–control study from one of seven production facilities found that lung cancer cases had shorter work tenures and lower lifetime cumulative beryllium exposures than controls, but higher average and maximum exposures.118 Some controversy surrounds the association between beryllium exposure and human lung cancer.119–121 Cancer excesses were primarily associated with work during early production years, when exposures were greatest, and questions persist regarding why certain processing facilities had greater excesses than others. Concern has been raised regarding a possible contribution to lung cancer mortality from acid mists at certain facilities and from residual confounding due to cigarette smoking. The IARC has concluded that there is sufficient evidence that beryllium is carcinogenic to humans.122

Cadmium (see Chapter 39.4) Primary occupational exposures to cadmium occur in the smelting and refining of copper, zinc, and lead, as well as in electroplating, welding, and the manufacture of batteries, plastics, and paints. Human epidemiologic study results have been mixed regarding the potential of cadmium exposure to cause lung cancer in humans. Elevated risk has been observed among US cadmium-exposed production workers,123,124 and a study of cadmium recovery workers found elevated lung cancer risk among those exposed to both cadmium and arsenic.125 However, the same study did not show elevated lung cancer risk among workers exposed solely to cadmium, and a study of copper

cadmium alloy workers did not reveal a positive correlation between cadmium exposure and lung cancer risk.126 Swedish nickel-cadmium battery workers were shown to have an elevated lung cancer risk, with an SMR of 176 (95% CI, 101–287), but there was no apparent dose response between cadmium or nickel and lung cancer risk.127 The IARC concluded in 1993 that there was sufficient evidence that cadmium is carcinogenic to humans.122

Man-made vitreous fibers (MMVF) (see Chapter 46) MMVF include rock wool, slag wool, glass fibers, and ceramic fibers. These materials are used primarily for insulation and have received particular scrutiny because the occupational groups exposed to them are those that have borne the legacy of past widespread asbestos use. There are two large study populations of MMVF-exposed workers: an American cohort of over 16,000 male workers at 17 US production plants128 and a European cohort of over 22,000 workers in 14 plants.129 Only one cohort study has examined workers exposed to refractory ceramic fibers (RCF).64,130 In the large American cohort, the overall lung cancer SMR was 1.13 (95% CI, 1.03–1.23). Glass wool workers had an SMR of 1.12 (95% CI, 1.00–1.24), while rock/slag wool workers had an SMR of 1.36 (95% CI, 1.06–1.71).128 In the large European cohort, overall SMR for workers with 1 year or more of employment was 1.05 (95% CI, 1.02–1.09). Glass wool workers had an SMR of 1.27 (95% CI, 1.07–1.50), while rock/slag wool workers had an SMR of 1.34 (95% CI, 1.08–1.63).129 A recent study of over 32,000 workers employed for 1 year or more at any of 10 US fiberglass manufacturing facilities found a 6% excess mortality from respiratory cancer (P = 0.05). There was a statistically non-significant 3% excess mortality among longer-term (>5 years) workers. Duration of exposure and cumulative exposure did not appear to be associated with increased risk of respiratory cancer.131 Several nested case–control studies examining glass wool and rock/slag wool exposures have allowed more detailed exposure and dose–response assessments, and have shown mixed results.132–134 RCF, potent carcinogens in animal models, have not been shown to be carcinogenic in human studies.130 However, the limited size of the exposed cohort and its relatively young age do not allow one to conclude that such fibers are without risk. A recent risk analysis has concluded that deaths in the RCF cohort are significantly below that which would be expected if RCF had the potency of either crocidolite or amosite. Mortality was also reported to be lower than would be expected if RCF had the potency of chrysotile, but the difference was not statistically significant.135 The current human epidemiologic evidence regarding lung cancer risk is strongest for rock wool and slag wool. Results regarding glass wool are mixed. Though recent follow-up data on exposed workers suggest a small risk, dose–response relationships have not been consistent. It is not possible to draw positive or negative conclusions from the limited epidemiologic study of RCF. The IARC considers MMVF to be possible human carcinogens (Group 2B).136

Other Respiratory Tract Cancers 735

A

B

Ambient air pollution (see Chapter 51) Considerable evidence has accumulated to suggest that exposure to ambient air pollution may increase lung cancer risk. Studies addressing the issue are extremely challenging due to a number of issues: pollution levels are difficult to characterize over the long periods of time which must be considered when evaluating lung cancer incidence; factors which affect exposure levels to the individual are challenging to measure, as are confounding exposures over large population groups; the air pollutants measured may themselves not impart cancer risk, and may or may not correlate well with potentially carcinogenic agents; and most measurement data represent average levels over large areas. Still, compelling results have accumulated from case– control and cohort studies, which suggest relative risks for lung cancer in the range of 1.5.137 In general, more recent studies have better controlled for differences in smoking among subjects, and some have tagged increased lung cancer risk to elevations of specific pollutants, such as fine or sulfate particulates.138,139 Pope et al have recently reported on the lung cancer mortality in a prospective cohort of approximately 1.2 million adults enrolled in 1982.140 Fine particulate and sulfur oxide-related pollution were associated with all-cause, lung cancer, and cardiopulmonary mortality. Each 10 μg/m3 elevation in fine particulate air pollution was associated with an approximately 8% increased risk of lung cancer mortality. The authors reported no associations with coarse particle fraction or total suspended particles.

OTHER RESPIRATORY TRACT CANCERS Mesothelioma The link between malignant mesothelioma and asbestos exposure was first suggested in case reports from the 1940s. Further evidence was gathered by Wagner and associates in 1960 in a study of diffuse pleural mesotheliomas associated with asbestos mining in South Africa.141 In 1964, Selikoff and colleagues reported a surprisingly high incidence of death from mesothelioma in a cohort of 632 insulation workers.9 In a larger cohort of 17,800 asbestos insulation workers followed from 1967 to 1976, nearly 8% of deaths were due to the tumor.142 Subsequent studies

Figure 30.1.1: Gross anatomy of malignant mesothelioma. A: Right lung encased with malignant pleural tumor. B: Peritoneal tumor.

have found consistently elevated rates of mesothelioma in asbestos workers. Diffuse malignant mesothelioma is a rare tumor of the serosal membranes of the lung and abdomen (Fig. 30.1.1). Incidence rates in American men range from four to 11 cases per million per year, and in American women, from one to four cases per million per year. Between 50% and 100% of all malignant mesotheliomas in North America and South Africa are attributable to asbestos exposure. In Turkey, the fibrous mineral zeolite has also been associated with a large epidemic of cases. Pleural mesotheliomas are more common than peritoneal mesotheliomas and account for nearly 90% of cases. Studies demonstrate an average latency period from first exposure to onset of disease of 30 to 35 years. Although higher cumulative exposures to asbestos generally correlate with a greater risk of malignant mesothelioma, studies have revealed that the tumor may occur with short-term (1 to 2 years) exposures and lower exposure levels. Household contacts of asbestos workers who may have washed asbestos-contaminated clothing, for example, are a recognized risk group.143 Smoking does not appear to increase the risk of mesothelioma. There is no evidence that asbestos increases the risk of solitary benign mesothelioma. Asbestos fiber type may influence the incidence of malignant mesothelioma. Lower rates have been reported in those exposed only to chrysotile, and some studies of chrysotile miners have not revealed elevated risk of mesothelioma.144,145 Malignant mesothelioma may be more common with exposure to the longer crocidolite fibers than with exposure to amosite and anthophylite. A recent autopsy study employing analysis of retained asbestos fibers in young people who died of mesothelioma has estimated that 80% of the disease is due to amosite and crocidolite.146 Others have suggested that chrysotile asbestos is the main cause of pleural mesothelioma, citing the necessity to consider longer follow-up periods for exposed cohorts, limited data regarding mesotheliomas among exposed household members, and more widespread exposure to chrysotile.147 A reasonable synthesis of the current data is that chrysotile imparts a lower risk of mesothelioma than do the amphiboles, but that chrysotile-exposed workers are at risk for mesothelioma.148 Peritoneal mesothelioma has not been associated with chrysotile exposure alone. There is concern that other fibrous substances may induce mesothelioma. Erionite exposure appears to profoundly increase risk of mesothelioma, while MMVF do not. A recent study examined a cohort of 162 immigrants to Sweden from Karain, a Turkish village in which there is

736 Malignancies of the Respiratory Tract and Pluera low-level environmental exposure to erionite, a fibrous zeolite. Of 18 deaths which occurred in the cohort, 14 (78%) were due to malignant pleural mesothelioma. In addition, there were five patients with mesothelioma who were still alive. Risks of mesothelioma were elevated 135-fold and 1336-fold, respectively, in men and women, compared with an age-matched Swedish population.149 In contrast, no excesses of mesothelioma have been detected in the American cohort of MMVF production workers.150 A recent German case–control study could neither detect nor exclude an association between MMVF and mesothelioma.151

Clinical features Typical symptoms of malignant mesothelioma include chest pain, dyspnea, weight loss, and cough. Fever also may be present. Chest pain of a dull gnawing quality is the most common symptom and tends to localize over the involved side, although radiation to the arm or shoulder may occur. Cough is usually not a prominent symptom until the tumor has invaded the mediastinum. Pleural effusion is the most common physical finding, and chest radiographs usually reveal unilateral fluid with persistent pleural thickening and nodularity following fluid removal. Early on, thoracic structures often are drawn towards the tumor by its dense desmoplastic growth (Fig. 30.1.2). In advanced disease, there may be enlargement of the affected hemithorax, bulging of intercostal spaces, and displacement of the trachea and mediastinum to the unaffected side. Local tumor growth also may depress the diaphragm, giving the impression of splenomegaly or hepatomegaly. Supraclavicular and axillary lymph node enlargement, subcutaneous nodules in the chest wall, and rib tenderness also may be seen in advanced disease.

Figure 30.1.2: Plain chest X-ray study showing right-sided pleural malignant mesothelioma, with shift of thoracic contents.

Diagnosis Diagnosis often is delayed due to the indistinct clinical features of early disease. Despite the fact that the overwhelming majority of cases of malignant mesothelioma are caused by asbestos exposure, asbestosis or other nonmalignant effects are present in only about 20% of individuals diagnosed with the tumor. Pleural or peritoneal fluid is always exudative but of variable cellularity. Cytologic distinction of malignant cells from reactive mesothelial cells is often difficult. The presence of hyaluronic acid in fluid may be useful. Ultimately, the diagnosis depends on defining the anatomy of the tumor and obtaining definitive tissue. In cases in which plain X-ray studies of chest and abdomen may be non-diagnostic, computed tomography (CT) scans are often very revealing (Fig. 30.1.3). Biopsy is definitive but may be difficult to interpret, even by experienced readers. The spindle cell types (Fig. 30.1.4) are generally differentiable from benign lesions and other tumors, but some mesotheliomas have striking adenomatous features, making them difficult to differentiate from metastatic adenocarcinoma. Immunohistochemical stains to exclude GI or GU cell origin and confirm mesenchymal surface markers are now widely employed to differentiate the two.

Treatment No effective therapy exists for malignant mesothelioma. Also, there is no demonstrated benefit of early diagnosis beyond the timely acquisition of compensation benefits for the occupationally exposed. Average survival is 12 to 15 months from onset of symptoms, and most patients die within 1 year of diagnosis; there are a few long-term survivors, however, for unclear reasons. Death usually occurs from extension of the tumor into adjacent struc-

Figure 30.1.3: Malignant peritoneal mesothelioma encasing abdominal organs on CT scan.

Other Respiratory Tract Cancers 737

Figure 30.1.4: Histology of spindle cell malignant mesothelioma (H & E stain, high power).

tures in the chest or the abdomen; metastases occur uncommonly. Hypoglycemia and hemolytic anemia have been reported as paraneoplastic complications. In general, palliative care should be emphasized, although surgery, chemotherapy, and radiotherapy have potential value in individual cases.

Upper respiratory cancer Known and probable carcinogenic agents are summarized in Table 30.1.3.

Sinonasal and nasopharyngeal cancers The first recognized cases of occupational sinonasal cancer were reported in a Welsh nickel refinery in 1932. Since that time a number of industrial processes and specific exposures have been associated with the tumor. Sinonasal cancers are rare, occurring at a rate of less than 10 cases per million per year in the US. Squamous cell cancers, which make up about half of all sinonasal tumors, and adenocarcinomas have been linked with occupational exposures. Adenomas, lymphomas, melanomas, and adenoid cystic carcinomas make up the remainder of sinonasal cancers. The primary exposures linked to sinonasal cancer are nickel, wood dust, formaldehyde, cutting oils, and chroAgents Sinonasal cancer Known carcinogens: Nickel Wood dust Probable carcinogens: Chromium Cutting oils Formaldehyde Laryngeal cancer Probable carcinogens: Asbestos Cutting oils

Exposure settings

Nickel refining Furniture making Chromium pigment production Machinists Production of formaldehyde resins, particle board, apparel Insulating, shipyard work Machinists

Table 30.1.3 Upper respiratory tract carcinogens

mium. Cancers in nickel workers have occurred most frequently in the nose and ethmoid sinuses, and usually are of squamous histology.152 Most epidemiologic studies were carried out in nickel refineries where exposures occurred to nickel oxides, nickel sulfide dust, soluble nickel compounds, and nickel carbonyl. Wood dust has been recognized as a sinonasal carcinogen since 1965, when a strikingly high incidence of the tumor occurred in an English furniture factory, and several case–control studies have documented increased risk among wood dustexposed occupations since then.153,154 The most frequently occurring tumor histology following wood dust exposure is adenocarcinoma. Wood dusts contain a number of biologically active substances as well as preservatives, pesticides, and fungal proteins. The active carcinogen in wood dust is unknown, although PAHs may play a role because wood often is charred in the furniture-manufacturing process. A number of case–control studies have evaluated risk of sinonasal cancer among formaldehyde-exposed workers.155–157 Pooled analyses suggest that risk of both adenocarcinoma and squamous cell carcinoma is increased. Risk of nasopharyngeal cancer has also been shown to be significantly elevated. Formaldehyde may potentiate the carcinogenic potential of wood dust exposure. An association between cutting oils and sinonasal cancer has been demonstrated in a small number of case–control studies, though a recent investigation found a significant association with nasopharyngeal cancer only.158 Occupations in which an increased risk of sinonasal tumors have been reported include shoe manufacturing and repair, textile and clothing manufacture, agricultural work, and food preparation. Leather dusts, natural fibers, pesticides, and PAHs have been suggested as possible etiologic agents. Exposure to wood smoke under poorly ventilated conditions in the home has been proposed as an environmental etiology, which may explain higher rates of nasopharyngeal carcinoma among Chinese, other Asians, and natives of Alaska. Cigarette smoking has not been implicated as a major cause of sinonasal cancer. Although it is a rare tumor, sinonasal cancer should be suspected in a patient with significant exposures who complains of chronic nasal discharge, obstruction, fullness, intermittent bleeding, sinus pain, or pain in the upper teeth. The persistence of such symptoms despite usual clinical interventions for more typical sinus problems is cause for referral to an otorhinolaryngologist for further evaluation. The diagnostic evaluation, natural history, and treatments of occupational sinonasal cancer do not differ from non-occupationally related disease.

Laryngeal cancer Laryngeal cancer is considerably more common than sinonasal cancer, making up about 2% of the annual cancer incidence in the US. The primary risk factor is cigarette smoking, with some additional risk imparted by alcohol consumption. Occupational risk factors may include exposure to asbestos, cutting oils, acid mists, and nickel. Nearly all tumors are of squamous histology.

738 Malignancies of the Respiratory Tract and Pluera Increased rates of laryngeal cancer linked to asbestos exposure have been observed in miners, insulators, shipyard workers, and manufacturers of asbestos products. A recent meta-analysis of 69 asbestos-exposed occupational cohorts yielded SMRs for asbestos of 157 and 133 with and without latency, respectively.159 Greater risks have generally been found in older studies with less rigorous control for tobacco exposure.160 Results have been mixed with regard to a dose response between asbestos exposure and elevation of laryngeal cancer risk.161,162 Machinists have been shown to have an elevated risk of laryngeal cancer in several studies.163–165 Risk appears to be attributable to straight mineral oil exposure and may be enhanced during processes likely to produce PAHs. Acid mists also have been associated with laryngeal cancer in some, but not all, studies.166–168 Longduration, high-level exposures appear to impart greater risk than do short-term exposures to lower levels. Rates of laryngeal cancer among nickel workers have been elevated in some studies based on small numbers, but others have reported no significant increases.169–171 As with other respiratory tract cancers, occupationally related laryngeal cancer does not differ from non-occupational disease in its clinical presentation, natural history, or recommended therapy.

PREVENTION OF RESPIRATORY CANCER Given the lack of effective treatment for most respiratory cancers, the most effective method to reduce their considerable mortality rate is primary prevention. Identification of etiologic agents in the workplace is an ongoing task that requires physicians, industrial hygienists, epidemiologists, and policy makers to be sensitive to patterns of exposure and disease and to act prudently even when information is incomplete. Adherence to workplace standards is likely to yield decreasing rates of occupational respiratory cancer over the next several decades because many cancers today reflect exposures prior to the implementation of current standards. Still, standards must be continually reassessed as better epidemiologic studies yield new information. Smoking cessation efforts in the workplace also are vitally important. Secondary prevention in the form of mandated medical monitoring is an adjunctive approach to occupational disease prevention. The efficacy of lung cancer screening in high-risk populations and choice of diagnostic tools to use remains an area of controversy. Presently, the Occupational Safety and Health Administration (OSHA) mandates medical monitoring for workers exposed to the following carcinogens: asbestos, acrylonitrile, arsenic, bischloromethyl ether, cadmium, coke oven emissions, silica, and vinyl chloride. Medical monitoring in each case involves physical examination, chest radiograph, and ILO rating. Sputum cytology examination is required for some workers, including those exposed to arsenic or coke oven emissions. The asbestos standard requires chest radiographs every 5 years for the first 10 years after the initial exposure, and, after 10 years of latency, every 2 years

between the ages of 35 and 45 years and annually after age 45. NIOSH recommends medical monitoring in addition for workers exposed to beryllium, carbon black, chromium VI, coal tar products, inorganic nickel, chrysene, and coal gasification processes. Recommended monitoring includes history, chest examination, chest radiograph, and, in some cases, sputum cytology. Several studies have been performed to evaluate current tools for medical monitoring of lung cancer in the workplace. The value of the screening chest radiograph, even in workers with considerably elevated lung cancer risk, has yet to be firmly established, though re-analyzed results of clinical trials suggest that screening may improve clinical outcomes in detected cases.172,173 One study has shown that lung cancer cases detected through chest X-ray screening have a 5-year survival rate of 19% versus 10% among cases detected through conventional healthcare.174 Sputum cytology monitoring has been proposed as an alternative or supplemental monitoring tool. Lung cancers detected by sputum cytology tend to be squamous cell carcinomas of the major or segmental bronchus, whereas peripheral adenocarcinomas and large- and small-cell carcinomas more often are detected by X-rays. The addition of sputum cytology examination to radiographic screening has not brought about a decreased mortality from primary lung carcinoma. Sputum cytology results, however, may reveal mucosal changes that can identify individuals at increased risk. Low-dose spiral CT has been proposed as a more effective means of screening high-risk populations for lung cancer.175 In one study, 62 lung cancers were detected among 6341 participants. For single CT examinations, the mean detection rate (0.36%) was nearly 10 times greater than that achieved through chest X-ray screening.176,177 Of the lung cancer cases detected, 83% were pathologic state IA, and 66% were less than 15 mm in diameter. Conventional radiography did not detect 63% of the cancers detected by spiral CT. The reported sensitivity and specificity were 57% and 95%, respectively. Another study evaluated spiral CT screening of 1000 smokers 60 years of age and over, of whom 14% reported asbestos exposure. Screening initially identified 233 individuals with noncalcified nodules, among whom 27 lung cancers were ultimately resected. Malignant disease was detected four times more frequently with spiral CT than with chest radiography (2.7% versus 0.7%), and stage I tumors were detected six times more frequently (2.3% versus 0.4%).175,178,179 A screening program for 602 asbestos-exposed Finnish construction workers found abnormalities in 65 workers, among whom five lung cancers and one mesothelioma were detected. Sensitivity and specificity were estimated at 100% and 90%, respectively, with a positive predictive value of 8% in the studied population.180 The search for chemopreventive agents which might be administered to populations at high risk for lung cancer, such as heavy smokers or asbestotics, has not proven fruitful to date. Primary prevention trials with alpha-tocopherol with or without beta-carotene in the

Prevention of Respiratory Cancer 739 Alpha-Tocopherol, Beta-Carotene (ATBC) Trial in 29,133 smokers in Finland, and with beta-carotene and retinyl palmitate in the Beta-Carotene and Retinol Efficacy Trial (CARET) in 18,254 smokers, former smokers, and asbestosexposed workers in the US, have shown increases in lung cancer incidence, suggesting that beta-carotene itself or retinol might have carcinogenic properties.181–183 A randomized trial in a population of 22,071 male physicians, of whom 11% were current smokers and 39% former smokers, showed no significant difference in the number of lung cancer cases between those receiving placebo and those receiving supplemental beta-carotene.184 Despite the failures of such trials, epidemiologic evidence suggests that diets rich in fruits and vegetables confer some protection against respiratory and other cancers.185 Beta-carotene may also affect different epithelial tissues differently, as evidenced by its apparent protective effect against precancerous oral lesions and a statistically non-significant reduction in secondary head and neck cancers.186,187 Clearly, successes with chemoprevention, if they are to come, await a better understanding of the mechanisms underlying lower cancer rates among populations with apparently protective dietary habits.

References 1. Doll R, Peto R. The causes of cancer: quantitative estimates of avoidable risks of cancer in the U.S. today. J Natl Cancer Inst 1981; 66:1191-308. 2. Vineis P, Thomas T, Hayes R, et al. Proportion of lung cancers in males due to occupation in different areas of the US. Int J Cancer 1988; 42:851-6. 3. Morabia A, Markowitz S, Garibaldi K, Wunder E. Lung cancer and occupation: results of a multicentre case-control study. Br J Ind Med 1992; 49:721-7. 4. Steenland K, Loomis D, Shy C, Simonsen N. Review of occupational lung carcinogens. Am J Ind Med 1996; 29:474-90. 5. Lynch KM, Smith WA. Pulmonary asbestosis. III: Carcinoma of lung in asbesto-silicosis. Am J Cancer 1935; 56-64. 6. Merewether E. Chief Inspector of Factories. Annual Report for 1947. London: HMSO; 1949:79-81. 7. Gloyne SR. Pneumoconiosis. A historical survey of necropsy material in 1205 cases. Lancet 1951; 1:810. 8. Doll R. Mortality from lung cancer in asbestos workers. Br J Ind Med 1955; 12:81-6. 9. Selikoff IJ, Churg J, Hammond EC. Asbestos exposure and neoplasia. JAMA 1964; 188:22-6. 10. Armstrong B, DeKlerk N, Musk A, Hobbs M. Mortality in miners and millers in Western Australia. Br J Ind Med 1988; 45:5-13. 11. Dement J, Brown D, Okuin A. Lung cancer mortality among asbestos textile workers: a review and update. Ann Occup Hyg 1994; 38:525-32. 12. Gardner M, Powell C, Gardner A, Winter P, Fletcher A. Continuing high lung cancer mortality among ex-amosite factory workers and a pilot study of individual anti-smoking advice. J Soc Occup Med 1988; 38:69-72. 13. Raffn E, Lynge E, Korsgaard B. Incidence of lung cancer by histological type among asbestos cement workers in Denmark. Br J Ind Med 1993; 50:85-9. 14. Sluis-Cremer GK, Liddell F, Logan W, Bezuidenhout B. The mortality of amphibole miners in S. Africa. Br J Ind Med 1992; 49:566-75. 15. Cullen MR. Controversies in asbestos-related lung cancer. Occup Med 1987; 2:259-73

16. Lee PN. Relation between exposure to asbestos and smoking jointly and the risk of lung cancer. Occup Environ Med 2001; 58:145. 17. McFadden D, Wright JL, Wiggs B, et al. Smoking inhibits asbestos clearance. Am Rev Respir Dis 1986; 133:372-4. 18. Tron V, Wright JL, Harrison N, et al. Cigarette smoke makes airway and early parenchymal asbestos-induced lung disease worse in the guinea pig. Am Rev Respir Dis 1987; 136:271-5. 19. Gerde P, Scholander P. Adsorption of benzo(a)pyrene on to asbestos and manmade mineral fibres in an aqueous solution and in a biologic model solution. Br J Ind Med 1988; 45:682-8. 20. Berry G. Mortality of workers certified by pneumoconiosis medical panels as having asbestosis. Br J Ind Med 1981; 38:130-7. 21. Cookson W, Musk A, Glancy J, et al. Compensation, radiographic changes, and survival in applicants for asbestosis compensation. Br J Ind Med 1985; 42:461-8. 22. Hughes J, Weill H. Asbestosis as a precursor of asbestosrelated cancer: results of a prospective study. Br J Ind Med 1991; 48:229-33. 23. Weiss W. Asbestosis: a marker for the increased risk of lung cancer among workers exposed to asbestos. Chest 1999; 115:536-49. 24. Finkelstein MM. Radiographic asbestosis is not a prerequisite for asbestos-associated lung cancer in Ontario asbestoscement workers. Am J Ind Med 1997; 32:341-8. 25. Wilkinson P, Hansell DM, Janssenns J, et al. Is lung cancer associated with asbestos exposure when there are no small opacities on the chest radiograph? Lancet 1995; 345:1074-8. 26. Hillerdal G, Henderson DW. Asbestos, asbestosis, pleural plaques and lung cancer. Scand J Work Environ Health 1997; 23:93-103. 27. Weiss W. Asbestos and lobar site of lung cancer. Occup Environ Med 2000; 57:358. 28. Lee BW, Wain JC, Kelsey KT, et al. Association of cigarette smoking and asbestos exposure with location and histology of lung cancer. Am J Respir Crit Care Med 1998; 157:748-55. 29. Davis JMG, Addison J, Bolton RE, et al. Inhalation and injection studies in rats using dust samples from chrysotile asbestos prepared by a wet dispersion process. Exp Pathol 1986; 67:113-29 30. Yano E, Wang ZM, Wang XR, Wang MZ, Lan YJ. Cancer mortality among workers exposed to amphibole-free chrysotile asbestos. Am J Epidemiol 2001; 154:538-43. 31. Stanton MF, Layard M, Tegeris A, et al. Relation of particle dimension to carcinogenicity in amphibole asbestoses and other fibrous minerals. J Natl Cancer Inst 1981; 67:965-75. 32. Browne K. Is asbestos or asbestosis the cause of the increased risk of lung cancer in asbestos workers? Br J Ind Med 1986; 43:145-9. 33. Camus M, Siemiatycki J, Meek B. Nonoccupational exposure to chrysotile asbestos and the risk of lung cancer. N Engl J Med 1998; 338:1556-71. 34. Saupe E. Carcinoma of the lung in arsenic miners. Gewerbpathol 1930; 1:582. 35. Welch K, Higgins I, Oh M, Burchfiel C. Arsenic exposure, smoking, and respiratory cancer in copper smelter workers. Arch Environ Health 1982; 37:325-35. 36. Lee-Feldstein A. Cumulative exposure to arsenic and its relationship to respiratory cancer among copper smelter employees. J Occup Med 1986; 28:296-302. 37. Jarup L, Pershagen G, Wall S. Cumulative arsenic exposure and lung cancer in smelter workers: a dose-response study. Am J Ind Med 1989; 15:31-41. 38. Jarup L, Pershagen G. Arsenic exposure, smoking and lung cancer in smelter workers – a case-control study. Am J Epidemiol 1991; 134:545-51. 39. Taylor P, Qiao Y, Schatzkin A, et al. Relation of arsenic exposure to lung cancer among tin miners in Yunnan Province, China. Br J Ind Med 1989; 46:881-6.

740 Malignancies of the Respiratory Tract and Pluera 40. Ott M, Holder B, Gordon H. Respiratory cancer and occupational exposure to arsenicals. Arch Environ Health 1974; 29:250-5. 41. Englyst V, Lundstrom NG, Gerharsson L, Rylander L, Nordberg G. Lung cancer risks among lead smelter workers also exposed to arsenic. Sci Total Environ 2001; 273:77-82. 42. Simeonova PP, Luster MI. Mechanisms of arsenic carcinogenicity: genetic or epigenetic mechanisms? J Environ Pathol Toxicol Oncol 2000; 19:281-6. 43. Hertz-Piccotto I, Smith AH, Holzman D, Lipsett M, Alexeeff G. Synergism between occupational arsenic exposure and smoking in the induction of lung cancer. Epidemiology 1992; 3:23-31. 44. Figueroa WG, Raszkowski R, Weiss W. Lung cancer in chloromethyl methy ether workers. N Engl J Med 1973; 288:1096-7. 45. Collingwood KW, Pasternack BS, Shore RE. An industrywide study of respiratory cancer in chemical workers exposed to chloromethyl ethers. J Natl Cancer Inst 1987; 78:1127-36. 46. Blair A, Kazerouni N. Reactive chemicals and cancer. Cancer Causes Control 1997; 8:473-90. 47. Machle W, Gregorio F. Cancer of the respiratory tract in the US chromate producing industry. Public Health Rep 1948: 63:1114-27. 48. Rosenman KD, Stanbury M. Risk of lung cancer among former chromium smelter workers. Am J Ind Med 1996; 29:491-500. 49. Sorahan T, Burges DC, Hamilton L, Harrington JM. Lung cancer mortality in nickel/chromium platers, 1946-1965. Occup Environ Med 1998; 55:236-42. 50. Itoh T, Takahashi K, Okubo T. Mortality of chromium plating workers in Japan – a 16-year follow-up study. J UOEH 1996; 18:7-18. 51. Sjogren B, Hansen KS, Kjuus H, Persson PG. Exposure to stainless steel welding fumes and lung cancer: a metaanalysis. Occup Environ Med 1994; 51:335-6. 52. Langard S. Nickel-related cancer in welders. Sci Total Environ 1994; 148:303-9. 53. Moulin JJ. A meta-analysis of epidemiologic studies of lung cancer in welders. Scand J Work Environ Health 1997; 23:104-13. 54. Becker N. Cancer mortality among arc welders exposed to fumes containing chromium and nickel. Results of a third follow-up: 1989-1995. J Occup Environ Med 1999; 41:294-303. 55. Paddle GM. Metaanalysis as an epidemiological tool and its application to studies of chromium. Regul Toxicol Phamacol 1997; 26:S42-S50. 56. Pastides H, Austin R, Lemeshow S, Klar J, Mundt KA. A retrospective-cohort study of occupational exposure to hexavalent chromium. Am J Ind Med 1994; 25:663-75. 57. Korallus U, Ulm K, Steinmann-Steiner-Haldenstaett W. Bronchial carcinoma mortality in the German chromateproducing industry: the effects of process modification. Int Arch Occup Environ Health 1993; 65:171-8. 58. Davies JM, Easton DF, Bidstrup PL. Mortality from respiratory cancers and other causes in United Kingdom chromate production workers. Br J Ind Med 1991; 48:299-313. 59. Alexander BH, Checkoway H, Wechsler L, Heyer NJ, Muhm JM, O’Keeffe TP. Lung cancer in chromate-exposed aerospace workers. J Occup Environ Med 1996; 38:1253-8. 60. IARC. IARC Monogr Eval Carcinog Risks Hum 1987, Supplement 7. 61. Mancuso TF. Chromium as an industrial carcinogen: Part I. Am J Ind Med 1997; 31:129-39. 62. Gibb HJ, Lees PSJ, Pinsky PF, Rooney BC. Lung cancer among workers in chromium chemical production. Am J Ind Med 2000; 38:115-26. 63. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans: Silica, Some Silicates, Coal Dust and ParaAramid Fibrils. Lyon, 15-22 October 1996. IARC Monogr Eval Carcinog Risks Hum 1997; 68:1-475.

64. Steenland K, Stayner L. Silica, asbestos, man-made mineral fibers, and cancer. Cancer Causes Control 1997; 8:491-503. 65. Carta P, Aru G, Manca P. Mortality from lung cancer among silicotic patients in Sardinia: an update study with 10 more years of follow up. Occup Environ Med 2001; 58:786-93. 66. Steenland K, Sanderson W. Lung cancer among industrial sand workers exposed to crystalline silica. Am J Epidemiol 2001; 153:695-703. 67. Steenland K, Mannetje A, Boffetta P, et al. Pooled exposureresponse analyses and risk assessment for lung cancer in 10 cohorts of silica-exposed workers: an IARC multicentre study. Cancer Causes Control 2001; 12:773-84. 68. Case RA, Lea AJ. Mustard-gas poisoning, chronic bronchitis and lung cancer: investigation into possibility that poisoning by mustard gas in 1914-18 war might be factor in production of neoplasia. Br J Prev Soc Med 1955; 9:62-72. 69. Nishimoto Y, Yamakido M, Ishioka S, Shigenobu T, Yukutake M. Epidemiologic studies of lung cancer in Japanese mustard gas workers. In: Miller RW, ed. Unusual occurrences as clues to cancer etiology. Tokyo: Taylor & Francis; 1988:95-101. 70. Easton DF, Peto J, Doll R. Cancers of the respiratory tract in mustard gas workers. Br J Ind Med 1988; 45:652-9. 71. Bridge JC. Annual Report of the Chief Inspector of Factories and Workshops for 1932. London: HMSO; 1933:103-4. 72. ICNCM (International Committee on Nickel Carcinogenesis in Man). Report of the International Committee on Nickel Carcinogenesis in Man. Scand J Work Environ Health 1990; 16:1-84. 73. IARC. Chromium, nickel and welding. Monograph 49. Lyon: IARC; 1990. 74. Pott P. Chirurgical observations relative to the cataract, the polypus of the nose, the cancer of the scrotum, the different kinds of ruptures, and the mortification of the toes and feet. London: Hawes, Clark & Collins; 1775. 75. Lloyd JW, Lundin FE Jr, Redmond CK, Geiser PB. Longterm mortality study of steelworkers. IV. Mortality by work area. J Occup Med 1970; 12:151-7. 76. Costantino JP, Redmond CK, Bearden A. Occupationally related cancer risk among coke oven workers: 30 years of follow-up. J Occup Environ Med 1995; 37:597-604. 77. Maclaren WM, Hurley JF. Mortality of tar distillation workers. Scand J Work Environ Health 1987; 13:404-11. 78. Partanen T, Boffetta P. Cancer risk in asphalt workers and roofers: review and meta-analysis of epidemiologic studies. Am J Ind Med 1994; 26:721-40. 79. Evanoff BA, Gustavsson P, Hogstedt C. Mortality and incidence of cancer in a cohort of Swedish chimney sweeps: an extended follow up study. Br J Ind Med 1993; 50:450-9. 80. Steenland NK, Silverman DT, Hornung RW. Case-control study of lung cancer and truck driving in the Teamsters Union. Am J Public Health 1990; 80:670-4. 81. Steenland K, Deddens J, Stayner L. Diesel exhaust and lung cancer in the trucking industry: exposure-response analyses and risk assessment. Am J Ind Med 1998; 34:220-8. 82. Doll R, Vessey MP, Beasley RWR, et al. Mortality of gasworkers – final report of a prospective study. Br J Ind Med 1972; 29:394-406. 83. Andersen A, Dahlberg BE, Magnus K, Wannag A. Risk of cancer in the Norwegian aluminium industry. Int J Cancer 1982; 29:295-8. 84. Gibbs GW. Mortality of aluminum reduction plant workers, 1950 through 1977. J Occup Med 1985; 27:761-70. 85. Armstrong B, Tremblay C, Baris D, Theriault G. Lung cancer mortality and polynuclear aromatic hydrocarbons: a case-cohort study of aluminum production workers in Arvida, Quebec, Canada. Am J Epidemiol 1994; 139:250-62. 86. Gustavsson P, Plato N, Lidstrom EB, et al. Lung cancer and exposure to diesel exhaust among bus garage workers. Scand J Work Environ Health 1990; 16:348-54. 87. Health Effects Institute Diesel Epidemiology Expert Panel. Diesel emissions and lung cancer: epidemiology and

Prevention of Respiratory Cancer 741

88.

89.

90.

91.

92.

93. 94.

95.

96.

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

quantitative risk assessment. Cambridge, MA: Health Effects Institute; 1999. Romundstad P, Haldorsen T, Andersen A. Cancer incidence and cause specific mortality among workers in two Norwegian aluminium reduction plants. Am J Ind Med 2000; 37:175-83. Ronneberg A, Langmark F. Epidemiologic evidence of cancer in aluminum reduction plant workers. Am J Ind Med 1992; 22:573-90. Zhou BS, Wang TJ, Guan P, Wu JM. Indoor air pollution and pulmonary adenocarcinoma among females: a case-control study in Shenyang, China. Oncol Rep 2000; 7:1253-9. Seow A, Poh WT, The M, et al. Fumes from meat cooking and lung cancer risk in Chinese women. Cancer Epidemiol Biomarkers Prev 2000; 9:1215-21. Mumford JL, He XZ, Chapman RS, et al. Lung cancer and indoor air pollution in Xuan Wei, China. Science 1987; 235:217-20. Agricola G. De re metallica. Translated by Hoover H, Hoover L. New York: Dover; 1950. Harting FH, Hesse W. Der Lungenkrebs, die Bergkrankheit in der Schneeberger Gruben. Vgschr Med Gerish Off Sanit 1879; 30:296-309 and 31:102-32, 313-37. Lubin JH, Boice JD Jr, Edling C, et al. Lung cancer in radonexposed miners and estimation of risk from indoor exposure. J Natl Cancer Inst 1995; 87:817-27. Lubin JH, Qiao YQ, Taylor PR, et al. Quantitative evaluation of the radon and lung cancer association: a case-control study of Chinese tin miners. Cancer Res 1990; 50:174-80. Samet JM, Kutvirt DM, Waxweiler RJ, Key CR. Uranium mining and lung cancer in Navajo men. N Engl J Med 1984; 310:1481-4. Roscoe RJ, Steenland K, Halperin WE, et al. Lung cancer mortality among nonsmoking uranium miners exposed to radon daughters. JAMA 1989; 262:629-33. Ennemoser O, Oberdorfer E, Brunner P, et al. Mitigation of indoor radon in an area with unusually high radon concentrations. Health Phys 1995; 69:227-32. Levesque B, Gauvin D, McGrego RG, et al. Radon in residences: influences of geological and housing characteristics. Health Phys 1997; 72:907-14. Lubin JH, Boice JD Jr. Lung cancer risk from residential radon: meta analysis of eight epidemiologic studies. J Natl Cancer Inst 1997; 89:49-57. Field RW, Steck DJ, Smith BJ, et al. The Iowa radon lung cancer study – phase I: Residential radon gas exposure and lung cancer. Sci Total Environ 2001; 272:67-72. Kreienbrock L, Kreuzer M, Gerken M, et al. Case-control study on lung cancer and residential radon in western Germany. Am J Epidemiol 2001; 153:42-52. Alavanja MC, Lubin JH, Mahaffey JA, Brownson RC. Residential radon exposure and risk of lung cancer in Missouri. Am J Public Health 1999; 89:1042-8. Kreuzer M, Gerken M, Kreienbrock L, Wellmann J, Wichmann HE. Lung cancer in lifetime nonsmoking men – results of a case-control study in Germany. Br J Cancer 2001; 84:134-40. Lagarde F, Axelsson G, Damber L, Mellander H, Nyberg F, Pershagen G. Residential radon and lung cancer among never-smokers in Sweden. Epidemiology 2001; 12:396-404. O’Neill IK, Brunnemann KD, Dodet B, et al., eds. Environmental carcinogens: methods of analysis and exposure measurement. Volume 9 – Passive smoking. Lyon: IARC; 1987. (IARC scientific publications no. 81.) Brownson RC, Alavanja MC, Caporaso N, Somoes EJ, Chang JC. Epidemiology and prevention of lung cancer in nonsmokers. Epidemiol Rev 1998; 20:218-36. Respiratory health effects of passive smoking: lung cancer and other disorders. Washington, DC: Office of Health and Environment, Office of Research and Development, US

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

Environmental Protection Agency; 1992. (Publication no. EPA/600/6-90/006F.) Taylor R, Cumming R, Woodward A, Black M. Passive smoking and lung cancer: a cumulative meta-analysis. Aust NZ J Public Health 2001; 25:203-11. Hackshaw AK, Law MR, Wald NJ. The accumulated evidence on lung cancer and environmental tobacco smoke. BMJ 1997; 315:980-8. Brown KG. Lung cancer and environmental tobacco smoke: occupational risk to nonsmokers. Environ Health Perspect 1999; 107(Suppl 6):885-90. Reynolds P. Epidemiologic evidence for workplace ETS as a risk factor for lung cancer among nonsmokers: specific risk estimates. Environ Health Perspect 1999; 107(Suppl 6):865-72. Hammond SK. Exposure of US workers to environmental tobacco smoke. Environ Health Perspect 1999; 107(Suppl 2):329-40. Pirkle JL, Flegal KM, Bernert JT, et al. Exposure of the US population to environmental tobacco smoke. The Third National Health and Nutrition Examination Survey, 1988 to 1991. JAMA 1996; 275:1233-40. Steenland K, Ward E. Lung cancer incidence among patients with beryllium disease: a cohort mortality study. J Natl Cancer Inst 1991; 83:1380-5. Ward E, Okun A, Ruder A, Fingerhut M, Steenland K. A mortality study of workers at seven beryllium processing plants. Am J Ind Med 1992; 22:885-904. Sanderson WT, Ward EM, Steenland K, Petersen MR. Lung cancer case-control study of beryllium workers. Am J Ind Med 2001; 39:133-44. Deubner DC, Lockey JL, Kotin P, et al. Re: Lung cancer casecontrol study of beryllium workers. Am J Ind Med 2001; 40:284-5. Sanderson WT, Ward EM, Steenland K, et al. Re: Response to criticisms of “Lung cancer case-control study of beryllium workers”. Am J Ind Med 2001; 40:286-8. MacMahon B. The epidemiological evidence on the carcinogenicity of beryllium in humans. J Occup Med 1994; 36:15-24. IARC. Beryllium, cadmium, mercury and exposures in the glass manufacturing industry. Monograph 58. Lyon: IARC; 1993. Thun MT, Schnorr TM, Smith AB, Halperin W. Mortality among a cohort of US cadmium production workers – an update. J Natl Cancer Inst 1985; 74:325-33. Stayner L, Smith R, Thun M, Schnorr T, Lemen R. A doseresponse analysis and quantitative assessment of lung cancer risk and occupational cadmium exposure. Ann Epidemiol 1992; 2:177-94. Sorahan T, Lancashire RJ. Lung cancer mortality in a cohort of workers employed at a cadmium recovery plant in the United States: an analysis with detailed job histories. Occup Environ Med 1997; 54:194-201. Sorahan T, Lister A, Gilthorpe MS, Harrington JM. Mortality of copper cadmium alloy workers with special reference to lung cancer and non-malignant diseases of the respiratory system, 1946-92. Occup Environ Med 1995; 52:804-12. Jarup L, Bellander T, Hogstedt C, Spang G. Mortality and cancer incidence in Swedish battery workers exposed to cadmium and nickel. Occup Environ Med 1998; 55:755-9. Marsh G, Enterline P, Stone R, Henderson V. Mortality among a cohort of US man-made mineral fiber workers: 1985 follow-up. J Occup Med 1990; 32:594-604. Boffetta P, Saracci R, Andersen A, et al. Cancer mortality among man-made vitreous fiber production workers. Epidemiology 1997; 8:259-68. Lockey J, Lemasters G, Rice C, McKay R, Gartside P. A retrospective morbidity, mortality and nested case-control study of the respiratory health of individuals manufacturing refractory ceramic fiber and RCF products. Submitted to the

742 Malignancies of the Respiratory Tract and Pluera

131.

132.

133.

134.

135.

136.

137. 138.

139.

140.

141.

142.

143.

144.

145.

146.

147. 148.

149.

150.

151.

152.

Refractory Ceramic Fiber Coalition (RCFC), Washington DC, 1993. Marsh GM, Youk AO, Stone RA, et al. Historical cohort study of US man-made vitreous fiber production workers: I. 1992 fiberglass cohort follow-up: initial findings. J Occup Environ Med 2001; 43:741-56. Marsh G, Stone R, Youk A, et al. Mortality among US rock wool and slag wool workers: 1989 update. J Occup Safety Health Aust NZ 1996; 12:297-312. Wong O, Foliart D, Tent L. A case-control study of lung cancer in a cohort of workers potentially exposed to slag wool fibers. Br J Ind Med 1991; 48:818-24. Chiazze L, Watkins D, Fryar C, Kozone J. A case-control study of malignant and non-malignant respiratory disease among employees of a fibreglass manufacturing facility II: exposure assessment. Br J Ind Med 1993; 50:717-25. Walker AM, Maxim LD, Utell M. Risk analysis for mortality from respiratory tumors in a cohort of refractory ceramic fiber workers. Regul Toxicol Pharmacol 2002; 35:95-104. International Agency for Research on Cancer. Man-made mineral fibres and radon. IARC Monogr Eval Carcinog Risks Hum 1988; Vol. 43. Katsouyanni K, Pershagen G. Ambient air pollution exposure and cancer. Cancer Causes Control 1997; 8:284-91. Dockery DW, Pope CA 3rd, Xu X, et al. An association between air pollution and mortality in six US cities. N Engl J Med 1993; 329:1753-9. Pope CA 3rd, Thun MJ, Namboodiri MM, et al. Particulate air pollution as a predictor of mortality in a prospective study of US adults. Am J Respir Crit Care Med 1995; 151:669-74. Pope CA 3rd, Burnett RT, Thun MJ, et al. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 2002; 287:1132-41. Wagner JC, Sleggs CA, Marchand P. Diffuse pleural mesothelioma and asbestos exposure in the North Western Cape Province. Br J Ind Med 1960; 17:260-71. Selikoff I, Hammond E, Seidman H. Mortality experience of insulation workers in the US and Canada, 1943-1976. Ann N Y Acad Sci 1979; 330:91-116. Anderson HA, Lillis R, Daum SM, Selifoll IJ. Asbestosis among household contacts of asbestos factory workers. Ann N Y Acad Sci 1979; 330:387-99. Rees D, Myers J, Goodman E, et al. Case-control study of mesothelioma in South Africa. Am J Ind Med 1999; 35:213-22. Hodgson JT, Darnton A. The quantitative risks of mesothelioma and lung cancer in relation to asbestos exposure. Ann Occup Hyg 2000; 44:565-601. McDonald JC, Armstrong BG, Edwards CW, et al. Casereferent survey of young adults with mesothelioma: I. Lung fibre analysis. Ann Occup Hyg 2001; 45:513-18. Smith AH, Wright CC. Chrysotile asbestos is the main cause of pleural mesothelioma. Am J Ind Med 1996; 30:252-66. Stayner L, Dankovic D, Lemen R. Occupational exposure to chyrsotile asbestos and cancer risk: a review of the amphibole hypothesis. Am J Pub Health 1995; 86:179-86. Metintas M, Hillerdal G, Metintas S. Malignant mesothelioma due to environmental exposure to erionite: follow-up of a Turkish emigrant cohort. Eur Respir J 1999; 13:523-6. Marsh GM, Gula MJ, Youk AO, et al. Historical cohort study of US man-made vitreous fiber production workers: II. Mortality from mesothelioma. J Occup Environ Med 2001; 43:757-66. Rodelsperger K, Jockel KH, Pohlabeln H, Romer W, Woitowitz HJ. Asbestos and man-made vitreous fibers as risk factors for diffuse malignant mesothelioma: results from a German hospital-based case-control study. Am J Ind Med 2001; 39:262-75. Sunderman FW Jr, Morgan LG, Andersen A, Ashley D, Forouhar FA. Histopathology of sinonasal and lung cancers in nickel refinery workers. Ann Clin Lab Sci 1989; 19:44-50.

153. Gordon I, Boffetta P, Demers PA. A case study comparing a meta-analysis and a pooled analysis of studies of sinonasal cancer among wood workers. Epidemiology 1998; 9:518-24. 154. Mannetje A, Kogevinas M, Luce D, et al. Sinonasal cancer, occupation, and tobacco smoking in European women and men. Am J Ind Med 1999; 36:101-7. 155. Luce D, Gerin M, Leclerc A. Sinonasal cancer and occupational exposure to formaldehyde and other substances. Int J Cancer 1993; 53:224-31. 156. Partanen T. Formaldehyde exposure and respiratory cancer – a meta-analysis of the epidemiologic evidence. Scand J Work Environ Health 1993; 19:8-15. 157. Luc D, Leclerc A, Begin D. Sinonasal cancer and occupational exposures: a pooled analysis of 12 case-control studies. Cancer Causes Control 2002; 13:147-57. 158. Zhu K, Levine RS, Brann EA, Hall HI, Caplan LS, Gnepp DR. Case-control study evaluating the homogeneity and heterogeneity of risk factors between sinonasal and nasopharyngeal cancers. Int J Cancer 2002; 99:119-23. 159. Goodman M, Morgan RW, Ray R, Malloy CD, Zhao K. Cancer in asbestos-exposed occupational cohorts: a meta-analysis. Cancer Causes Control 1999; 10:453-65. 160. Kraus T, Drexler H, Weber A, Raithel HJ. The association of occupational asbestos dust exposure and laryngeal carcinoma. Isr J Med Sci 1995; 31:540-8. 161. Imbernon E, Goldberg M, Bonenfant S. Occupational respiratory cancer and exposure to asbestos: a case-control study in a cohort of workers in the electricity and gas industry. Am J Ind Med 1995; 28:339-52. 162. Gustavsson P, Jakobsson R, Johansson H. Occupational exposures and squamous cell carcinoma of the oral cavity, pharynx, larynx, and oesophagus: a case-control study in Sweden. Occup Environ Med 1998; 55:393-400. 163. Eisen EA, Tolbert PE, Hallockj MF, Monson RR, Smith TJ, Woskie SR. Mortality studies of machining fluid exposure in the automobile industry. III: A case-control study of larynx cancer. Am J Ind Med 1994; 26:185-202. 164. Wortley P, Vaughan TL, Davis S, Morgan MS, Thomas DB. A case-control study of occupational risk factors for laryngeal cancer. Br J Ind Med 1992; 49:837-44. 165. Russi M, Dubrow R, Flannery JT, Cullen MR, Mayne ST. Occupational exposure to machining fluids and laryngeal cancer risk: contrasting results using two separate control groups. Am J Ind Med 1997; 31:166-71. 166. Checkoway H, Heyer NJ, Demers PA. An updated mortality follow-up study of Florida phosphate industry workers. Am J Ind Med 1996; 30:452-60. 167. Steenland K. Laryngeal cancer incidence among workers exposed to acid mists (United States). Cancer Causes Control 1997; 8:34-8. 168. Soskolne CL, Jhangri GS, Siemiatycki J, et al. Occupational exposure to sulfuric acid in southern Ontario, Canada, in association with laryngeal cancer. Scand J Work Environ Health 1992; 18:225-32. 169. Burch JD, Howe GR, Miller AB, Semenciw R. Tobacco, alcohol, asbestos, and nickel in the etiology of cancer of the larynx: a case-control study. J Natl Cancer Inst 1981; 67:1219-224. 170. Shannon HS, Julian JA, Roberts RS. A mortality study of 11,500 nickel workers. J Natl Cancer Inst 1984; 73:1251-8. 171. Roberts RS, Julian JA, Muir DC, Shannon HS. A study of mortality in workers engaged in the mining, smelting, and refining of nickel. II: Mortality from cancer of the respiratory tract and kidney. Toxicol Ind Health 1989; 5:975-93. 172. Strauss G, Gleason R, Sugarbaker D. Chest X-ray screening improves outcome in lung cancer: a reappraisal of randomized trials on lung cancer screening. Chest 1995; 107(Suppl):270-9. 173. Strauss GM, Gleason RE, Sugarbaker DJ. Screening for lung cancer. Another look; a different view. Chest 1997; 111:754-68.

Prevention of Respiratory Cancer 743 174. Salomaa E, Liippo K, Taylor P, et al. Prognosis of patients with lung cancer found in a single chest radiograph screening. Chest 1998; 114:1514-18. 175. Tossavainen A. International expert meeting on new advances in the radiology and screening of asbestos-related diseases. Scand J Work Environ Health 2000; 26:449-54. 176. Sone S, Takashima S, Li F, et al. Mass screening for lung cancer with mobile spiral computed tomography scanner. Lancet 1998; 351:1242-5. 177. Sone S. Lung cancer screening using mobile low-dose computed tomography: results from Nagano project in Japan. In: Proceedings of an international expert meeting on new advances in radiology and screening of asbestos-related diseases. Helsinki: Finnish Institute of Occupational Health; 2000:33-46. People and Work Research Reports, no 36. 178. Henschke C, McCauley D, Uankelevitz D, et al. Early lung cancer action project: overall design and finding from baseline screening. Lancet 1999; 354:99-105. 179. Henschke C. Early lung cancer action project: findings on baseline and annual repeat screening CT. In: Proceedings of an international expert meeting on new advances in radiology and screening of asbestos-related diseases. Helsinki: Finnish Institute of Occupational Health; 2000:31-2. People and Work Research Reports, no. 36. 180. Vehmas T, Kivisaari L, Zitting A, et al. Computed tomography (CT) and high resolution CT for the early diagnosis of lung and pleural disease in workers exposed to asbestos: Finnish experiences. In: Proceedings of an

181.

182.

183.

184.

185. 186.

187.

international expert meeting on new advances in radiology and screening of asbestos-related diseases. Helsinki: Finnish Institute of Occupational Health; 2000:53-6. People and Work Research Reports, no 36. The Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study Group. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med 1994; 330:1029-35. Omenn GS, Goodman GE, Thornquist MD, et al. Risk factors for lung cancer and for intervention effects in CARET, the Beta-Carotene and Retinol Efficacy Trial. J Natl Cancer Inst 1996; 88:1550-9. Omenn GS. Chemoprevention of lung cancer is proving difficult and frustrating, requiring new approaches. J Natl Cancer Inst 2000; 92:959-60. Hennekens CH, Buring JE, Manson JE, et al. Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. N Engl J Med 1996; 334:1145-9. Ziegler RG, Mayne ST, Swanson CA. Nutrition and lung cancer. Cancer Causes Control 1996; 7:157-77. Mayne ST, Cartmel B, Baum M, et al. Randomized trial of supplemental beta-carotene to prevent second head and neck cancer. Cancer Res 2001; 61:1457-63. Garewal HS, Katz RV, Meyskens F, et al. Beta-carotene produces sustained remissions in patients with oral leukoplakia: results of a multicenter prospective trial. Arch Otolaryngol Head Neck Surg 1999; 125:1305-10.

30.2 Lymphohematopoietic Malignancies Howard M Kipen, Daniel Wartenberg Lymphohematopoietic (LH) malignancies are a heterogeneous group of neoplasms that arise from the clonal malignant transformation of various cell lines of the blood and lymphoid tissues. The transformed cell lines proliferate initially in the lymphoid tissues or bone marrow, where they may disturb normal immune function and hematopoiesis. Eventually, malignant cells spill over into the peripheral blood and other tissues, where traditional clinical syndromes become apparent. All LH neoplasms combined account for about 8% of the 1,300,000 new malignancies expected annually in the US. A 15% to 25% male predominance is evident for all subgroups (see Table 30.2.1). Leukemic clones are characterized by an arrest in maturation of the multipotent stem cell, which is now believed to be the effect of a prior genetic or chromosomal alteration. Although the malignant clone is not truly hyperplastic, its growth is unregulated and becomes independent of homeostatic feedback controls. The particular LH neoplasm that affects a given individual depends on the level of stem cell development at which the maturation block has occurred. For example, developmental blockade at the pluripotent stem cell level causes aplastic anemia (see Chapter 21). If the block occurs a stage later, acute myelogenous leukemia (AML) results. Later stages of maturation arrest result in progressively more differentiated cells up through acute megakaryocytic leukemia. In chronic myelogenous leukemia (CML), relatively mature, although dysfunctional, cells are produced by an incomplete maturation block; however, the cells from this line typically become unstable due to subsequent development of an earlier block, with eventual proliferation of promyelocytes and blasts, leading to crisis. Acute and chronic lymphocytic leukemias (ALL, CLL) are disorders of committed lymphopoietic stem cells, originating in lymphoid precursors of bone marrow, thymus, and lymph nodes. The solid tumors of the LH system, sometimes known as B-cell immunoproliferative disorders, include multiple myeloma (MM), Hodgkin’s disease (HD), and non-Hodgkin’s lymphoma (NHL). LH neoplasms are often regarded as sentinel health events, but as with all occupational and environmental disease, the exposure and risk factor history are of paramount importance because none of these conditions, nor their associated markers such as cytogenetic aberrations, is regarded as pathognomonic for a particular cause. Genetic abnormalities, both inherited and acquired, and environmental factors have been shown to be etiologically important.1,2 Families with increased incidence have been described, although in some cases, shared environmental influences have been present and certainly should always be sought. Identical twins have a high concordance for leukemia in the first year of life, but this concordance is not known to persist in later life, suggesting predominance of environmental factors. Individuals with some congenital disorders, such as Down’s, Bloom’s, Wiscott–Aldrich,

Fanconi’s, and Klinefelter’s syndromes, have an increased incidence of leukemia. Acquired genetic factors are discussed later. Variations by sex and race of the major LH neoplasms are shown in Table 30.2.1. Aplastic anemia from any cause is associated with a markedly increased risk for the development of acute leukemia. Previous treatment with alkylating agents, radiotherapy, or both, also predisposes an individual to various LH disorders, and therapeutic immunosuppression such as that accompanying transplantation produces an exceedingly high relative risk of NHL, over 40-fold within 2 years, with a latency as short as a few months.3

CLASSIFICATION The study of the distribution and causes of the LH neoplasms, as with all diseases, is limited by accurate nosology. As a unit, the LH disorders are a fairly welldelineated group of cancers comprising the following conditions.

THE ACUTE LEUKEMIAS Acute myelogenous leukemia Acute lymphocytic leukemia

THE LYMPHOMAS Hodgkin’s disease Non-Hodgkin’s lymphoma T-cell lymphoma (mycoses fungoides)

OTHER Multiple myeloma Myeloproliferative disorders (including chronic myelogenous leukemia, myelofibrosis with myeloid metaplasia, and polycythemia vera) Chronic lymphocytic leukemia A number of caveats are important. Nomenclatures within the LH neoplasms have changed frequently over the last three decades. The International Classification of Diseases (ICD) and clinical coding practices are not well synchronized with one another nor with trends in clinical diagnostic practice.1,4–6 Thus, deaths due to myelodysplastic syndrome (see Chapter 21) are sometimes classified as a type of AML and sometimes separately, and within the NHLs there have been numerous shifts in nomenclature, whereas the ICD has retained older terminology.2,7 Because in any population LH neoplasms are relatively rare, such inconsistency influences the relative incidence of specific LH cancers, and thus impedes epidemiologic recognition and investigation of etiologic associations. Consequently, the total number of LH neoplasms or leukemias based on death certificates usually is reliable, because LH tumors rarely are classified outside the overall leukemia and lymphoma rubric, but specific cell types still cannot be reliably ascertained from death certificates.

Laboratory Evaluation 745 Males

Leukemia Lymphocytic Acute Chronic Myelogenous Acute Chronic Lymphomas Hodgkin’s Non-Hogkin’s

Females

White

Black

White

Black

13.8 6.3 1.9 4.3 5.5 3.4 1.8 23.5 3.2 20.3

10.3 4.4 1.2 3.1 4.6 2.6 1.9 19.6 2.5 17.1

8.4 3.7 1.4 2.2 3.7 2.4 1.1 16.2 2.7 13.5

6.2 2.3 0.8 1.4 3.2 2.0 1.0 10.9 1.9 8.9

Table 30.2.1 Rates (per 100,000) of the major lymphohematopoietic malignancies, 1994–1998, based on nine registries, adjusted to 1970 US standard populations. (Data from Ries et al.86)

For practical reasons, in many epidemiologic analyses, the leukemias are considered as one. But the various LHs are a diverse group of conditions, both hematologically and etiologically. The emergence of various unifying understandings of pathogenesis such as oncogene activation adds some coherence to lumping together conditions whose distinguishing characteristics may be limited to subtle aspects of morphologic and functional cellular phenotype. Whereas virtually all LH tumors share the mechanistic similarity of one or more genes of lymphoid cellular origin having been abnormally activated and expressed, the mechanism may be a mutation, a translocation, a chromosome aberration, or acquisition of new viral DNA. Some have even hypothesized that a non-specific stimulus such as clonal hyperstimulation of B cells could eventuate in malignancy such as lymphoma or MM. The use of the broader term B-cell immunoproliferative disorders has been proposed for a combination of CLL, NHL, HD, and MM. The paradox is that, to the extent that the tumors are distinct entities with distinct sets of causes, imprecision in classification makes etiologic associations more difficult to observe with consistency, whereas to the extent that one cause may result in multiple distinct histologic abnormalities, splitting arbitrarily reduces epidemiologic power and the opportunity to observe statistically significant associations.

LABORATORY EVALUATION IN THE DIAGNOSIS OF LYMPHOHEMATOPOIETIC MALIGNANCIES OF OCCUPATIONAL OR ENVIRONMENTAL ORIGIN Leukemias usually are recognized when a white blood cell count is elevated in the presence of circulating blast cells in a symptomatic individual, although presentations with normal white blood cell counts or with cytopenias are not uncommon. Diagnosis is based on a finding of leukemic cells in either peripheral blood, bone marrow, or

extramedullary tissue. Examination of marrow classically shows an infiltrate of uniformly appearing blast cells, causing an overall hypercellular appearance with a reduction in normal formed elements. The therapeutically important distinction between AML and ALL is made by a combination of morphology (e.g., Auer’s rods in AML), reactions with cytochemical stains (e.g., TdT positive in ALL; lysosomal enzyme positive in AML), and cell surface immunophenotypic markers. CLL usually is associated with elevated numbers of apparently mature B lymphocytes in peripheral blood, although 5% have T-cell surface markers. Cytogenetic (karyotypic) characteristics are used to define subsets of therapeutic and prognostic significance within the major leukemic groups, and they also can be used to make etiologic inferences. Lymphomas are diagnosed and classified by histopathologic examination of resected tissue, because reliance on aspirates is associated with an unacceptably high error rate. Disagreements in classification between experienced pathologists using biopsy material is reported to be higher than for most types of solid tumors, although immunologic markers are increasingly applied. Immunophenotyping through use of monoclonal antibodies has shown that 60% of NHL is B-cell derived and 30% to 40% is of T-cell origin. Cytogenetic evaluation commonly reveals abnormalities, including translocations in proximity to known oncogenes such as c-myc, and such data are becoming more incorporated into diagnosis and classification. A number of studies have reported a statistically significant relationship between certain chromosome changes in AML and NHL and exposure to broad categories of occupational agents such as pesticides, solvents, petrochemicals, combustion products, and asbestos. This finding is consistent with the abnormal cytologic characteristics of secondary leukemia after cytotoxic and radiation therapy (see later), although such associations have been variably noted, with positive findings in Scandinavia and relatively more negative findings in the US and Canada. The most frequently implicated abnormalities are -5 and -7, which have been associated previously with leukemia occurring after prior administration of chemotherapy.8–12 This work is enabled and complicated by rapid advances in chromosome banding techniques, which now demonstrate some sort of visible chromosome rearrangement in nearly all cases of AML, in contrast to much lower rates in earlier years. The stage of the leukemia at the time of analysis also complicates the picture because the karyotype is known to evolve during the natural history of AML and other leukemias. Although these types of epidemiologic investigations are relatively new, they offer a tantalizing approach to the exploration and clarification of the relationship between endogenous carcinogenic factors such as oncogenes and activation by environmental factors. This approach assumes one can distinguish the chromosome changes due to the environment factor from those due to progression and transformation of the disease, and has not yet been widely applied to resolution of environmental etiology problems. A slightly different use of laboratory abnormalities relies on

746 Lymphohematopoietic Malignancies the possibility that chromosome changes such as sister chromatid exchange (SCE) may be predictive of neoplasms. Ethylene oxide, an animal carcinogen and probable human leukemogen, has been shown to cause increases in SCE in exposed workers,13 although not all studies have been positive.14–16 Benzene, a known human leukemogen, causes chromosomal abnormalities in exposed individuals.7,17 Of course, all of these chromosomal markers may be thought of as exposure indices, but much work remains to be performed to improve their specificity, and they must be interpreted with care in clinical settings.

carcinogenesis remain an area of scientific ferment. The very high risk of AML after high-dose benzene-induced aplastic anemia may be an example of the influence of the exposure rate, but there is no clear indication that the leukemogenic properties of benzene require clinical marrow injury. Study of US and Chinese cohorts has led to some convergence of the dose–response relationships between benzene exposure and AML, extending evidence for causal relationships to exposures reported to be below 40 parts per million-years (ppm-years) of cumulative exposure.7

ACUTE MYELOGENOUS LEUKEMIA OR ACUTE NON-LYMPHOCYTIC LEUKEMIA

Clinical features

Epidemiology The 31,500 annual incident leukemias in the US comprise about 3% of all malignant neoplasms. Of these, myeloid types represent about 45%, with lymphoid types about 37%, and the remainder other and unspecified. Ionizing radiation, benzene, and cytotoxic drugs are well-established acute leukemogens. Each is discussed in detail later in the chapter. In all three cases, investigations have progressed from whether they are leukemogens to questions of dose–response, mechanism of action, and risks posed by lower-dose environmental exposure. Ethylene oxide is now classified as a human carcinogen, and suspected to cause AML in addition to other LH neoplasms (see Chapter 40); styrene, 1,3-butadiene, vinyl chloride, paints, and nitrites have been reported causes. Industries that have used any of these agents (e.g., printing, rubber manufacturing, and painting) may be implicated, although benzene exposure is often implicated as the most potent among the mixtures. A recent case–control study confirmed well-known risks and raised new questions.18 It is an accepted generalization that the leukemias have much shorter latencies than the 10- to 30-year periods accepted for solid tumors. Acute leukemia risk in atomic bomb radiation survivors peaked at 2 to 5 years after detonation and declined after 10 years. After chemotherapy, the peak incidence is at 5 to 8 years and there appears to be a decline in risk by 10 to 15 years (as opposed to the persistent and progressively higher increased risk for solid tumors). Based on these high-dose, high-intensity exposure situations, some investigators have inferred that risks from environmental exposures to benzene, ionizing radiation, and perhaps other leukemogens diminish after a certain latency has been achieved. There are no clear epidemiologic data to support this. Dose–response relations for acute leukemogens remain controversial as well. It is possible that the pattern and rate of exposure, as well as simple cumulative dose according to Haber’s law, influence the risk of leukemia from ionizing radiation, the risk increasing as exposure is concentrated in time. This finding has not been clearly demonstrated for chemical leukemogens, and appropriate models of

Of potentially great mechanistic significance for epidemiologic studies of occupational and environmental agents is the marked increase in clonal cytogenetic abnormalities, with 7q- or 5q- detectable in 70% of AMLs secondary to chemotherapy versus 10% of de-novo AMLs.9 These leukemias secondary to alkylating agents or radiotherapy are usually characterized by a latency period of 5–10 years which may be preceded by a myelodysplastic phase. These particular chromosomal markers are also associated with particularly poor prognoses. The converse finding that the aberrations most common in de-novo AML are rare in the secondary cases lends credence to the potential causal mechanistic specificity of cytogenetic evaluation. Interestingly, some feel that de-novo AML in older adults bears cytogenetic similarities to secondary AMLs, leading to the hypothesis that AML and myelodysplastic syndrome developing in the elderly may reflect cumulative mutagenic environmental exposures in genetically susceptible individuals.19 As stated earlier, field epidemiologic studies so far have yielded inconsistent results with respect to environmental and occupational agents and their specific associations with characteristic cytogenetic markers.10–12,20

Diagnosis, prognosis, and treatment Cytotoxic chemotherapy with transfusion support has led to increasing numbers of initial remissions, currently 60% to 80%. Maintenance therapy is needed for some months. Between 10% and 30% of patients are free of disease at 5 years and apparently are cured. Secondary leukemias are becoming an increasingly prominent issue as survival rates increase.9

Screening There is no practical approach to screening or early diagnosis for either prevention or reduction of morbidity. Screening based on cytogenetic abnormalities in peripheral blood lymphocytes or DNA adducts has been suggested but is not validated and should be considered only in experimental settings.21 Routine complete blood counts (CBCs) to detect benzene-induced marrow depression, and urinary phenols to detect excessive benzene exposure are not likely to be sensitive enough to be useful at current levels of exposure in the developed world,

Non-Hodgkin’s Lymphoma 747 except for indicating serious breaches of industrial hygiene, leading to overexposures greater than 10-fold the permissible exposure limit (PEL).

ACUTE LYMPHOCYTIC LEUKEMIA Epidemiology ALL is primarily a disease limited to the first two decades of life, with peak incidence at ages 2 to 9 years, although there is a smaller late peak in incidence after age 60. It is the type of leukemia most closely associated with several congenital disorders characterized by chromosomal abnormalities. However, because ALL does not affect the majority of patients with Down’s, Fanconi’s, Bloom’s, and Klinefelter’s syndromes, an environmental contribution to cause is still strongly implied. Studies of the effect of paternal and maternal occupations on childhood leukemia risk usually have not been cell-type specific (30% of childhood leukemias are ANLL) and, in fact, usually combine leukemias and lymphomas. Paternal occupations linked to the combined tumor types include nuclear industry workers, motor vehicle mechanics, hydrocarbon-exposed occupations, and medical and social service workers. In mothers, leukemia is linked to employment as a pharmacist, textile worker, chemical-exposed worker, hydrocarbon-exposed worker, and work in a domestic, hotel, or catering job (for a review, see Colt & Blair22). Solvent exposures at home have also been implicated.23 The data for the nuclear industry have the most biologic plausibility and consistency with worker outcomes. Although some studies have suggested that paternal exposure through employment at nuclear facilities increases childhood leukemia risk, alternative hypotheses have also been offered for the observed cluster of childhood leukemias around the Seascale nuclear facility in England (see below).

Clinical features ALL typically is exhibited by constitutional symptoms, easy bruising, and lymphadenopathy due to infiltration and replacement of bone marrow and lymphatic tissue by an abnormal lymphoid precursor. This usually is a null cell but occasionally has T-cell or, very rarely, B-cell markers.

Diagnosis, prognosis, and therapy Diagnosis is arrived at by examination of peripheral blood and bone marrow for excessive lymphoblasts. Combination chemotherapy currently results in a 4- to 5-year disease-free survival with a significant long-term cure rate.

Screening No screening is indicated.

HODGKIN’S DISEASE Epidemiology HD (lymphoma) is a tumor of histiocytic or reticulum cells, which is pathologically distinguished by the presence of Reed–Sternberg cells, as well as the somewhat atypical clinical features of an infection. There are about 7400 new cases annually in the US, with an average age at diagnosis of 32 years. It has the unusual epidemiologic feature of a pronounced bimodal age distribution with peaks around age 20 and age 70 in developed countries. In contrast to NHL and MM, the incidence is not linearly increased with age. Although disease presentation patterns in young adults still suggest an infectious pattern, this is not suggested in older patients, and some occupational associations have been discovered, although not with consistency. The roles of ionizing radiation, benzene, and other environmental agents are considered later; however, a recent review emphasized the paucity of confirmed association with environmental agents.24

Clinical presentation HD usually is characterized by lymph node enlargement, either asymptomatic or associated with constitutional symptoms. Staging is performed based on the extent of nodal and extranodal involvement, and whether or not disease is present on both sides of the diaphragm. A functional T-lymphocyte defect usually is present, and normochromic-normocytic anemia is often present.

Diagnosis and treatment Accurate diagnosis usually requires a surgical biopsy because aspiration rarely is adequate. Thorough exploration using physical examination, radiologic, and nuclear medicine modalities to detect all possible areas of involvement is required for accurate staging. Although radiotherapy is difficult and demanding owing to shielding requirements to reduce acute complications as well as risk of secondary malignancies and hypogonadism, over 70% of patients are now considered curable by radiotherapy, with or without combination chemotherapy.

Screening There is no applicable screening for HD.

NON-HODGKIN’S LYMPHOMA Classification NHL is characterized by monoclonal neoplasms associated with proliferation of lymphocytes, macrophages, and their precursor and derivative cells. They are dominantly of B-cell phenotypic lineage, although some may have mixed B and T markers, and T-cell types may be particularly malignant. Classification has rapidly and competitively evolved in this field from the histologically based Rappaport

748 Lymphohematopoietic Malignancies Classification of 1966 to the immunologically based Lukes–Collins Classification of 1973 to 1974 and, in 1982, to the more clinically and prognostically based Working Formulation of NHLs for Clinical Usage. Further classifications have been proposed in an attempt to delineate individual diseases likely to have distinct etiologies.5,25 The implications for studies of etiology are difficult to specify but will surely complicate assessment of causal factors for a tumor observed today in light of epidemiologic information collected based on earlier nomenclatures, including the pre-Rappaport data, on which much of current epidemiologic understanding is based.

Epidemiology NHL accounts for approximately 3% of new cancer cases annually in the US and has an annual incidence rate of about 16/100,000 based on 1994 to 1998 Surveillance, Epidemiology, and End Results (SEER) tumor registry estimates. The incidence of NHL is 50% higher in males than in females, and it is much higher in persons beyond the fifth decade. These characteristics plus a steep secular rise have been used to argue for etiologic attribution to occupational and chemical exposures, although detailed analysis does not yield a completely satisfactory explanation.26 Except for Burkitt’s lymphoma, which is highly prevalent in equatorial Africa, rates are higher in whites than in blacks, and they also increase with socioeconomic status in most studies. Clusters of T-cell lymphoma in southwestern Japan and the Caribbean, where 10% to 15% of the population is seropositive, are strongly associated with endemic human T-cell lymphocytic virus (HTLV-1) infection.27,28 Burkitt’s is strongly associated with EB virus. Known risks apart from socioeconomic status include hydantoin therapy and immune suppression. The latter may be either a primary (genetic) defect, secondary to drug therapy (e.g., prophylaxis for transplant rejection), or acquired from infection (HIV, HTLVs), autoimmune disorders, or therapeutic irradiation. Although impaired immune surveillance has been proposed to explain the increased incidence associated with the above-mentioned factors (up to 40- to 100-fold in the case of immunosuppressive therapy), there are other hypotheses of the origin of B-cell neoplasms that may be more germane to the role played by occupational and environmental chemical exposures. This line of reasoning maintains that lymphomas may develop as a result of imbalanced immunoregulation rather than deficient immune surveillance. Whereas the immune impairment is clinically apparent in the congenital or therapeutic immunodeficiencies, the impairments associated with NHL developing in the general population are hypothesized to be much more subtle. It is hypothesized that individual and, perhaps, idiosyncratic reactions to common physical, chemical, and viral agents may lead to lymphoma, without frank immune deficiency. Oncogene activation or immune gene dysregulation due to mutagens and clastogens also may play a role. Occupations for which the risk of NHL has been reported to be elevated include rubber workers, veterinarians, uranium miners,

asbestos-exposed workers, lumberjacks, metal workers, female textile workers, chemists, benzene-exposed workers, and farmers (see discussion later).29,30

Clinical presentation The clinical presentation of occupation-related NHL is not known to be different from de-novo cases.

Diagnosis, prognosis, and therapy Examination of lymph node material usually is preferable to examination of involved extranodal tissue. Selection of therapy is more dependent on histologic type (Burkett’s, follicular, diffuse) than on staging, in contrast to the dominant importance of staging for HD. The dominant favorable prognostic indicator is the presence of a follicular pattern in the nodal architecture, because patients with this characteristic finding follow a more indolent course than do those with a diffuse pattern of equivalent cytologic characteristics.

Screening No screening techniques are clinically applicable to this disease.

MYCOSIS FUNGOIDES (see Chapter 30.9) This rare cutaneous T-cell lymphoma has been linked to metals, solvents, arsenic, and bismuth in case reports. Industries such as petrochemical, rubber, metals, machinery production, printing, and textiles also have been suggested to have an association. The construction industry, too, was implicated in one study, which showed a particularly high risk among machinists. Viral etiologies have also been considered.31

Clinical presentation, diagnosis, prognosis, and treatment The clinical presentation, diagnosis, prognosis, and treatment of occupation-related mycosis fungoides are not known to be any different from those of de-novo cases.

Screening Although, in general, skin cancer screening may be indicated, especially for outdoor workers, mycosis fungoides is too rare to warrant special attention.

MULTIPLE MYELOMA Epidemiology MM, with 15,000 new cases annually in the US, accounts for 1% of all malignant diseases and 15% of the LH

Myeloproliferative Disorders 749 malignancies that are relevant to this chapter. Its annual incidence is about twice that for HD. Less than 2% of MM occurs in patients younger than 40 years old, and the peak incidence occurs in patients 70 to 80 years old. Rates in non-whites are twice those of whites. Prior to 1949, it was classified as a bone tumor. In the US and other industrialized countries, there has been a striking increase in the age-standardized incidence over the past three to four decades.32 Although some investigators maintain that this increase is attributable to diagnostic bias, it also is compatible with and suggestive of environmental and occupational factors as etiologic agents for MM. Ionizing radiation and occupational benzene exposure are well linked to increased myeloma risk and are discussed later. Excess risks, mostly less than twofold, also have been reported for petroleum refinery and petrochemical workers, farmers, woodworkers, food workers, printers, and those exposed to arsenic, lead, cutting oils, pesticides, and paint-related compounds.31

Clinical presentation The clinical presentation of patients with occupationrelated MM is no different than in de-novo cases.

Diagnosis, prognosis, and treatment The classic diagnostic triad is marrow plasmacytosis (>10%), lytic bone lesions, and a monoclonal immunoglobulin spike in serum or urine. Although 10% of patients have an indolent and very slowly progressive course, the vast majority require both chemotherapy for the myeloma and supportive care for complications such as hypercalcemia, renal failure, infections, and anemia. Fifteen percent of patients die within 3 months of diagnosis, with a subsequent death rate of about 15% per year. Typically, there is a 2- to 5-year chronic course before refractory cytopenias and associated complications develop. Five percent of patients die with acute leukemia; whether this finding might represent an interaction with some of the common (leukemogenic) etiologic agents is unknown.

Screening There is no screening relevant to the occupational setting.

MYELOPROLIFERATIVE DISORDERS Myeloproliferative disorders are neoplastic disorders of the multipotent hematopoietic stem cell. Included are CML, polycythemia vera, essential thrombocytosis, agnogenic myeloid metaplasia with myelofibrosis, paroxysmal nocturnal hemoglobinuria (PNH), and, for purposes of this discussion, myelodysplastic syndrome, frequently called preleukemia or smoldering leukemia. Other than CML and myelodysplastic syndrome, the diseases run a prolonged

course over many years. We discuss CML, myelofibrosis, and polycythemia vera in detail because of relatively stronger links to environmental agents, although environmental and occupational causes of these disorders are not well studied. Myelodysplastic syndrome may be considered to have the same etiologic risk factors as AML (see also Chapter 21).

Chronic myelogenous leukemia Epidemiology Ionizing radiation exposure is clearly linked to the development of CML33 and benzene is linked to CML but slightly less clearly than it is to AML.34 Rubber workers and electrical workers have been found to be at increased risk for this particular malignant phenotype,35–37 electrical workers perhaps due to non-ionizing radiation exposure. The incidence of CML is roughly comparable to that for AML during the first five decades (about 50/100,000) and then doubles after age 60, as opposed to AML, which increases over sixfold.38

Clinical features CML is clinically characterized by an accumulation of both immature and mature granulocytic cells, leading to suppression of normal myelopoiesis and erythropoiesis. Twenty percent of cases are diagnosed by an elevated white blood cell (WBC) count in the absence of symptoms. Lymphadenopathy is rare. CML is characterized in 90% of cases by the well-known cytogenetic marker, the Philadelphia chromosome, resulting from a reciprocal translocation between the long arms of chromosomes 22 and 9. The breakpoints are near the proto-oncogenes, c-sis on 22 and c-abl on 9. Although this abnormality persists in all phases of the illness, the blastic phase usually is characterized by additional chromosome abnormalities, including aneuploidy, reflecting the more malignant character of this phase.39

Diagnosis, prognosis, and treatment The high WBC count, marrow aspirate with a Philadelphia chromosome, and depressed leukocyte alkaline phosphatase activity are all considered diagnostic hallmarks of CML. Therapy for the chronic phase usually includes moderate doses of alkylating agents, most commonly busulfan, whose most serious complication is prolonged myelosuppression. Transplant approaches have been added over the past decade.40

Screening No screening measures are applicable.

Myelofibrosis with myeloid metaplasia Epidemiology Myelofibrosis with myeloid metaplasia (agnogenic myeloid metaplasia) is another clonal stem cell disor-

750 Lymphohematopoietic Malignancies der resulting in marrow fibrosis in association with proliferation of granulocytes, erythrocyte precursors, and megakaryocytes in extramedullary sites. The incidence appears to increase in individuals with prior ionizing radiation exposure (including radioactive phosphorus [32P]-treated polycythemia vera) and has been reported to occur with benzene exposure41

Clinical features Myelofibrosis usually is associated with constitutional symptoms and anemia (over 50% of patients are anemic at diagnosis), with virtually all patients developing anemia at some time during the course of the disease, due to decreased erythropoiesis, splenic pooling, and shortened red blood cell survival. High WBC counts are common, leading to a difficulty in differentiating this condition from CML, in which 10% to 15% of patients also may have marrow fibrosis.

Diagnosis, treatment, and prognosis Bone marrow biopsy is required. Polycythemia vera and CML are the major considerations in the differential diagnosis. No therapy has been demonstrated to prolong life, but androgens have been used for anemia, myelosuppressive therapy for marked splenomegaly or thrombocytosis, and allopurinol to control hyperuricemia from tissue breakdown. The natural history is prolonged, with a median survival rate of 5 years from diagnosis. One-quarter of patients live 15 years after diagnosis.42

Polycythemia vera Epidemiology Polycythemia vera is characterized by elevated hemoglobin levels in a setting of increased production of all marrow cell lines. Splenomegaly is common. It is a disease of adults, with a slight male predominance and no known familial aggregation. Information on occupational or environmental etiology is lacking. However, because this is a clonal myeloproliferative disorder, often associated with chromosomal abnormalities, myelotoxic factors may well be operative.

Clinical features No differences have been found in cases suspected to be caused by environmental factors versus those arising de novo.

Diagnosis, prognosis, and treatment The diagnosis, prognosis, and treatment are independent of etiologic consideration. In cases treated with myelosuppressive agents or 32P, cytogenetic abnormalities may increase from the approximately 10% rate reported for untreated patients.

Screening No screening measures are applicable.

CHRONIC LYMPHOCYTIC LEUKEMIA This disease of later life, characterized by high circulating levels of mature-looking lymphocytes, is actually a variant of well-differentiated NHL. Generally, it has a favorable prognosis and often requires no specific therapy. Although some environmental associations have been noted, this disorder has not been convincingly linked to any myelotoxic agent. Notably, sufficient data exist to rule out an association with ionizing radiation, at least in Japanese and perhaps other Asian populations, based on data from the atomic bomb.

OCCUPATIONAL AND ENVIRONMENTAL RISK FACTORS FOR LYMPHOHEMATOPOIETIC MALIGNANCIES There are four well-accepted risk factors for LH malignancies: • ionizing radiation; • benzene inhalation; • agricultural work; and • cytotoxic and immunosuppressive drugs. Less clear-cut evidence is suggestive of an association between LH malignancies and certain non-agricultural occupations, exposure to electromagnetic fields (one type of non-ionizing radiation), and exposure to solvents, notably trichloroethylene (TCE) in drinking water. We consider each in turn.

Known causes Ionizing radiation (Table 30.2.2) The importance of ionizing radiation as an etiologic agent for LH malignancies has been known since the early part of this century from studies of radiologists. The most compelling evidence for this association has come from studies of survivors of the atomic bomb blasts in Hiroshima and Nagasaki and patients treated for ankylosing spondylitis. For both of these types of exposures, leukemias (other than CLL) were noted as early as 3 years after exposure, with peak incidence occurring 5 to 10 years after exposure, and additional excess cases were still being diagnosed for three decades after. There also is limited evidence for leukemia risk for occupational exposure to ionizing radiation. Studies of military personnel on maneuvers at a nuclear bomb test showed statistically elevated leukemia incidence and mortality, as do the most recent studies of workers at nuclear facilities. A review and pooled analysis of nuclear worker studies conducted by the International Agency for Research on Cancer43 found statistically significant excesses for leukemia and MM in more than one study of individual facilities. Their pooled analysis found statistically significant excess relative risks for leukemia excluding CLL and

Occupational and Environmental Risk Factors for Lymphohematopoietic Malignancies 751 Risk factor Ionizing radiation

Specific exposure Radiologists

Paternal occupation in nuclear facility Residential magnetic fields

Radiologist RR ~ 10 for leukemias, relative to other physicians Noted a positive response for leukemias Peak leukemia incidence 5 years after first course of treatment RR ~ 2.5 for leukemias SMR ~ 1.6 for leukemias RR ~ 2–6 for leukemias RR < 2; children’s OR elevated for ALL Local ecologic study; SIR ~ 1.5 for leukemias Broad ecologic study; no association found RR ~ 2–6 RR ~ 2–3

Occupational magnetic fields

Risk varies by occupation

Atomic bomb blasts Treatment for ankylosing spondylitis Nuclear bomb test workers Nuclear facility workers Naval shipyard workers Bomb test fallout Ingestion of radium-containing water Ingestion of radium-containing water Non-ionizing radiation

Comments

Reference March, 194483 Beebe et al., 197887 Court Brown & Doll, 196568 Caldwell et al., 198388 Cardis et al., 199543 Stern et al., 198689 Stevens et al., 199090; Archer, 198791 Lyman et al., 198548; O’Brien et al., 198792 Cohen, 199193 Gardner et al., 199052 Wartenberg, 200176; Ahlbom et al., 200077; Greenland et al., 200078 Savitz & Loomis, 199571; Floderus et al., 199372; Theriault et al., 199415

RR, relative risk; SIR, standardized incidence ratio; SMR, standardized mortality ratio.

Table 30.2.2 Selected studies demonstrating the association between radiation exposure and lymphohematopoietic malignancies

for MM. The leukemia effect was consistent with, but smaller than, values estimated from the studies of atomic bomb survivors reported in the BEIR V report from the National Research Council.44 All LH neoplasms studied in this Japanese population reveal a high incidence following radiation exposure, with the notable exception of CLL. The absence of positive data on this neoplasm from the atomic bomb data makes it one of the rare proven negative findings that can be asserted in environmental epidemiologic causation, or perhaps an indication that Japanese individuals are not susceptible for presumably genetic reasons. No safe threshold has been demonstrated, but direct evidence for radiation carcinogenesis has been derived almost entirely from studies of single high doses in excess of 50 rads. The case–control study with the best dosimetry for diagnostic X-ray exposure showed no increased risk for leukemia at highly fractionated doses up to 300 rads,45 although two earlier questionnaire studies did show an increased risk.46,47 Another suggested environmental cause of leukemia is ingestion of radium-containing groun dwater. One study that generated this hypothesis noted the geographic overlap of counties in Florida with both relatively high radium concentrations in the groun dwater and relatively high leukemia incidence rates.48 This observation is ecologic and particularly subject to confounding, which could explain the findings. Suggestive but not clear-cut results have followed in subsequent studies.49,50 Several studies have investigated the possible association between inhaled radon exposure and leukemia. The suggested mechanism is that radon inhalation may result in a small amount of irradiation to the bone marrow that could lead to leukemia. A recent weight-of-evidence review of 19 ecologic studies, six miner cohort studies, and eight

case–control studies concluded that while an ecologic association exists, it is not supported by the other studies and is unlikely to be etiologic.51 One unanticipated occupational risk for leukemia has been suggested by studies of a local cancer excess (cluster) near a nuclear plant in England. After adjusting for antenatal X-ray exposure, Gardner and associates found an association between childhood leukemia risk and paternal occupation in the nuclear plant, the risk increasing with larger cumulative paternal exposure.52 Although no such preconception risk was noted for survivors of the Hiroshima and Nagasaki atomic bomb blasts, the authors note that exposure to radiation from these atomic bombs was instantaneous, whereas that for workers is accumulated over many years. Laboratory studies of mice show similar paternal heritability of increased tumor risk, including tumors of the lung and lymphocytic leukemia. An alternative explanation involving an infectious etiology has been proposed for this and related clusters. Kinlen and colleagues have suggested that the children in this community might have been at higher than normal risk because they were resident in an isolated, rural town without frequent exposure to infectious agents that are common among children in less isolated locales, impeding the establishment of herd immunity.53–55 Mixing with populations from more urban areas, with wider exposure to viral agents, resulted in increased contacts between susceptible and infected individuals, promoting epidemics of underlying infections. Then, if the populations were large enough, these infectious epidemics could have resulted in extra cases of leukemia, an extremely rare response to a common and widespread infection. Regions with high population densities would be more likely to have established herd immunity, thereby preventing the infectious epidemics and subsequent leukemias. Many

752 Lymphohematopoietic Malignancies subsequent ecologic studies in other communities are consistent with this hypothesis. Greaves developed a related hypothesis proposing an infectious etiology for childhood ALL, suggesting that children were at greater risk if they lived in more affluent areas because with greater affluence came more effective hygiene and thus reduced exposure to infectious agents in earliest childhood.56–58 Like Kinlen, he argued that for susceptible children, in this case affluent, early childhood exposure to infectious agents underlay a substantial number of childhood leukemia cases. The Greaves hypothesis suggests that the increased risk is due to decreased immune responsiveness and thus exposure to any or many infectious agents would confer similar risk.56–58 More detailed studies and identification of specific infectious agents are needed for confirmation of any of these competing hypotheses.

Benzene (Table 30.2.3) Benzene is the best-recognized chemical leukemogen in the workplace and environment. Benzene’s profound abilities to depress marrow function were recognized around the turn of the last century and actually led to its unsuccessful use as an antileukemic agent, persisting into the 1930s. Based on series of cases observed from the 1940s through the 1960s, Italian and Turkish hematologists suggested that there may be a relationship between acute leukemia and occupational exposure to benzene-based inks and glues in the rotogravure printing and shoemaking industries, respectively.59–61 Exposures were not well quantified, but both industries are known to have resulted in substantial numbers of cases of marrow depression as well, implying substantial exposure. In the late 1970s and 1980s, cohort studies began to emerge, focusing on industrial rubber workers, refinery workers, and chemical workers. The most convincing and widely cited study involves a cohort of about 1200 rubber workers,62 who at the most recent follow-up had nine cases of myelogenous leukemia, compared with 2.7 expected.7 Clearly, the issue concerning the use of benzene now is not

Known Pancytopenia: aplastic anemia94 Acute myelogenous leukemia and variants7,63,95 Suspected Chronic myelogenous leukemia7,61 Chronic lymphocytic leukemia7,36,37,60,80 Paroxysmal nocturnal hemoglobinuria60 Multiple myeloma7,96 Reported Acute lymphoblastic leukemia59 Myelofibrosis and myeloid metaplasia97 Non-Hodgkin’s disease98 Thrombocythemia97 Table 30.2.3 Causal relation between benzene exposure and various hematologic disorders

whether it is a cause of leukemia but rather what level of risk accompanies a given level of exposure. Under a US Supreme Court requirement to show the magnitude of the effect of a reduction in the Occupational Safety and Health Administration (OSHA) benzene PEL from 10 ppm to 1 ppm, the National Institute for Occupational Safety and Health (NIOSH) performed a quantitative risk assessment on the previously described group of rubber workers.63 A dose–response was noted, including the projection that even 1 ppm of exposure for 40 years would lead to a 70% increase in leukemia risk (95% confidence interval [CI], 1.1–2.5). More recent observations of a large benzene-exposed cohort of Chinese workers from multiple industries seems to confirm this projection and suggests increased risk even below the 40 ppm-year cumulative level.7 Epidemiologic evidence from relatively low-dose occupations, such as service station attendants and auto mechanics, has suggested a relationship of benzene with leukemia, but the evidence is not regarded as conclusive.64 Beyond aplastic anemia and AML, there are varying levels of evidence supporting a role for benzene in the causation of numerous other LH malignancies, which are summarized in Table 30.2.3.

Agricultural exposures A large amount of epidemiologic literature addresses the incidence of LH neoplasms in farmers and in children exposed to pesticides.65,66 These studies partially overlap those of other agricultural workers, such as lumberjacks, forestry workers, wood mill workers, meat packers, and ranchers of poultry and cattle. Although various studies report excesses of leukemias, myelomas, HD, and NHL, there are discrepancies between well-studied geographic areas.67 Technically rigorous studies have reported high rates of NHL in agricultural workers exposed to phenoxyacetic acid herbicides, chlorophenols (e.g., 2,4,5-T and 2,4-D), or both in the US, Canada, and Sweden since 1977. However, other reliable studies from New Zealand have repeatedly failed to find these associations. A possible interpretation of the discrepancy is the role of various confounding factors such as co-carcinogens and oncogenes in the various populations, although no specific explanation has been documented. A reasonable conclusion at this time is that NHL is probably causally associated with exposure to the herbicides listed above; HD, MM, and leukemias cannot be linked specifically to herbicides, although the higher occupational incidence among agricultural workers as a whole is clear. Hopefully unresolved questions will emerge from ongoing NCI/NIOSH studies of agricultural health.

Medical exposures Although not strictly occupational or environmental, LH malignancies due to medical therapy provide a critical context for consideration of broader causes, both epidemiologically and mechanistically. CLL is the only LH malignancy (and one of only a handful of cancers) whose risk is not increased by therapeutic levels of ionizing radiation, at least not in the Japanese population.

Occupational and Environmental Risk Factors for Lymphohematopoietic Malignancies 753 The classic cohort68 of 14,558 young men who were therapeutically irradiated for ankylosing spondylitis in the 1930s showed 52 leukemias versus 5.5 expected, with the risk cresting at 4 to 5 years and diminishing 8 to 9 years after exposure.

Cytotoxic drugs Alkylating agents and, to a lesser extent, antimetabolite therapy significantly increase the risk for AML, often with the identifiable 5q- or 7q- markers, and myelodysplastic syndrome. More specific immunosuppressive therapies preferentially promote NHL, with the relative risk (RR) estimated at 40- to 100-fold.29 There is substantial cross-over of LH malignancy phenotype between the different classes of agents in spite of the useful distinction. Although there has been interest in the effects of cytotoxic agents on nurses, pharmacists, and others who are occupationally exposed, evidence of biologic effects has not been proved.69 Nonetheless, rigid precautions to limit exposures are well justified by the risk in medically exposed individuals.70

Suspect causes Non-ionizing radiation Concern also has been raised over the leukemia incidence among workers exposed to electric and magnetic fields (EMF).15,71,72 The risk was first documented in a case–control study of children living in homes with high magnetic fields, but, as yet, no mechanism for this risk has been established. Prompted by population-based studies, a series of occupational mortality studies have shown that workers in electrical, electronic, and telecommunications occupations and ham radio operators are at increased risk for ALL, AML, CML, and CLL. Because no viable mechanism has been postulated, there is much controversy over the definition of relevant exposure, although magnetic fields rather than electrical fields are implicated. Exposure metrics under consideration include peak, average, time-weighted average, and variability measures, although the data from different studies are contradictory (see Chapter 33.2). Two recent expert panel reviews of the EMF issue were conducted in the US.73,74 The first,73 under the sponsorship of the National Academy of Sciences, using a consensus process, reported that ‘no conclusive and consistent evidence shows that exposures to residential electric and magnetic fields produce cancer’, but also asserted that ‘an association between residential wiring configurations. . . and childhood leukemia persists in multiple studies. . .’. Interpretations of these statements varied. The second panel,74 convened by the National Institute of Environmental Health Sciences, using a majority rule process, concluded that ‘Extremely low frequency (ELF) EMF are possibly carcinogenic to humans (Group 2B)’. This was based principally on ‘the results of studies on childhood leukemia in residential environments and on CLL [chronic lymphocytic leukemia] in adults in occupational settings’. In addition, the participants

stated that the in-vitro and mechanistic data provide weak support based on studies at very high levels of exposures (>100 uT). To summarize this body of literature analytically, a variety of meta-analyses and pooled analyses were conducted. The most recent meta-analyses75,76 reported that the risk for childhood brain tumors was elevated, although not statistically significant, while the risk for leukemia was elevated and marginally statistically significant, particularly at the higher exposure cut points. There was some evidence that supported an exposure–response gradient. The two pooled analyses of childhood leukemia77,78 reported no overall effect but, again, elevated risks at the highest exposure categories. In sum, there still is much controversy over whether exposure to EMFs causes cancer. The strongest and most consistent adverse effect data from EMF exposure are found for childhood leukemia. Although the most common childhood cancer, this is a very rare disease, with an incidence rate of approximately 1.5/100,000. If EMFs are carcinogenic, the excess risk suggested by the combined studies for exposures four times greater than background are in the range 50%–100%. However, there is added concern because exposures are ubiquitous, and rare high residential exposures can be greater than 50 times background while rare high occupational exposures, although typically only short term, can reach 1000 to 40,000 times background.

Asbestos Beginning in the 1960s, case reports and case series have linked individuals with asbestosis to various immune disorders, both malignant and non-malignant. Clinical epidemiologic studies have linked asbestos exposure and asbestosis to increased prevalences of rheumatoid factor and other indices of immune activation and regulation. Exposure has been linked to NHL, CLL, and acute leukemias, although data are inconsistent and limited (for a review, see Becker et al.79). This remains an intriguing area for further investigation, possibly directed at co-factors necessary for an effect to emerge.

Rubber industry and other occupations Rubber industry workers have been reported to have increased risks for AML, CML, and MM.80 The extent to which this factor is independent of benzene exposure has not been definitively worked out due to the complexities of characterization of lifelong retrospective exposure. In this regard, the carcinogenic monomer 1,3 butadiene (see Chapter 41) may be an independent risk factor, although this is not clearly demonstrated.81,82 Chemical workers, machinists, and painters have been reported to have an increased incidence of LH malignancies and, in some cases, to have increased chromosomal abnormalities.18,83,84 Especially noteworthy is the suspected relationship between ethylene oxide and AML (see Chapter 40 and Steenland and associates85). Although the evidence is not as clear cut as for benzene or radiation exposure, risks have been demonstrated in various

754 Lymphohematopoietic Malignancies settings and some of the inconsistencies could be explained by exposure misclassification as well as by the more commonly posited causal relationships. Confounding benzene exposures are frequently a problem in the interpretation of this literature based on historical chemical exposures.

Trichloroethylene and other solvent exposures Several solvents have been investigated for possible carcinogenicity. TCE is an organic chemical that has been used in dry cleaning, for metal degreasing, and as a solvent for oils and resins. It has been shown to cause liver and kidney cancer in experimental animals. Evidence of excess cancer rates among occupational cohorts with the most rigorous exposure assessment is found for kidney cancer (relative risk [RR] = 1.7; 95% CI, 1.1–2.7), liver cancer (RR = 1.9; 95% CI, 1.0–3.4), and NHL (RR = 1.5; 95% CI, 0.9–2.3), as well as for cervical cancer, HD, and MM.65 However, since few studies isolate TCE exposure, results are likely confounded by exposure to other solvents and other risk factors. Several studies of communities whose drinking water is contaminated with solvents including TCE show excess rates of leukemia and a few show excess NHL.

References 1. Linet MS. The leukemias: epidemiologic aspects. New York: Oxford University Press; 1985. 2. Tsongas TA. Occupational factors in the epidemiology of chemically induced lymphoid and hemopoietic cancers. In: Irons RD, ed. Toxicology of the blood and bone marrow. New York: Raven Press; 1985:149-77. 3. Goldstein BD, Kipen HM. Cancer secondary to chemotherapy. In: Witmer CM, Snyder RR, Jollow DJ, et al., eds. Biologic reactive intermediates IV: molecular and cellular effects and their impact on human health. New York: Plenum Press; 1991:619-25. 4. Linet MS, Cartwright RA. The leukemias. In: Schottenfeld D, Fraumeni FR Jr, eds. Cancer epidemiology and prevention. New York: Oxford University Press; 1996:841-92. 5. Mueller NE. Hodgkin’s disease. In: Schottenfeld D, Fraumeni FR Jr, eds. Cancer epidemiology and prevention. New York: Oxford University Press; 1996:893-919. 6. Scherr PA, Mueller NE. Non-Hodgkin’s lymphomas. In: Schottenfeld D, Fraumeni FR Jr, eds. Cancer epidemiology and prevention. New York: Oxford University Press; 1996: 920-45. 7. Hayes R, Songnian Y, Dosemeci M, et al. Benzene and lymphohematopoietic malignancies in humans. Am J Ind Med 2001; 40:117-26. 8. Fagioli F. Distinct cytogenetic and clinicopathologic features in acute myeloid leukemia after occupational exposure to pesticides and organic solvents. Cancer 1992; 70:77-85. 9. Dann EJ, Rowe JM. Biology and therapy of secondary leukemias. Best Pract Res Clin Haematol 2001; 14:119-37. 10. Crane MM, Keating MJ, Trujillo JM, et al. Environmental exposures in cytogenetically defined subsets of acute nonlymphocytic leukemia [published erratum appears in JAMA 1990; 263:662]. JAMA 1989; 262:634-9. 11. Rodella S, Ciccone G. Cytogenetics and occupational exposures in acute nonlymphocytic leukemia and myelodysplastic syndrome. Scand J Work Environ Health 1993; 19:369-74.

12. West RR. Cytogenetic abnormalities in the myelodysplastic syndromes and occupational or environmental exposure. Blood 2000; 95:2093-7. 13. Matanoski GM, Stockwell HG, Diamond EL, et al. A cohort mortality study of painters and allied tradesmen. Scand J Work Environ Health 1986; 12:16-21. 14. Tates AD. Biologic effect monitoring in industrial workers following incidental exposure to high concentrations of ethylene oxide. Mutat Res 1995; 329:63-77. 15. Theriault GP, Goldberg M, Miller AB, et al. Cancer risks associated with occupational exposure to magnetic fields among electric utility workers in Ontario and Quebec, Canada, and France: 1970-1989 [published erratum appears in Am J Epidemiol 1994; 139:1053]. Am J Epidemiol 1994; 139:550-72. 16. Dellarco VL, Generoso WM, Sega GA, et al. Review of the mutagenicity of ethylene oxide. Environ Mol Mutagen 1990; 16:85-103. 17. Zhang L. The nature of chromosomal aberrations detected in humans exposed to benzene. Crit Rev Toxicol 2002; 32: 1-42. 18. Blair A, Zheng T, Linos A, et al. Occupation and leukemia: a population-based case-study in Iowa and Minnesota. Am J Ind Med 2000; 40:3-14. 19. Willman CL. Molecular evaluation of acute myeloid leukemias. Semin Hematol 1999; 36:390-400. 20. Crane MM, Strom SS, Halabi S, et al. Correlation between selected environmental exposures and karyotype in acute myelocytic leukemia. Cancer Epidemiol Biomarkers Prev 1996; 5:639-44. 21. Perera FP. Molecular epidemiology: on the path to prevention? J Natl Cancer Inst 2000; 92:602-12. 22. Colt JS, Blair S. Parental occupational exposures and risk of childhood cancer. Envrion Health Perspect 1998; 106(Suppl):909-25. 23. Freedman DM, Stewart P, Kleinerman RA, et al. Household solvent exposures and childhood acute lymphoblastic leukemia. Am J Public Health 2001; 91:564-7. 24. Mayer J, Warburton D, Jeffrey AM, et al. Biologic markers in ethylene oxide-exposed workers and controls. Mutat Res 1991; 248:163-76. 25. Jaffe ES, Raffeld M, Medeiros LJ, et al. An overview of the classifications of non-Hodgkin’s lymphomas: an integration of morphological and phenotypical concepts. Cancer Res 1992; 52(Suppl):5447s-5452s. 26. Hartge P, Devesa SS. Quantification of the impact of known risk factors on time trends in non-Hodgkin’s lymphoma incidence. Cancer Res 1992; 52(Suppl): 5566s-5569s. 27. Ferreira OC Jr, Planelles V, Rosenblatt JD. Human T-cell leukemia viruses: epidemiology, biology and pathogenesis. Blood Rev 1997; 111:91-104. 28. Blattner WA. Human retroviruses: their role in cancer. Proc Assoc Am Physicians 1999; 111:563-72. 29. Beral V, Newton R. Overview of the epidemiology of immunodeficiency-associated cancers. J Natl Cancer Inst Monogr 1998; 23:1-6. 30. Baris D, Zahm SH. Epidemiology of lymphomas. Curr Opin Oncol 2000; 12:383-94. 31. Morales Suarez-Varela MM, Llopis Gonzalez A, Marquina Vila A, et al. Mycosis fungoides: review of epidemiological observations. Dermatology 2000; 201:21-8. 32. Davis DL, Hoel D, Fox J, et al. International trends in cancer mortality in France, West Germany, Italy, Japan, England, Wales, and the USA. Lancet 1990; 336:474-81. 33. Aguiar RC. Therapy-related chronic myeloid leukemia: an epidemiological, clinical and pathogenetic appraisal [published erratum appears in Leuk Lymphoma 1998; 30:665]. Leuk Lymphoma 1998; 29:17-26. 34. McDevitt JJ, Lees PS, McDiarmid MA. Exposure of hospital pharmacists and nurses to antineoplastic agents. J Occup Med 1993; 35:57-60.

Occupational and Environmental Risk Factors for Lymphohematopoietic Malignancies 755 35. McMichael AJ. Carcinogenicity of benzene, toluene and xylene: epidemiological and experimental evidence. IARC Sci Pub 1988; 85:3-18. 36. McMichael AJ, Spirtas R, Gamble JF, et al. Mortality among rubber workers: relationship to specific jobs. J Occup Med 1976; 18:178-85. 37. McMichael AJ, Spirtas R, Kupper LL. An epidemiologic study of mortality within a cohort of rubber workers. J Occup Med 1974; 16:458-64. 38. Finch SC, Linet MS. Chronic leukemia. Baillière’s Clin Haematol 1992; 5:27-56. 39. Chase A, Huntly BJ, Cross NC. Cytogenetics of chronic myeloid leukemia. Best Pract Res Clin Haematol 2001; 14: 553-71. 40. Faderl S, Kantarjian HM, Talpaz M, et al. New treatment approaches for chronic myelogenous leukemia. Semin Oncol 2000; 27:578-86. 41. Tondel M, Persson B, Cartensen J. Myelofibrosis and benzene exposure. Occup Med (Lond) 1995; 45:51-2. 42. Bench AJ, Cross NC, Huntly BJ, et al. Myeloproliferative disorders. Best Pract Res Clin Haematol 2001; 14:531-51. 43. Cardis E, Gilbert ES, Carpenter L, et al. Combined analysis of cancer mortality among nuclear industry workers in Canada, the United Kingdom and the United States of America. Lyons, France: International Agency for Research on Cancer; 1995. 44. National Research Council. Health effects of exposure to low levels of ionizing radiation – Beir V. National Research Council, Committee on the Biologic Effects of Ionizing Radiations, Board on Radiation Effects Research, and Commission on Life Sciences. Washington DC: National Academy Press; 1990. 45. Linos A, Gray JE, Orvis AL, et al. Low-dose radiation and leukemia. N Engl J Med 1980; 302:1101-5. 46. Stewart A, Pennybacker W, Barber R. Adult leukemia and diagnostic x-rays. BMJ 1962; 2:882-90. 47. Gibson R, Graham S, Lillienfeld A, et al. Irradiation in the epidemiology of leukemia among adults. J Natl Cancer Inst 1972; 48:301-11. 48. Lyman GH, Lyman CH, Johnson W. Association of leukemia with radium groun dwater contamination. JAMA 1985; 254:621-6. 49. Collman GW, Loomis DP, Sandler DP. Childhood cancer mortality and radon concentration in drinking water in North Carolina. Br J Cancer 1991; 63:626-9. 50. Fuortes L. Leukemia incidence and radioactivity in drinking water in 59 Iowa towns. Am J Public Health 1990; 80:1261-2. 51. Laurier D, Valenty M, Tirmarche M. Radon exposure and the risk of leukemia: a review of epidemiological studies. Health Phys 2001; 81:272-88. 52. Gardner MJ, Snee MP, Hall AJ, et al. Results of case-control study of leukaemia and lymphoma among young people near a Sellafield nuclear plant in West Cumbria. BMJ 1990; 300:423-9. 53. Kinlen L. Evidence for an infective cause of childhood leukemia: comparison of a Scottish new town with nuclear reprocessing sites in Britain. Lancet 1988; ii:1323-7. 54. Kinlen LJ, Dickson M, Stiller CA. Childhood leukaemia and non-Hodgkin’s lymphoma near large rurual construction sites, with a comparison with Sellafield nuclear site. BMJ 1995; 310:763-8. 55. Kinlen L. Infection, childhood leukemia and the Seascale cluster. Radiol Protect Bull 2000; 226:9-18. 56. Greaves MF. Speculations on the cause of childhood acute lymphoblastic leukemia. Leukemia 1988; 2:120-5. 57. Greaves MF, Alexander FE. An infectious etiology for common acute lymphoblastic leukemia in childhood? Leukemia 1993; 7:349-60. 58. Greaves MF. Childhood leukemia. BMJ 2002; 324:283-7. 59. Aksoy M, Erdem S, Dincol G. Types of leukaemia in a chronic benzene poisoning: a study in thiry-four patients. Acta Haematol 1976; 55:65-72.

60. Aksoy M. Different types of malignancies due to occupational exposure to benzene: a review of recent observations in Turkey. Environ Res 1980; 23:181-90. 61. Vigliani EC, Forni A. Benzene and leukemia. Environ Res 1976; 11:122-7. 62. Infante PF, Rinsky RA, Wagoner JK, et al. Leukaemia in benzene workers. Lancet 1977; 2:76-8. 63. Rinsky R, Smith A, Hornung R, et al. Benzene and leukemia – an epidemiologic risk assessment. N Engl J Med 1987; 316:1044-50. 64. Hotz P. Biologic monitoring of vehicle mechanics and other workers exposed to low concentrations of benzene. Int Arch Occup Environ Health 1997; 70:29-40. 65. Zahm SH. Pesticides and childhood cancer. Environ Health Perspect 1998; 106(Suppl):893-908. 66. Daniels JL. Pesticides and childhood cancers. Environ Health Perspect 1997; 105:1068-77. 67. Blair A, Zahm SH. Agricultural exposures and cancer. Environ Health Perspect 1995; 103(Suppl):205-8. 68. Court Brown WM, Doll R. Mortality from cancer and other causes after radiotherapy for ankylosing spondylitis. BMJ 1965; 2:1327-32. 69. Sessink PJ, Bos RP. Drugs hazardous to healthcare workers. Evaluation of methods of monitoring occupational exposure to cytostatic drugs. Drug Safety 1999; 20:347-59. 70. McCunney RJ. Hodgkin’s disease, work and the environment. J Occup Environ Med 1999; 41:36-46. 71. Savitz DA, Loomis DP. Magnetic field exposure in relation to leukemia and brain cancer mortality among electric utility workers [published erratum appears in Am J Epidemiol 1996; 144:205] . Am J Epidemiol 1995; 141:123-34. 72. Floderus B, Persson T, Stenlund C, et al. Occupational exposure to electromagnetic fields in relation to leukemkia and brain tumors: a case-control study in Sweden. Cancer Causes Control 1993; 4:465-76. 73. National Research Council. Possible health effects of exposure to residential electric and magnetic fields. Committee on the possible effects of electromagnetic fields on biologic systems. Washington DC: National Academy Press; 1997:1-356. 74. Portier C, Wolfe M. NIEHS Working Group. Assessment of health effects from exposure to power-line frequency electric and magnetic fields. NIH Publication No. 98. Research Triangle Park, North Carolina: NIEHS, NIH, USDHHS, PHS; 1998:1-508. 75. Wartenberg D, Dietrich F, Goldberg R, et al. A meta-analysis of studies of childhood cancer and residential exposure to magnetic fields. A report to the National Institute of Environmental Health Sciences. 2002. 76. Wartenberg D. Residential EMF exposure and childhood leukemia: meta-analysis and population attributable risk. Bioelectricmagnetics 2001; Suppl 5:S86-S104. 77. Ahlbom A, Day N, Feychting M, et al. A pooled analysis of magnetic fields and childhood leukemia. Br J Cancer 2000; 83:692-8. 78. Greenland S, Sheppard AR, Kaune WT, et al. A pooled analysis of magnetic fields, wire codes and childhood leukemia. Epidemiology 2000; 11:624-34. 79. Becker N, Berger J, Bolm-Audorff U. Asbestos exposure and malignant lymphomas – a review of the epidemiological literature. Int Arch Occup Environ Health 2001; 74:459-69. 80. McMichael AJ, Spirtas R, Kupper LL, et al. Solvent exposure and leukemia among rubber workers: an epidemiologic study. J Occup Med 1975; 17:234-9. 81. Acquavella JF, Leonard RC. A review of the epidemiology of 1,3-butadiene and chloroprene. Chem Biol Interact 2001; 135-136:43-52. 82. Lemen RA, Meinhardt TJ, Crandall MS, et al. Environmental epidemiologic investigations in the styrene-butadiene rubber production industry. Environ Health Perspect 1990; 86:103-6. 83. March HC. Leukemia in radiologists. Radiology 1944; 43:275-8. 84. Lynge E, Antitila A, Hemminki K. Organic solvents and cancer. Cancer Causes Control 1997; 8:406-19.

756 Lymphohematopoietic Malignancies 85. Steenland K, Stayner L, Greife A, et al. Mortality among workers exposed to ethylene oxide. N Engl J Med 1991; 324:1402-7. 86. Ries LAG, Eisner MP, Kosary CL, et al., eds. SEER Cancer Statistics Review, 1973-1998. Bethesda, Maryland: National Cancer Institute; 2001. 87. Beebe GW, Kato H, Land CE. Studies of mortality of A-bomb survivors. VI. Mortality and radiation dose. Radiat Res 1978; 75:138-201. 88. Caldwell GG, Kelley D, Zack M, et al. Mortality and cancer frequency among military nuclear test (Smoky) participants, 1957 through 1979. JAMA 1983; 250:620-4. 89. Stern FB, Waxweiler RA, Beaumont JJ, et al. A case-control study of leukemia at a naval nuclear shipyard. Am J Epidemiol 1986; 123:980-92. 90. Stevens W, Thomas DC, Lyon JL, et al. Leukemia in Utah and radioactive fallout from the Nevada test site. JAMA 1990; 264:585-91. 91. Archer VE. Association of nuclear fallout with leukemia in the United States. Arch Environ Health 1987; 42:263-71.

92. O’Brien TR, Decoufle P, Rhodes PH. Leukemia and ground water contamination [letter to the editor]. JAMA 1987; 257:317. 93. Cohen BL. Radon exposure in homes and cancer [letter to the editor]. Lancet 1991; 337:790-1. 94. Aksoy O, Dincol K, Erdem S, et al. Details of blood changes in 32 patients with pancytopenia associated with long-term exposure to benzene. Br J Ind Med 1972; 29:56-64. 95. Rinsky RA, Smith AB, Young RJ. Leukemia in benzene workers. Am J Ind Med 1981; 2:217-45. 96. Goldstein B. Is exposure to benzene a cause of human multiple myeloma? Ann N Y Acad Sci 1990; 609:225-30. 97. Goldstein BD. Clinical hematotoxicity of benzene. In: Benzene: occupational and environmental hazards – scientific update. Princeton, New Jersey: Princeton Scientific; 1989:55-67. 98. Vianna NJ. Lymphomas and occupational benzene exposure. Lancet 1979; i:1394-5.

30.3 Bladder Cancer Avima M Ruder, Tania Carreón, Elizabeth M Ward, Paul A Schulte, William Halperin It is well documented that the etiology of bladder cancer involves environmental risk factors. Occupational risks may account for 21–27% of bladder cancers among men in the United States,1,2 an estimated 40,000 cases in 2001,3 and 11% among the estimated 15,000 cases in women in 2001.3,4 Occupational exposure to aromatic amines has been known to cause bladder cancer since Rehn identified the first few cases in workers in the new organic chemical industry in 1895.5 Since that time, numerous occupations and specific substances have been associated with an increased risk of bladder cancer (Tables 30.3.1 and 30.3.2). Beyond the certainties that specific aromatic amines have been demonstrated to be human occupational bladder carcinogens and that a broad range of occupations are at risk of bladder cancer, a well-informed approach to the prevention and management of bladder cancer depends on appreciating various controversies involved in its primary and secondary prevention and treatment. The reader is referred to a report of a national conference held in 19896 to delineate these issues. Many of these controversies, such as the relevance to human bladder cancer of findings from animal studies, the line between benign and malignant tumors, the appropriate screening regimen for workers exposed to bladder carcinogens, and whether early detection is worthwhile, remain relevant more than a decade later. In addition to incidence and survival differences by social class, race, and gender,7 new developments in understanding inherited risk factors, such as acetylator status and intermediate biomarkers, influence understanding of etiology, prevention and management.

ETIOLOGY Epidemiology Bladder cancer is the eighth most common neoplasm worldwide. An estimated 261,000 new cases of bladder cancer occur each year, with four-fifths of the cases occurring in men.8 Incidence rates of bladder cancer vary about 10-fold, with higher rates in Western Europe and North America, and lower rates in Eastern Europe and several Asian countries.9 Cigarette smoking accounts for 47% of bladder cancer among men and 37% among women; smokers have twice the risk of non-smokers.3 The contributions of coffee drinking and alcohol consumption to bladder cancer are equivocal. The World Health Organization estimates that bladder cancer cases can be subdivided into two broad categories based on etiology. (1) Those caused by tobacco and industrial carcinogens are predominantly transitional cell carcinomas and are common in industrialized countries. (2) Those due to bilharzia, human papilloma virus, and schistosomiasis infection10,11 are more likely to be squamous cell carcinomas and are found chiefly in subtropical

and tropical countries. Thus, both types are, in principle, preventable.12 Bladder cancer is historically the neoplastic disease most strongly linked to occupational exposure to chemicals. Several occupations have been suspected to increase the risk of bladder cancer, but strong associations only exist for dye workers, aromatic amine manufacturing workers, leather workers, rubber workers, painters, truck drivers, and aluminum workers.13 Table 30.3.1 presents the epidemiologic studies of occupational associations. Occupational exposures to chemicals such as arylamines, polycyclic aromatic hydrocarbons and other industry-related agents may explain some of the risk associated with these occupations.14 Some of the relative risks for exposed workers are substantially higher than the two-fold increased risk for smokers. Table 30.3.2 includes specific chemicals or processes classified by IARC as carcinogenic to the human or canine bladder, combined with data on working populations potentially at risk. Over 200 chemicals have been confirmed as rodent bladder carcinogens; however, rodent bioassays do not appear to be a good model for human bladder cancer15 (see ‘Animal studies’ section below). A relationship between bladder cancer and exposure to chemical dyes was first established in 1895.5 In the late 1930s, oral administration of the industrial arylamine 2-naphthylamine was shown to induce bladder cancer in dogs.16 In 1954, Case et al. reported a 20-fold excess of bladder cancer in arylamine-exposed individuals, compared to the general population of England and Wales.17 Since then, the most investigated bladder carcinogens have been 2-naphthylamine, benzidine, and 4-aminobiphenyl. Several studies have reported increased risks of bladder cancer in workers exposed to 2-naphthylamine and 4-aminobiphenyl, and have been reviewed elsewhere.18 One newly recognized factor that may have an occupational component is fluid intake. In a study of about 50,000 men, it was observed that drinking more fluids was associated with a significantly decreased rate of bladder cancer.19 These results are consistent with the urogenous contact hypothesis, which proposed that the level of DNA adducts to 4-aminobiphenyl (a carcinogenic amine) could be decreased by increased fluid intake.20,20a Occupations that have exposure to carcinogens and limit the workers’ ability to drink liquids could be at risk. There is no known distinctive histologic feature for occupational bladder cancer. However, in some heavily exposed cohorts, cases regularly occur at ages 15 years younger than the median age at diagnosis of the general population,17,21 in which the rate of bladder cancer increases substantially with age. Usually, the interval from first exposure to onset of symptoms is decades long; however, occupational cases have occurred surprisingly early after exposure, which substantiates the argument that cases with only a few years of latency may be occupationally induced. It has been estimated that the latency for chemically induced bladder cancer ranges from 4 to 45 years.21,22

758 Bladder Cancer

Study site, type, population

Occupation /Industry Aluminum smelterers (10 yrs) Armed services Auto workers (>10 yrs) Butchers Carpenters Carpenters Chemical mfg. workers Chemical mfg. workers

Increased morbidity relative risk

Increased mortality relative risk (95% CI)

Canada, NCC, men

3.9 (1.6–9.6)*

Canada, CC, men US, CC, African–American men Sweden, CC, men US, CC, men US, CC, white men UK, CC, men US, C, white men US, C, African–American men Sweden, CC, men US, CC, white women UK, CC, men China, CR, men US, CC, African–American men

1.8 (1.2–2.7) 4.7 (1.7–10.7) 1.3 (1.0–1.6) 11.1 (3.3–37.0) 1.4 (1.1–1.8) 2.2 (1.7–3.0) 2.6 (1.1–6.4)* 5.0 (2.2–11.3)* 1.3 (1.1–1.5) 2.1 (0.9–5.1)*

US, CC, white men US, C, Latino men

1.6 (1.1–2.5)*

Crafts workers Dental technicians Drivers Drivers

China, CR, women Sweden, CC, men Argentina, CC, men Denmark, CC

1.2 2.5 5.3 1.6

(1.0–1.4) (1.3–4.3) (2.3–12.2) (1.1–2.3)*

Drivers† Drivers, railroad Drivers, taxi Drivers, truck Drivers, truck Drivers, truck Drivers, truck

US, CC, white Germany, CC, US, CC, white US, CC, white Germany, CC, US, CC, white UK, CC, men

1.2 3.0 6.3 2.1 1.8 1.5

(1.1–1.4)* (1.2–8.8) (1.6–29.3)* (1.4–4.4)* (1.1–2.8)* (1.1–2.0)

Dry cleaners Dry cleaners Dye mfg. workers

US, CC, African–American men 2.8 (1.1–7.4) US, C, African–American men 5.1 (1.4–13.1) China, C, men 11.1 (3.6–25.9) 31.5 (20.4–46.4) 17.5 (7.5–34.5) UK, CC, men 3.5 (2.2–5.3) UK, CC, men 2.6 (1.8–3.7) US, C, white men 5.2 (1.4–13.2)

Chemical workers† Chemical workers Clerical workers Clerical workers Clerical workers (>10 yrs) Construction workers Construction workers

Dye mfg. workers Dye mfg. workers Dye mfg. workers

men men men men men men

Howe et al., 198059 Silverman et al., 19891 Malker et al., 198760 current smokers Schumacher et al., 198961 smoking Silverman et al., 19892 Boyko et al., 198562 age Schulte et al, 198563 age Malker et al., 198760 smoking Silverman et al., 19904

age, smoking smoking

sex, age, smoking smoking age, smoking age, smoking age, smoking PMR 2.0 (P 10 yrs) Tailors Tailors Tailors, dressmakers Technicians

Europe, CC, women

8.1 (2.1–32.0)

age, smoking

‘t Mannetje et al., 199985

Germany, CC, men Canada, CC, men Europe, CC, women US, C, African–American women

2.7 (1.1–6.6) 3.9 (1.3–14.2) 1.4 (1.0–2.1)

latency⭓ 8 yrs smoking

Claude et al., 198869 Risch et al., 198874 ‘t Mannetje et al., 1999 Schultz & Loomis, 200066

US, C, Asian men

smoking

Schultz & Loomis, 200066

PMR 2.1 (0.8–5.6)

Schulz & Loomis, 200066

PMR 5.3 (1.6–16.8)

PMR 1.8 (0.8–3.8)

Table 30.3.1 (Cont’d) Occupations associated with increased risk of bladder cancer

Silverman et al., 19904

760 Bladder Cancer

Occupation /Industry

Study site, type, population

Increased morbidity relative risk

Increased mortality relative risk (95% CI)

Controlled for (95% CI)

Reference

Telephone & telegraph ops Textile workers

US, CC, white men

1.9 (0.9–4.0)*

smoking

Silverman et al., 19892

Spain, CC, women

6.4 (1.3–30.0)

Gonzalez et al., 198989

Textile workers

Spain, CC, men

1.9 (1.1–3.1)

Textile workers

Denmark, CC

1.7 (1.1–2.4)

smoking, other jobs smoking, other jobs sex, age, smoking

Textile workers† Tobacco processors Turners (lathe operators) Turners (lathe operators) Upholsterers Weavers Weavers

Italy, CC, women Europe, CC, women UK, CC, men Germany, CC, men Germany, CC, men Germany, CC, men Spain, CC, men

1.9 3.1 1.5 2.3 2.7 2.7 3.5

Weavers

Spain, CC, women

21.2 (1.5–298)

Weavers Welders Welders, oxyacetylene

US, CC, white men Canada, CC, men Italy, C, men

3.5 (1.3–9.3) 2.8 (1.1–8.8)

(0.9–4.2) (1.1–9.3) (1.2–1.8) (1.0–5.6)* (1.1–6.6) (1.1–6.6) (1.3–9.3)

age, smoking

smoking, other jobs smoking, other jobs smoking 3.7 (1.2–8.6)

age

Gonzalez et al., 198989 Jensen et al.. 198768 Maffi & Vineis, 198686 ‘t Mannetje et al., 199985 Cartwright, 198264 Claude et al., 198869 Claude et al., 198869 Claude et al., 198869 Gonzalez et al., 198989 Gonzalez et al., 198989 Silverman et al., 19892 Howe et al., 198059 Merlo et al., 198983

Only occupations explicitly mentioned in a study are included. Study types: CC (case-control), C (cohort), NCC (nested case-control), CR (linkage of case registry and census data) * Dose–response demonstrated † Higher risks were found for some subcategories of workers

Table 30.3.1 (Cont’d) Occupations associated with increased risk of bladder cancer

Based on mortality data from England and Wales, it was estimated that bladder cancer due to occupational exposures was responsible for half of the rate difference between high and low social classes because these exposures are concentrated in blue-collar jobs.23 However, no evidence was provided associating the blue-collar jobs with exposure to bladder carcinogens.

Metabolic polymorphisms and bladder cancer Interindividual variation is common for many metabolic enzymes. In some cases, the variability has been attributed to inherited polymorphisms.24 Phenotypic and genotypic tests have shown that variation in xenobiotic metabolizing enzymes is associated with cancer risk and may have an influence on human susceptibility to genotoxic agents. A limited number of studies of bladder cancer genetic susceptibility in populations exposed occupationally to arylamines has been published.25–29 The results of a metaanalysis of all studies of acetylation status and bladder cancer in the general population suggest that certain groups with the NAT2 slow-acetylation phenotype are at greater risk of bladder cancer.30 Additional studies are needed to establish if individuals could be at higher risk of bladder cancer given the presence of certain alleles that make them more susceptible. In the workplace, various metabolic polymorphisms could be acting in combination with occupational toxicants to produce risk. For occupational bladder cancer, polymorphic genotypes in the NAT (N-acetyltransferase) and GST (glutathione S-transferase) families of genes have been explored. Their joint effect, together with the effect of other genotypes, has not yet

been investigated. Moreover, the metabolic differences between monoarylamines and diarylamines, such as benzidine, warrant careful attention to the specific compounds to which each worker is exposed.

Animal models Use of animal models to predict human bladder carcinogens has been problematic. Rats and mice are not susceptible to bladder cancer by most aromatic amines, including some highly potent occupational carcinogens. For aromatic amines, the Syrian hamster and dog were found to be better predictors for human bladder cancer than mice or rats, but this may not necessarily be the case for other chemical classes.15 There has been considerable debate about the relevance to human bladder cancer of bladder tumors associated with urinary calculi in mice or rats or calcium phosphate-containing precipitates in rats. The International Agency for Research on Cancer (IARC), which has a formal program for identification of carcinogenic hazards to humans, has issued a consensus report on this subject.31 Based on this consensus report, the IARC monograph program has recently classified saccharin in Group 3 (not classifiable as to carcinogenicity in humans) because the mechanism through which it is thought to cause bladder cancer in rats, formation of a calcium phosphate-containing precipitate, is not relevant to humans.32 Numerous classes of genotoxic chemicals have been identified as bladder carcinogens in rodents and some of these have been identified in humans, most notably, aromatic amines, nitrosamines, and cyclophosphamide. In contrast, non-genetoxic chemicals appear to be highly specific with regard to species strain, diet, agent, dose, and mechanism.

Clinical Aspects 761

Compound name [variant name]

CAS#

IARC group

Source(s)

Potentially exposed occupations, numbers*

Aluminum (production) Arsenic & arsenic compounds

7429–90–5† 7440–38–2

1 1

Boffetta Wilbourn

Auramine dye manufacturing Benzidine Benzidine-based dyes Direct Black 38 [2,7-Naphthalenedisulfonic acid, 4-amino-3- {[4′-((2,4diaminophenyl)azo) (1,1′-biphenyl)-4-yl]azo}5-hydroxy-6-(phenylazo)-disodium salt] Direct Blue 6 [2,7-Naphthalenedisulfonic acid, 3,3′-[(1,1′-biphenyl)- ((4,4′-diylbis(azo)]bis(5amino-4-hydroxy)-, tetrasodium salt] Benzidine, 3,3′-dichloro4-Biphenylamine [4-Aminobiphenyl]

492–80–8† 92–87–5

1 1 2A

Boffetta Boffetta IARC 29; IARC Supp 7

91–94–1 92–67–1

2A 1

Wilbourn Boffetta; Wilbourn

494–03–1

1

Wilbourn

50–18–0

1

pharmaceutical mfg. workers, oncology nurses, pharmacists 27,171 (NOES)

65996–93–2

1

roofers – 142,600 (BLS)

Coal tars

65996–89–6

1

Diesel engine exhaust p-Dimethylaminoazobenzene [Brilliant yellow] Magenta dye manufacture 4,4′-Methylene bis (2-chloroaniline) [MOCA] Mineral oils, untreated and mildly treated 2-Naphthylamine Phenacetin [p-Acetophenetidide] Analgesic mixtures containing phenacetin

various 60–11–7 632–99–5† 101–14–4 various 91–59–8 62–44–2 various

2A 2B 1 2A 1 1 2A 1

Wilbourn; Boffetta Wilbourn; IARC Supp 7 Wilbourn Wilbourn Boffetta Wilbourn Wilbourn Boffetta Wilbourn Wilbourn

Rubber industry (certain occupations) Tobacco smoke p-Chloro-o-toluidine

various† 95–69–2

1 1 2A

o-Toluidine [o-Aminoazotoluene]

97–56–3

2A

Chemotherapy agents Chlornaphazine [(N,N-Bis(2-chloro-ethyl)-2naphthylamine)] Cyclophosphamide [2H-1,3,2Oxazaphosphorine, 2-(bis(2-chloroethyl) amino)tetrahydro, 2-oxide] Coal-tar pitches

1937–37–7

al production workers arsenical pesticide manufacturing workers; pesticide users dye mfg. workers 1,554 (NOES) 28,442 (NOES) dye manufacturing workers; dye-using workers – 44,500 (BLS)

2602–46–2

Boffetta Wilbourn IARC, 2000; Wilbourn IARC, 2000; Wilbourn

roofers – 142,600 (BLS) truck drivers – 2,500,000+ (BLS) dye-using workers – 44,500 (BLS) dye mfg. workers metal machinists – 1 million+ (BLS) 275 (NOES) 17,658 (NOES) pharmaceutical mfg. workers, oncology nurses, pharmacists rubber additives workers ‡ rubber additives workers & dye workers

An RTECS search for substances associated with bladder tumors produced the original list, adapted from Ruder et al., 1990.84 IARC classifications are adapted from Wilbourn et al. 199915 and Boffetta et al 1997.23 Updates through Monograph 84 by the authors of this chapter. IARC ratings: 1, definite human bladder carcinogen; 2A, probable human bladder carcinogen. Compounds rated by Wilbourn et al.15 as canine bladder carcinogens are included, but not those rated as exclusively rodent bladder carcinogens. * Estimated numbers are from two sources: 1. (NOES) Chemical-specific numbers of workers potentially exposed are from the 1981–1983 National Occupational Exposure Survey. Note that all these workers would be reaching > 20 years latency about 2001–3; 2. (BLS) Estimates of numbers of workers in occupational categories are from the Bureau of Labor Statistics 1999 National Occupational Employment and Wage Estimates (http://stats.bls.gov/oes). These data do not consider turnover among employees, which would lead to a greater number of persons exposed than is suggested by estimates from specific points in time. † These manufacturing processes involve exposure to a number of chemicals. ‡ No study to date has linked passive smoking to bladder cancer; several have found no association.

Table 30.3.2 Known and suspected human bladder carcinogens and estimated numbers of potentially exposed US workers

Use of molecular and genetic mechanism information may be helpful in identifying possible mechanisms involved for these non-genotoxic chemicals and, therefore, can be important for a rational evaluation of human risk.

CLINICAL ASPECTS Pathology Transitional cell carcinomas (TCC) are graded by the World Health Organization histologically by the degree of abnormality of the tissues33 and staged by the American Joint Committee on Cancer – Union Internationale Contre

le Cancer by the extent to which they have spread.34 One problem in the pathology of bladder tumors is the somewhat ambiguous line between the benign and the malignant, distinguishing between a papilloma, a papillary tumor with delicate fibrovascular stroma covered by a layer of epithelial cells indistinguishable from normal bladder mucosa, and a papillary carcinoma. This distinction should be viewed as a region where borders shift between pathologists, between institutions, and given the circumstances, even successive biopsies of the patient. Some so-called benign papillomas display effects of inflammation and reactive or regenerative conditions so that they are classified by some pathologists as anaplastic,

762 Bladder Cancer although most of them do not behave as malignant tumors. To make the situation more complicated, some papillary tumors do become aggressive. Robinson and Hall33 summarized the results of several studies: about 2–5% of ‘benign’ papillomas progress to carcinoma. The use of biochemical, molecular and genetic characteristics of cells is beginning to provide pathologists with a way to reduce these uncertainties. Strong associations of various markers with progression, invasiveness, and metastatic potential may provide a way to distinguish between pathologic subtypes of bladder cancer in the future.35

Therapy Strategies for diagnosis and therapy of occupational bladder cancer do not differ from those for bladder cancer resulting from non-occupational etiologies. A comprehensive guide to current treatment options by stage and grade has been assembled by the National Cancer Institute.36 Treatment strategies do not appear to differ by histological type. Occupational bladder cancer does have a public health component: putative bladder carcinogens and highrisk populations that need to be followed with screening have been identified when exceptional cases of bladder cancer (i.e., in young non-smokers) were diagnosed.

Survival The survival of patients with bladder cancer depends on the grade of anaplasia of the tumor and the stage of tumor invasion at time of diagnosis. SEER 1992–1997 5-year survival rates range from 94.5% for localized disease in white males to 0% for distant disease in African-American males. At each stage, women fare more poorly than men (except African-American women with distant disease) and African-Americans more poorly than whites. From 1974 to 1997 overall survival has improved from 48% to 65% in African-Americans and from 74% to 82% in whites. During the same period, incidence increased slightly overall from 14.6 to 16.7/100,000, in men from 25.6 to 29.0, and in women from 6.3 to 7.4.37 Survival also appears to depend on social class and race, which, of course, are somewhat correlated. A review of five studies on bladder cancer survival found about a 20% discrepancy overall between higher and lower income patients.38 African-Americans are less likely than whites to develop bladder cancer; however, once diagnosed, African-Americans experience poorer survival: 64% 5-year survival vs. 82% for whites.3

PREVENTION Techniques for early detection Cytoscopy is effective in identifying visible tumors in the bladder. A cytoscope is a slender tube with a lens and a light that is inserted through the urethra, allowing the physician to visually inspect the urethra and bladder.

Cytoscopy is invasive and not employed for asymptomatic individuals.39 Urine cytology is the accepted technique for detection of bladder cancer in asymptomatic individuals. Urine cytology microscopically identifies the presence of abnormal, malignant cells, which are shed into the urine of patients with bladder cancer.39 Cytologic screening for bladder cancer has a sensitivity of about 70% and a specificity of 90–95%, depending on the grade and stage of the tumor,40 which is comparable to screening tests for cervical cancer, breast cancer, and colon cancer.41 Bladder cytology is effective in detecting preclinical stages of aggressive tumors and is substantially less effective in detecting low-grade tumors. There is widespread agreement that superficial well-differentiated papillary tumors rarely can be diagnosed definitively from voided urine cytology. In summary, cytology may be used to detect aggressive tumors, but these tumors may be advanced by the time they are discovered by this method. Cytology is less effective for low-grade tumors, which, although they are less aggressive, it would be desirable to find.42 The greatest determinant of the sensitivity of urine cytology is the level of cytopathologist expertise. Ancillary techniques have been tested to improve the sensitivity of urine cytology. Of the large variety of methods, the most promising techniques appear to be DNA flow cytometry and image analysis for the detection of nuclear aneuploidy. Other sensitive methods include immunocytochemistry to detect the presence of antigens that are commonly expressed in neoplastic urothelium but not in the normal urothelium, such as the Lewis X antigen, and immunohistochemical analysis for the detection of p53 overexpression.43 Hematuria screening, by urinalysis or by dipstick – a positive reaction for blood on urine-reagent-strip testing of asymptomatic people44 – may be a more effective method than cytology for detecting low-grade early stage bladder tumors. The dilemmas with testing for hematuria are: (1) although almost all bladder tumors eventually cause hematuria, an infrequent examination may not be adequately sensitive, and (2) hematuria due to bladder cancer may be intermittent.45 Although more frequent examinations increase the sensitivity of the test for bladder cancer, this method decreases the specificity of the test because it will detect other, non-malignant conditions causing hematuria, including cystitis, kidney disease, and urinary calculi.3 The debate then focuses on the predictive value of a positive test result for hematuria or the probability that a positive test result will reflect bladder cancer rather than another problem. It has been suggested that 5–10% of patients with hematuria have bladder cancer and 10–20% have some other serious urinary tract disease. As a condition becomes more prevalent in a population, the predictive value of a positive test increases. Exposure of an individual to an occupational carcinogen, as well as the individual’s age and other risk factors for bladder cancer, should ensure a higher underlying prevalence of bladder cancer and thus increase the predictive value of a screen for hematuria.45 The American Urological

Prevention 763 Association recommends that asymptomatic microhematuria be evaluated only when associated with a risk (such as occupational exposure to carcinogens) of disease.46 Tests of genetic factors are now being developed or assessed for evaluation of risk factors for bladder cancer. In addition to variations in metabolic phenotypes such as N-acetyltransferase, there are other genetic factors and acquired factors, such as recessive alleles for oncogenes, mutated tumor suppressor genes, and growth factors, that may place individuals at increased risk for bladder cancer independent of occupational exposure.47 These genetic factors could add to any occupational risks for bladder cancer or multiply those risks. It is likely that the rapid pace of research will result in the identification of new predictive or prognostic markers in the near future. A number of new techniques are being tested for use in bladder cancer screening. Most markers appear to have an advantage over urine cytology in terms of sensitivity, especially for detecting low-grade superficial tumors. However, most markers tend to be less specific than cytology, yielding more false-positive results. This scenario is more common in patients with concurrent bladder inflammation or other benign bladder conditions. A summary of the sensitivity, specificity and limitations of these methods is presented in Table 30.3.3. The nuclear matrix protein (NMP) 22 test detects and measures urinary levels of a particular NMP called NuMA (nuclear mitotic apparatus). The NMP22 assay appears to be useful only to monitor, with high accuracy, for recurrence in patients with a past history of bladder cancer.48 The bladder tumor antigen (BTA) test detects the presence of a bladder tumor antigen in the urine of patients with bladder cancer. The BTA stat and the BTA trak assays are qualitative and quantitative assays, respectively. Both detect a human complement factor H-related protein in the urine, and both accurately identify two-thirds of patients with bladder cancer. Both are limited because of the high number of false-positive reactions compared to urine cytology in low-grade bladder tumors.39 The fibrin and fibrinogen degradation products (FDP) test is positive in two-thirds of patients with bladder cancer. The FDP test detects the degradation product of an

Test

Sensitivity

Urine cytology BTA test

17–70% 29–40%

90–95% 68–91%

BTA stat test BTA trak test NMP22 FDP test Telomerase

67–87% 72% 48–80% 40–68% 70–86%

40–70% 43–48% 64–80% 80–96% 60–90%

HA-HAase test

90–92%

80–84%

extravascular fibrin clot produced by tumors. It is more sensitive than urine cytology and has a high specificity.49 Telomerase is an essential enzyme for cellular immortality and tumorigenesis. The telomeric repeat amplification protocol assay for telomerase in exfoliated cells can be used as a tumor marker. However, the low stability of telomerase in urine affects test sensitivity.50 Inflammatory cells and stem cells have telomerase activity, and may be the source of false-positive tests. Hyaluronic acid is a glycosaminoglycan that promotes tumor metastasis. High levels are detectable in the urine of patients with bladder cancer. Patients with high-grade TCC have elevated urinary hyaluronidase activity. A combination of both tests (HA-HAase test) yields a higher sensitivity than the sensitivity of individual tests,51 but it has no better sensitivity than urine cytology for detecting low-grade lesions. Recently, detection of survivin in urine has been suggested as a predictive molecular marker of bladder cancer. Survivin is an enzyme inhibitor of apoptosis that is selectively overexpressed in human cancers, but undetectable in most normal adult tissues. In a patient series, the sensitivity of the urine survivin test for new or recurrent bladder cancer was 100%, and the specificity was 90–100%, depending on the population tested.52

Test batteries Combining tests can increase their sensitivity and specificity.41 Series testing is used to increase specificity and reduce the number of false-positive results. Parallel testing is used to increase sensitivity and reduce the number of false-negative results. A new US/European research consortium wants to create a simple, cost-effective, non-invasive diagnostic test to replace cytoscopy and cytology, initiating multicenter trials to find which of seven molecular markers, alone or in combination, is the most accurate detector of bladder cancer.53

Screening programs There are two reasons for screening a population exposed to a known or suspect bladder carcinogen. First, individu-

Specificity

Limitations Poor criteria to identify low-grade TCC* Low detection of grade I TCC. Poorer predictive value than urine cytology High false positive with gross hematuria, prostate cancer, BCG High false positive with UTI, stones, instrumentation High false positive with gross hematuria High false positive with gross hematuria False negatives with gross hematuria, false positives with inflammation, complicated assay not widely available No detection of grade I TCC

Modified from Brown, 200043 * Abbreviations: BCG = bacillus Calmette-Guérin, BTA = bladder tumor antigen, FDP = fibrin/fibrinogen degradation products, HA-HAase = hyaluronic acid/hyaluronidase, NMP = nuclear matrix protein, TCC = transitional cell carcinoma, UTI = urinary tract infection

Table 30.3.3 Sensitivity and specificity of non-invasive bladder tumor markers

764 Bladder Cancer als may be screened so that their tumors can be detected early when they are more readily treated, resulting in less morbidity and higher survival rates. This type of screening is for the personal benefit of the individuals. The second rationale for screening is to detect disease in a population at the earliest time possible in order to ensure that more primary methods of disease prevention, such as engineering controls and use of personal protective devices, are effectively incorporated to prevent exposure.54 The two motivations should be kept in mind in appreciating a consensus view that was reached at the 1989 conference on screening for bladder cancer in highrisk groups. For populations exposed to known carcinogens at high levels, cytologic examination and testing for hematuria was recommended at 6-month intervals. The rationale for including hematuria was to ensure the acceptability of the screening program by ensuring that low-grade tumors would be detected that otherwise may be missed by cytology. For low-exposure groups, as may be found in patients suffering from conditions as a result of environmental exposures, cytology was recommended 2 years after the first exposure, then every 5 years thereafter. For a suspect carcinogen, at high-exposure levels, cytology was recommended every 6 months, as well as measurement of hematuria to detect low-grade tumors. The argument for detecting low-grade tumors, even though there may be limited personal benefit for the individual because most such tumors are less aggressive, was to provide information that exposure had not been adequately controlled. The panel was not enthusiastic about any recommendations for a suspect carcinogen at low levels of exposure.55 When weighing the benefits of a strategy of early detection, be it for the personal benefit of the worker or for the benefit of the workforce, it is necessary to consider the extent to which false-positive findings will be involved. A screening modality that leads to a disproportionate number of unnecessary follow-up and diagnostic procedures may not be cost effective or personally desirable. Moreover, the lengthening of the lead time, although possibly providing an extended opportunity for therapeutic intervention, also could provide a longer period of anxiety and distress for the worker.54 In contrast to the recommendations of the bladder cancer conference consensus panel, the US Preventive Services Task Force concluded that there was insufficient evidence that hematuria and cytology screening improved the prognosis for those found to have cancer, even within high-risk groups.56 If the efficacy of hematuria and cytology screening is not established, then in monitoring high-risk populations, cystoscopy should be reserved as a diagnostic test in individuals who had positive results on cytology and hematuria. However, it should be remembered that in at least one high-risk group, the MBOCA cohort, bladder cancer was diagnosed in two individuals who had had negative hematuria by dipstick and cytology screening.57 Perhaps the non-invasive screening batteries now under development will end this dilemma.

SUMMARY Despite substitution and process changes to prevent or reduce worker exposure, a substantial number of workers continue to be exposed to bladder carcinogens. Much larger numbers have been exposed to bladder carcinogens in the past. Some of these workers may still be at risk for bladder cancer. Improved screening options for high-risk groups, as well as better treatment options, should continue to improve survival and quality of life for these individuals.

References 1. Silverman DT, Levin LI, Hoover RN. Occupational risks of bladder cancer in the United States: II. Non-white men. J Natl Cancer Inst 1989; 81:1480–3. 2. Silverman DT, Levin LI, Hoover RN, Hartge P. Occupational risks of bladder cancer in the United States: I. White men. J Natl Cancer Inst 1989; 81:1472–9. 3. American Cancer Society. Cancer facts and figures 2001. Atlanta: ACS, 2001. 4. Silverman DT, Levin LI, Hoover RN. Occupational risks of bladder cancer among white women in the United States. Am J Epidemiol 1990; 132:453–61. 5. Rehn L. Blasengeschwülste bei fuchsin-arbeitern [Urinary bladder tumors in dye workers]. Arch Klin Chir 1895; 50:588–600. 6. Schulte P, Halperin W, Ward E, Ruder A. Bladder cancer screening in high risk groups. J Occup Med 1990; 32:787–945. 7. Hartge P, Harvey ED, Linehan WM, et al. Unexplained excess risk of bladder cancer in men. J Natl Cancer Inst 1990; 82: 1636–40. 8. Parkin DM, Pisani P, Ferlay J. Global cancer statistics. CA Cancer J Clin 1999; 49:33–64. 9. Silverman DT, Hartge P, Morrison AS, Devesa SS. Epidemiology of bladder cancer. Hematol Oncol Clin North Am 1992; 6:1–30. 10. EL Mawla MD, EL Bolkainy MN, Khaled HM. Bladder cancer in Africa: update. Sem Oncol 2001; 28:174–8. 11. Lopez-Beltran A, Escudero AL. Human papillomavirus and bladder cancer. Biomed Pharmacother 1997; 51:252–7. 12. Koroltchouk V, Stanley K, Sthernsward J, Mott K. Bladder cancer: approaches to prevention and control. Bull WHO 1987; 65:513–20. 13. Silverman DT, Morrison AS, Devesa SS. Bladder cancer. In: Cancer epidemiology and prevention, 2nd edn. New York: Oxford University Press 1996;1156–79. 14. Zhang Z-F, Steineck G. Epidemiology and etiology of bladder cancer. In: Raghavan D, Scher HI, Leibel SA, Lang PH, eds. Principles and practice of genitourinary oncology. Philadelphia: Lippincott-Raven, 1997;215–22. 15. Wilbourn JD, Partensky C, Rise JM. Agents that induce epithelial neoplasms of the urinary bladder, renal cortex and thyroid follicular lining in experimental animals and humans: summary of data from IARC Monographs volumes 1–69. In: Capen CC, Dybing E, Rice JM, Wilbourn JD, eds. Species differences in thyroid, kidney and urinary bladder carcinogenesis. IARC Scientific Publications No. 147. Lyon: IARC, 1999;191–209. 16. Hueper WC, Wiley FH, Wolfe HD. Experimental production of bladder tumors in dogs by administration of betanaphthylamine. J Industr Hyg Toxicol 1938; 20:46–84. 17. Case RAM, Hosker ME, MacDonald DB, Pearson JT. Tumors of the urinary bladder in workmen engaged in the manufacture and use of certain dyestuff intermediaries in the British chemical industry. Br J Ind Med 1954; 11:75–104. 18. Vineis P, Pirastu R. Aromatic amines and cancer. Cancer Causes Control 1997; 8:346–55.

Summary 765 19. Michaud DS, Spiegelman D, Clinton SK, et al. Fluid intake and the risk of bladder cancer in men. N Engl J Med 1999; 340:1390–7. 20. Oyasu R, Hupp ML. The etiology of cancer of the bladder. Surg Gynecol Obstet 1974; 138:97–108. 20a. Melicow MM. Tumors of the bladder: a multifaceted problem. J Urol 1974; 112: 467–78. 21. Schulte PA, Ringen K, Hemstreet GP, Ward E. Occupational cancer of the urinary tract. Occup Med 1987; 2:85–107. 22. Wallace OMA. Occupational urothelial cancer. Br J Urol 1988; 61:175–82. 23. Boffetta P, Kogevinas M, Westerholm P, Saracci R. Exposure to occupational carcinogens and social class differences in cancer occurrence. IARC Sci Publ 1997; 138:331–41. 24. Hayes RB. Genetic susceptibility and occupational cancer. Med Lav 1995; 86: 206–13. 25. Cartwright RA, Glashan RW, Rogers HJ, et al. Role of Nacetyltransferase phenotype in bladder cancer. Lancet 1982; ii(8303):842–5. 26. Hanke J, Krajewska B. Acetylation phenotypes and bladder cancer. J Occup Med 1990; 32: 917–18. 27. Hayes RB, Bi W, Rothman N, et al. N-Acetylation phenotype and genotype and risk of bladder cancer in benzidine-exposed workers. Carcinogenesis 1993; 14:675–8. 28. Rothman N, Hayes RB, Zenser TV, et al. The glutathione S-transferase M1 (GSTM1) null genotype and benzidineassociated bladder cancer, urine mutagenicity, and exfoliated urothelial cell DNA adducts. Cancer Epidemiol Biomark Prevent 1996; 5:979–81. 29. Shinka T, Ogura H, Morita T, Nishikawa T, Fujinaga T, Ohkama T. Relationship between glutathione S-transferase M1 deficiency and urothelial cancer in dye workers exposed to aromatic amines. J Urol 1998; 159:380–3. 30. Marcus PM, Vineis P, Rothman N. NAT2 slow acetylation and bladder cancer risk: a meta-analysis of 22 case-control studies conducted in the general population. Pharmacogenetics 2000; 10:115–22. 31. Capen CC, Dybing E, Rice JM, Wilbourn JD, and Workshop Participants. Consensus report. In: Capen CC, Dybing E, Rice JM, Wilbourn JD, eds. Species differences in thyroid, kidney and urinary bladder carcinogenesis. IARC Scientific Publications No. 147. Lyon: IARC, 1999;1–14. 32. International Agency for Research on Cancer. Saccharin and its salts. Lyon: IARC, 2003. http://wwwcie.iarc.fr/htdocs/ monographs/vol73/73-19.html 33. Robinson MC, Hall RR. Histopathology of urothelial cancer: consensus or controversy? In: Hall RR, ed. Clinical management of bladder cancer. New York: Oxford University Press, 1999;25–65. 34. Droller MJ. Clinical presentation, investigation, and staging of bladder cancer. In: Raghavan D, Scher HI, Leibel SA, Lang PH, eds. Principles and practice of genitourinary oncology. Philadelphia: Lippincott-Raven 1997;249–59. 35. Fradet Y, Cordon-Cardo C. Tumor markers in the management of bladder cancer. In: Raghavan D, Scher HI, Leibel SA, Lang PH, eds. Principles and practice of genitourinary oncology. Philadelphia: Lippincott-Raven 1997;231–8. 36. National Cancer Institute. PDQ website, December 15, 2003. http://www.nci.nih.gov/cancerinfo/pdq/cancerdatabase. 37. Ries LAG, Eisner MP, Kosary CL, et al., eds. SEER Cancer Statistics Review, 1973–1998. Bethesda, MD: National Cancer Institute, 2001. 38. Kogevinas M, Porta M. Socioeconomic differences in cancer survival: a review of the evidence. IARC Sci Publ 1997; 138:177–206. 39. Pirtskalaishvili G, Konety BR, Getzenberg RH. Update on urine-based markers for bladder cancer. Postgrad Med J 1999; 106:85–94. 40. Hemstreet GP, Bonner RB, Hurst RE, Rao JY. Cytology of bladder cancer. In: Vogelzang NJ, Scardino PT, Shipley WV, Coffey DS, eds. Comprehensive textbook of genitourinary

41. 42. 43. 44.

45. 46.

47. 48.

49.

50.

51.

52.

53. 54.

55. 56.

57.

58.

59.

60.

61. 62. 63.

64.

65.

oncology. Philadelphia: Lippincott Williams & Wilkins, 2000;322–32. Hulka B. Principles of bladder cancer screening in an intervention trial. J Occup Med 1990; 32:812–6. Farrow G. Urine cytology in the detection of bladder cancer: a critical approach. J Occup Med 1990; 32:817–21. Brown FM. Urine cytology: is it still the gold standard for screening? Urol Clin North Am 2000; 27:25–37. Hall RR. 1999 The diagnosis of bladder cancer. In: Hall RR, ed. Clinical management of bladder cancer. New York: Oxford University Press 1999;1–20. Messing E, Vaillancourt A. Hematuria screening for bladder cancer. J Occup Med 1990; 32:838–46. Grossfeld GD, Wolf JS, Litwan MS, et al. Asymptomatic microscopic hematuria in adults: summary of the AUA best practice policy recommendations. Am Fam Phys 2001; 63:114–54. Koenig F, Jung K, Schnorr D, Loening SA. Urinary markers of malignancy. Clin Chim Acta 2000; 297:191–205. Soloway MS, Briggman JV, Carpinito GA, et al. Use of a new tumor marker, urinary NMP22, in the detection of occult or rapidly recurring transitional cell carcinoma of the urinary tract following surgical treatment. J Urol 1996; 156:363–7. Schmetter BS, Habicht KK, Lamm DL, et al. A multicenter trial evaluation of the fibrin/fibrinogen degradation products test for detection and monitoring of bladder cancer. J Urol 1997; 158:801–5. Lokeshwar VB, Soloway MS. Current bladder tumor tests: does their projected utility fulfill clinical necessity? J Urol 2001; 165:1067–77. Lokeshwar VB, Block NL. HA-HAase urine test. A sensitive and specific method for detecting bladder cancer and evaluating its grade. Urol Clin North Am 2000; 27:53–61. Smith SD, Wheeler MA, Plescia J, Colberg JW, Weiss RM, Altieri DC. Urine detection of survivin and diagnosis of bladder cancer. J Am Med Assoc 2001; 285:324–8. Agres T. Finding a better way to identify bladder cancer. The Scientist 2001; 15 (June 25): 21. Halperin WE, Ratcliffe J, Frazier TM, Wilson L, Becker SP, Schulte PA. Medical screening in the workplace: proposed principles. J Occup Med 1986; 28:547–52. Halperin W, Cartwright RA, Farrow GM, et al. Where do we go from here? J Occup Med 1990; 32: 936–45. United States Preventive Services Task Force. Guide to Clinical Preventive Services, 2nd edn, 1996. December 15, 2003. http://www.ahcpr.gov/clinic/cpsix.htm Ward E, Halperin W, Thun M, et al. Screening workers exposed to 4,4′-methylenebis(2-chloroaniline) for bladder cancer by cystoscopy. J Occup Med 1990; 32:865–68. Armstrong BG, Tremblay CG, Cyr D, Theriault GP. Estimating the relationship between exposure to tar volatiles and the incidence of bladder cancer in aluminium smelter workers. Scand J Work Environ Health 1986; 12:483–93. Howe GR, Burch JD, Miller AB, et al. Tobacco use, occupation, coffee, various nutrients and bladder cancer. J Natl Cancer Inst 1980; 64:701–3. Malker HSR, McLaughlin JK, Silverman DT, et al. Occupational risks for bladder cancer among men in Sweden. Cancer Res 1987; 47:6763–6. Schumacher MC, Slattery ML. Occupation and bladder cancer in Utah. Am J Ind Med 1989; 16:89–102. Boyko RW, Cartwright RA, Glashan RW. Bladder cancer in dye manufacturing workers. J Occup Med 1985; 27:799–803. Schulte PA, Ringen K, Hemstreet GP, et al. Risk assessment of a cohort exposed to aromatic amines. Initial results. J Occup Med 1985; 27:115–21. Cartwright R. Occupational bladder cancer and cigarette smoking in West Yorkshire. Scand J Work Environ Health 1982; 8:79–82. Zheng W, McLaughlin JK, Gao YT, Silverman DT, Gao RN, Blot WJ. Bladder cancer and occupation in Shanghai, 1980–1984. Am J Ind Med 1992; 21:877–85.

766 Bladder Cancer 66. Schulz M, Loomis D. Occupational bladder cancer mortality among racial and ethnic minorities in 21 states. Am J Ind Med 2000; 38:90–8. 67. Isovich J, Castellrito R, Esteve J, et al. Tobacco smoking, occupational exposure and bladder cancer in Argentina. Int J Cancer 1987; 49:734–40. 68. Jensen OM, Wahrendorf J, Knudsen JB, et al. The Copenhagen case-referent study on bladder cancer. Risks among drivers, painters and certain other occupations. Scand J Work Environ Health 1987; 13:129–34. 69. Claude JC, Prenzal-Beyme RR, Kunze E. Occupation and risk of cancer of the lower urinary tract among men. A casecontrol study. Int J Cancer 1988; 41:371–9. 70. Baxter PJ, McDowall ME. Occupation and cancer in London. An investigation into nasal and bladder cancer using the Cancer Atlas. Br J Ind Med 1986; 43:44–9. 71. Ruder AM, Ward EM, Brown DP. Mortality in dry cleaning workers: an update. Am J Ind Med 2001; 39:121–32. 72. Bi W, Hayes RB, Feng P, et al. Mortality and incidence of bladder cancer in benzidine-exposed workers in China. Am J Ind Med 1992; 21:481–9. 73. Sathiakumar N, Delzell E. An updated mortality study of workers at a dye and resin manufacturing plant. J Occup Environ Med 2000; 42:762–71. 74. Risch HA, Burch JD, Miller AB, et al. Occupational factors and the incidence of cancer of the bladder in Canada. Br J Ind Med 1988; 45:361–7. 75. Bulbulyan MA, Figgs LW, Zahm SH, et al. Cancer incidence and mortality among beta-naphthylamine and benzidine dye workers in Moscow. Int J Epidemiol 1995; 24:266–75. 76. Schoenberg JB, Stemhagen A, Mogielnicki AP, Altman R, Abe T, Mason TJ. Case-control study of bladder cancer in New Jersey. I Occupational exposure in white males. J Nat Cancer Inst 1984; 72:973–81. 77. Brownson RC, Chang JC, Davis JR. Occupation, smoking and alcohol in the epidemiology of bladder cancer. Am J Public Health 1987; 77:1298–300. 78. Decoufle P. Cancer risks associated with employment in the leather and leather products industry. Arch Environ Health 1979; 34:33–7.

79. Guberan E, Usel M, Raymond L, et al. Disability, mortality and incidence of cancer among Geneva painters and electricians: a historical prospective study. Br J Ind Med 1988; 46:16–23. 80. Miller BA, Silverman DT, Hoover RN, Blair A. Cancer risk among artistic painters. Am J Ind Med 1986; 9:281–7. 81. Steenland K, Palu S. Cohort mortality study of 57,000 painters and other union members: a 15 year update. Occup Environ Med 1999; 56:315–21. 82. Popp W, Schmieding W, Speck M, Vahrenholz C, Norpoth K. Incidence of bladder cancer in a cohort of workers exposed to 4-chloro-o-toluidine while synthesizing chlordimeform. Br J Ind Med 1992; 49:529–31. 83. Merlo F, Costantini M, Doria M. Cause specific mortality among workers exposed to welding fumes and gases: a historical prospective study. Sangyo I Daigaku Zasshi 1989; 11 (Suppl):302–15. 84. Ruder A, Fine L, Sundin D. National estimates of occupational exposure to animal bladder tumorigens. J Occup Med 1990; 32:797–805. 85. ‘t Mannetje A, Kogeninas M. Chang-Claude J, et al. Occupation and bladder cancer in European women. Cancer Causes Control 1999; 10:209–17. 86. Maffi L, Vineis P. Occupation and bladder cancer in females. Med Lav 1986; 77:511–4. 87. Silverman DT, Hoover RN, Mason TJ, Swanson GM. Motor exhaust-related occupations and bladder cancer. Cancer Res 1986; 46:2113–6. 88 Silverman DT, Hoover RN, Albert S, Graff KM. Occupation and cancer of the lower urinary tract in Detroit. J Natl Cancer Inst 1983; 70:237–45. 89. Gonzalez CA, Lopez-Abente G, Errezola M, et al. Occupation and bladder cancer in Spain: a multi-centre case-control study. Int J Epidemiol 1989; 18: 569–77. 90. Ward E, Carpenter, A, Markowitz S, Roberts D, Halperin W. Excess number of bladder cancers in workers exposed to orthotoluidine and aniline. J Natl Cancer Inst 1991; 83:501–6.

30.4 Renal Cancer Stuart L Shalat, Sandra N Mohr Renal cancer is relatively rare, accounting for only 3% of all cancers diagnosed in the US. It is the 10th leading cause of cancer in men and the 14th in women. Perhaps because it has received less attention than other, more common tumors, the etiology of renal cancer, with respect to both general and occupationally related causes, is still not well understood. However, approximately one-third to one-half of all cases of renal cell carcinoma occur in patients with von Hippel–Lindau disease.1 Other risk factors associated with renal cell carcinoma include smoking, obesity, history of kidney stones, and long-term use of phenacetin.2 Examination of risk of renal cancer in the workplace has included a variety of exposures, including metals (arsenic, cadmium, lead, and uranium), polyaromatic hydrocarbons (PAHs), organic solvents (including chlorinated compounds), asbestos, and other factors. Therefore, although renal cancer may be rare, the exposures that may confer increased risk to workers are not. In the US in 1996, there were approximately 30,600 incident cases of renal cancer with approximately 12,000 deaths.3 The disease is more common in men than women. Black men and women are at greater risk for the disease than whites. In 1997, age-adjusted incidence rates in the US were 12.2/100,000 in white males, 15.9/100,000 in black men, 6.4/100,000 in white women, and 7.9/100,000 in black women (SEER, 2000). The incidence rates for renal cancer peak in the sixth and seventh decades of life. Epidemiologic data from the National Cancer Institute’s Surveillance Epidemiology and End Results (SEER) pro-gram followed from 1973 to 1997 suggest that the incidence of cancer in the kidney is rising. There has been an increase of 1.7% per year for men and 2.2% for women. This increase, however, also may be due to improvements in diagnostic records. Only 41% of male cases and 46% of female cases were localized at the time of diagnosis. The 5-year survival rate for renal cancer has shown slight improvement from approximately 50% in the early 1970s (all gender and race groups) to approximately 61% for cases diagnosed in the early 1990s.

CLINICAL ASPECTS There are no distinguishing characteristics with regard to occupational etiology in the diagnosis and clinical course of renal cell carcinoma. Approximately 60% of cases present with hematuria, either gross or microscopic. The classic triad of flank pain, palpable abdominal mass, and hematuria is found in only 10% of patients. Most of the other presenting symptoms are systemic, including fever, weight loss, anemia, hypertension, and erythrocytosis. At the time of diagnosis, 30% of individuals have extensive metastatic disease. Renal tumors usually are detected by intravenous pyelography with tomography, or by ultrasound, followed up by a computed tomography (CT) scan. The CT scan is considered comparable to renal arteriography for the staging and characterization of the disease.

Classification Renal cell carcinoma, also known as renal adenocarcinoma or hypernephroma, accounts for 90% of tumors in the kidney. The cancer arises from cells of the proximal convoluted tubule and its associated connective tissue. The Mainz classification is now widely utilized for renal cancer, as a result of several cytogenetic studies.4 In this system, there are three histologically distinct forms of renal cell carcinoma: • clear cell type (cholesterol-filled cells); • chromophilic (eosinophilic or basophil-staining); and • chromophobic. Renal pelvic cancers involve the collecting systems of the kidney, and account for less than 10% of all renal neoplasms. These include collecting duct carcinoma and renal oncocytoma. The cancers usually are morphologically identical to transitional cell carcinomas seen in the bladder, and are associated with them in roughly 50% of cases. Wilms’ tumor, or nephroblastoma, is the third most common form of childhood cancer and represents about 5% of all kidney cancers. These tumors contain a variety of cell types, all derived from the mesoderm. The epithelial cell types may present as immature tubules or glomeruli.

Therapy As discussed above, the annual number of new diagnoses of renal cell carcinoma is increasing. For patients with localized primary tumors, surgical resection remains the mainstay of therapy. Since renal cell carcinoma is characterized by a lack of early warning signs, a majority of patients present with metastatic disease. Several reviews of available treatment regimens have concluded that renal cell carcinoma is resistant to chemotherapy.5–7 The result is that patients with distant metastatic disease (Stage IV) have a 5-year survival of less than 10%.8 Recent studies have suggested possible improved survival for advanced renal cell carcinoma from interferon and/or interleukin (IL)-2 therapy, and, more recently, allogenic stem-cell transplantation; however, far more research is required and new agents and treatment programs need to be considered.5,9

GENERAL EPIDEMIOLOGY Several epidemiologic studies have attempted to identify risk factors for renal cell carcinoma, which are shown in Table 30.4.1. These risk factors have been evaluated mostly through the use of case–control studies. Because of the high case-fatality rate occurring 1 year after diagnosis, the majority of these studies have had a fairly low response rate of 60% to 73%. In order to prevent a high percentage of nonresponse, some investigations have used proxy interviews

768 Renal Cancer Established Cigarette smoking Phenacetin analgesics Hereditary disorders von Hippel–Lindau disease

Suspected

Reported*

Obesity Hypertension and/or antihypertensive medications Northeastern European ancestry Long-term hemodialysis

Meat consumption Kidney infections/stones Acetaminophen analgesics

* Causal association requires further research

Table 30.4.1 Non-occupational risk factors for renal cell carcinoma

to substitute for deceased patients. This strategy, however, raises the problem of differential recall bias between proxy and direct patient interviews. These issues need to be taken into account when interpreting case–control studies of renal cancer. Cigarette smoking has been consistently associated with renal cancer across several epidemiologic studies. In a population-based case–control study of 495 cases and 697 controls, McLaughlin and coworkers found an odds ratio for smoking of 1.6 for men, and 1.9 for women.10 There was also a positive dose–response relationship in both sexes for years of tobacco use. Several other studies also found significant odds ratios for smoking in the 1.5–2.0 range. A large study from Australia recently confirmed the positive association for smoking, and estimated an attributable risk for renal cell cancer of 30% for men and 24% for women.11 Obesity has also been suggested as a risk factor for renal cell carcinoma. In a study by Yu and colleagues which assessed the effect of obesity, as measured by Quetelet’s index 10 years before the onset of cancer, significant increases in odds ratios for both men and women were observed.12 Maclure and Hankinson found that obesity was a predisposing factor for renal cell carcinoma among women in a case–control study of renal cell cancer.13 A nested case–control study (cases, n = 37; controls, n = 148) of refinery and petrochemical workers used body mass index (BMI) and found a statistically significant increase in risk of kidney cancer.14 A positive association of premorbid hypertension and the risk of renal cell cancer has been demonstrated, but it remains unclear whether the effect is due to an increase in blood pressure or use of antihypertensive drugs. McCredie and Stewart found that both beta-blocker use and hypertension for more than 2 years were independently associated with renal cell cancer, but the use of diuretics was not.15 In contrast, Yu and colleagues found that hypertension was not significantly associated with renal cell cancer after controlling for diuretic use in women.12 Several prospective studies have also investigated the relationship of hypertension, antihypertensive medications, and renal cell cancer, but only one of these demonstrated an effect due to antihypertensive medication independent of high blood pressure.16 In the same occupational study of refinery and petrochemical workers as previously mentioned, high blood pressure resulted in a large increase in risk, even when BMI was taken into account (OR = 4.5; 95% CI, 0.8–26).14 Phenacetin-containing analgesics have been consistently related to renal cell carcinoma in several epidemiologic

studies. Acetaminophen has not yet been investigated as extensively. While Sandler and colleagues demonstrated an association of acetaminophen with chronic kidney disease, the risk of renal cancer associated with this analgesic is not currently known.17 Hereditary forms of renal cell cancer have also been identified. These cancers tend to occur earlier and are more likely to be bilateral and multifocal. It has been hypothesized that tumor development in individuals with these cancers may be associated with an alteration of a tumor suppressor gene at the short arm of chromosome 3. Other risk factors that have been found to be associated with renal cell carcinoma include meat consumption, northeastern European ancestry, and prior kidney stones and infections. Coffee consumption has not been identified as a risk factor. Long-term users of renal dialysis, and those who develop acquired cystic kidney disease, have also been shown to have elevated numbers of renal cell cancer in several recent studies. These latter risk factors suggest that prior injury may predispose an individual to the development of renal cancer.

TOXICOLOGIC STUDIES Many known carcinogens – for example, asbestos and lead – have been found to cause renal cancer in animals. Animal studies in mice also have shown a clear dose– response relationship between unleaded petrol vapor and renal cell cancer, suggesting a carcinogenic role for gasoline. Several organic compounds have been shown to induce renal cancer in animals, including vinylidene chloride, dichloroacetylene, nitrosamines, PAHs, aromatic amines, coumarins, and various dyes. A toxicologic study by Walker and associates identified a gene that predisposes rats to develop renal cancer after exposure to a chemical carcinogen.18 Those rats that carried a single mutation on chromosome 4 had a 70-fold increased risk of developing renal cortical tumors compared with wild rats. These results suggest the involvement of tumor suppressor genes in the development of this cancer; a mutation in one allele predisposes susceptibility to the development of cancer if a spontaneous mutation by an environmental carcinogen occurs in another allele. This study provides valuable insight into the mechanisms of carcinogenesis in kidney cancer. Some studies have examined the role of glutathione S-transferase (GST) and variations in some of the GST genes in the etiology of renal cell carcinoma. A population-based case–control study found that those with GSTT1 null

Occupational Studies 769 gene were almost five times as likely to develop renal cell carcinoma.19

OCCUPATIONAL STUDIES Epidemiologic research on the role of occupational factors in the development of renal cancer has been conducted through both large occupational cohort studies of cancer mortality and case–control studies. The former have the advantage of relatively specific information on exposure, while usually containing minimal information on potential confounders (i.e., smoking, BMI), while the latter usually are able to address the confounders but are often limited by imprecise information on occupational exposures. In addition, these studies pos-sess other weaknesses, such as difficulty in assessing exposures, recall bias, and lack of statistical power due to the relative rarity of the cancer. While earlier large occupational cohort mortality studies were suggestive of increased cancer risk, recent studies have specifically addressed this issue. Since these studies have often not distinguished between different types of renal cancer, the following section only discusses occupational factors in relation to renal cell carcinoma. Table 30.4.2 shows the major exposures and occupations that have been examined in relation to renal cell carcinoma to date.

Hydrocarbons Polyaromatic hydrocarbons Little direct work has been conducted over the last decade to expand our knowledge on the role of PAHs in renal cell carcinoma. The combustion of organic materials, such as coke or coal, has been associated with renal cancer in several studies. Redmond and coworkers studied men who were employed between 1950 and 1955 at one

Occupations Aluminum pot-room workers Carbon electrode bakers Coke oven workers Dry cleaners Electric utility transformer repairmen Ferrosilicon and silicon metal workers Formaldehyde production Foundry workers Gas station attendants Insulators Lead battery manufacturing Lead smelter workers Newspaper pressmen Petrochemical and oil refinery workers Phenol production Rocket-test stand workers Sheet metal workers Uranium processing workers Welders

of 10 coke plants.20 After adjusting for potential confounders, they observed a relative risk of 7.5 for renal cell cancers among coke oven workers. Those workers were presumed to be exposed to byproducts of coke and coal combustion. Coal tar pitch and petroleum coke combustion volatiles also appeared to be involved in the high rate of renal cancer mortality seen in Rockette and Arena’s study of workers from 14 aluminum reduction plants.21 Those individuals who worked in an area where carbon anodes were prebaked were exposed to volatile agents containing PAHs and showed an increased standardized mortality ratio (SMR) for renal cell carcinoma. In a case–control study in Canada, the authors found an increased risk of 1.7 for exposure to burning coal, after controlling for smoking.22 Exposure to coal tar pitch also was found to be significantly associated with increased risk for renal cell cancer.

Petroleum-based hydrocarbons Since the International Agency for Research on Cancer (IARC) conducted its last comprehensive review of epidemiologic studies of oil refinery workers in 1989, and classified the workplace environment as containing probable human carcinogens, a number of new studies, updates of cohorts, and meta-analyses have been published.14,23–27 Petroleum products such as gasoline, chemicals found in distilled fuel oils or oil mist, and PAHs have been suggested as the possible carcinogens. Gasoline, in particular, has been suspected because of its high aromatic content. Recently, two papers examined both active and terminated workers and retired workers separately.23,28 These studies have observed a small increased risk for kidney cancer. In the active and current workers this reached statistical significance, with the point estimates for the SMRs virtually identical (active and terminated workers, SMR = 144; retirees, SMR = 140). A recent case–control

Chemicals

Probable

PAHs PAHs PAHs Perchloroethylene Organic solvents, PCBs Metal fume, organic solvents Formaldehyde Asbestos, metal fume, PAHs Gasoline Asbestos Lead Lead, arsenic Lead, organic solvents PAHs, gasoline, aliphatic hydrocarbons Phenol Hydrazine, TCE TCE Uranium, TCE Asbestos, cadmium, organic solvents

X X X

PAHs, polyaromatic hydrocarbons; PCBs, polychlorinated biphenyls; TCE, trichloroethylene

Table 30.4.2 Occupational risk factors for renal cell carcinoma

Possible

X X X X X X X X X X X X X X X X

770 Renal Cancer study of kidney cancer in Canada found odds ratios of 3.9 (95% CI, 1.6–9.8) and 3.6 (95% CI, 1.4–8.8) in workers with an exposure to jet fuel or aviation gasoline.29 Although further investigation is still clearly needed in this area, the available research does suggest that exposure to petroleum products in oil refineries, as well as in distribution, is likely to increase risk of renal cancer.

Organic solvents A number of case–control studies have also found a positive association between hydrocarbons and renal cell carcinoma. McLaughlin and colleagues demonstrated a statistically significant increased relative risk of 2.6 for workers with 20 years or greater on the job.10 In a case–control study of 210 cases of renal cell carcinoma, Kadamani and colleagues similarly found increased risk for moderate hydrocarbon exposure, with an odds ratio of 3.5 with high cumulative solvent exposure.30 A greater than 3.4-fold risk was also associated with exposure to mixed non-chlorinated solvents in a large Finnish study of 672 cases of renal cell cancer. Whether other specific organic materials are renal carcinogens in human populations continues to be an area of active investigation. A review of the question of the potential role of trichloroethylene and perchloroethylene in renal cell cancer failed to reach any definitive conclusions.31 The authors evaluated seven cohort studies and six case–control studies. Only three of the cohort studies and two of the case–control studies observed an excess, and in only one cohort study was the result statistically significant. It was concluded that future studies need to have large populations and quantitative exposure estimates to further characterize dose-related risk. Other studies have been conducted with regard to the potential role of gene interaction (GST) in workers exposed to trichloroethylene.32 Elevated odds ratios were observed for both those with GSTM1+ (OR = 2.7; 95% CI, 1.2–6.3) and those with GSTT1+ (OR = 4.2; 95% CI, 1.2–14.9). A more general review of this subject, that examined both the mechanistic issues as well as the occupational epidemiologic data, concluded that the issue was far from settled conclusively.33 With respect to other specific organic solvents, a cluster of renal cancer cases was observed in a group of workers employed in the repair and refurbishment of electrical transformers.34 The exposures included organic solvent systems used in epoxy coatings for transformers, as well as polychlorinated biphenyls (PCBs), with which the transformers were filled. The authors suggested that the solvents may have acted as initiators, with the PCBs potentially acting as promoters. A study of mortality in 8854 workers exposed to acrylamide observed only small increases in risk of kidney cancer (SMR 1.1; 95% CI, 0.7–1.6).35 The study also failed to observe a dose response; however, this is not surprising with only 22 cases of kidney cancer in the cohort. Perhaps more importantly, when workers with mean intensity of exposure >0.02 mg/m3 were considered (n = 5), the SMR increased to 1.7; although not statistically significant, this is suggestive of a possible causal linkage.

Metals Cadmium Cadmium, a potential animal carcinogen and a nephrotoxin, has also been associated with an increased risk of renal cancer. A case–control study by Kolonel identified 64 renal cancer cases in white males admitted to one hospital and 269 controls.36 A relative risk of 2.5 was found for workers exposed occupationally to cadmium. In addition, a synergistic effect was demonstrated between smoking and cadmium exposure, with a relative risk of 4.4. A 1991 study in Finland of 338 cases of renal cell carcinoma also found a suggestion of elevated risks in individuals exposed to cadmium.37 While studies are suggestive of an association of cadmium and renal cancer, further research is required to characterize this risk.

Lead A mortality study in a cohort of lead smelter workers has been reported by Selevan and coworkers.38 The exposure to lead in the smelting process was considered to be heavy, averaging over twice the consensus standard established in 1965 for air-borne lead exposure of 200 μg/m3. An SMR of 301 (98 to 703), associated with five deaths, was found for kidney cancer in this study. In addition, a significant SMR of 392 for chronic kidney disease was found for exposure lasting more than 20 years. Results for both chronic renal disease and renal carcinoma have been variable in subsequent studies. A recent review and meta-analysis by Steenland and Boffetta of eight studies of lead smelter and battery workers – 40 total cases of renal carcinoma – found little evidence of lead as a risk factor for kidney cancer, observing a relative overall risk of 1.01 (95% CI, 0.7–1.4), with only two of the eight studies demonstrating a risk greater than 1.39 The existing evidence for lead as a renal carcinogen thus remains inconclusive.

Steel, aluminum, ferrosilicon and silicon metal Studies at steel and aluminum reduction plants have shown elevated rates of renal cancer in some workers. In addition, a recent study of ferrosilicon and silicon metal facilities in Norway observed a small excess of cancer of the kidney and ureter in both furnace (standardized incidence ratio [SIR] = 1.3; 95% CI, 0.66–2.38) and nonfurnace (SIR = 1.7; 95% CI, 1.03–2.55) workers.40 As discussed earlier, however, these risks are thought to be due to PAHs generated in processing, as well as organic solvents used in cleaning the metals; the role of these metals remains unclear.

Asbestos A large number of studies have been carried out to examine the mortality of workers exposed to asbestos. While they have primarily focused on lung cancer, a number of studies have observed increased risk of kidney cancer as well. Selikoff’s study of 17,800 insulators found 18 deaths due to renal cancer when 8.1 were expected, yielding an SMR of 222, which was statistically siginificant.41 These

Wilms’ Tumor 771

There have been two studies that have observed an increased risk of renal cell carcinoma in workers processing uranium.43,44 While the former study observed a dose– response association with radiation exposure in general, studies of ionizing radiation have not noted similar increases in risk.45–49 In addition to ionizing radiation, other chemical exposures related to the processing may be confounders. Interestingly, in a study of 2514 workers at a uranium processing plant, in addition to renal cell carcinoma, the authors also observed increased mortality for chronic nephritis (OR = 1.9; 95% CI, 0.8–3.8).43

likely to be bilateral and show nephroblastomatosis on pathologic examination. The genetic form is not familial in nature but reflects mutations in the germ cells before conception. A case–control study of 88 Wilms’ tumor patients identified both maternal and paternal risk factors for Wilms’ tumor.51 The increase was highest in those whose fathers held jobs as machinists and welders during preconception (OR = 5.3; P = 0.006). Maternal use of hair dyes in the preconception period revealed a statistically significant odds ratio of 3.6. Other gestational risk factors included hypertension during pregnancy, tea drinking, and vaginal infection. Older maternal age was associated with the genetic form of Wilms’ tumor. In a study carried out in England of paternal exposures and childhood cancer, the only statistically significant finding was for exposure to pesticides (proportional mortality ratio [PMR] = 1.6; 95% CI, 1.2–2.2).52 A review was carried out of 48 published epidemiologic studies regarding childhood cancer and parental occupation.53 Of the 48 studies, only five addressed the issue of Wilms’ tumor, the previously cited Bunin et al. study51 among them. One study attributed the increased risk to parental exposure to lead, the others to various occupations, including manufacturing of wood and furniture, iron and metal structures, and electrical contracting.54 Another study found associations between maternal exposure to PAHs and Wilms’ tumor.51 The overall status of parental exposure and Wilms’ tumor, certainly suggestive of an association, warrants further investigation.

Other associated occupations

References

results were supported by two other long-term cohort studies of use of asbestos products in Italian factory and shipyard workers. A recent meta-analysis of 37 cohort studies was carried out at the IARC.42 As is often the case with cohort studies, the actual number of kidney cancer deaths was limited (n = 169). The pooled SMR for these studies was 1.1 (95% CI, 0.9–1.6). The positive epidemiologic studies are supported by several toxicologic experiments of asbestos exposure. It is clear that it is possible for asbestos fibers to reach the kidney, presumably through the blood or lymphatic system. Autopsy studies of individuals have found asbestos bodies in the kidney. Amphibole fibers have also been demonstrated in urine. In conclusion, while there is evidence for a causal association between asbestos exposure and renal cancer, the magnitude of the risk is likely to be relatively small compared with that for lung cancer.

Uranium

Several other occupations have been suggested to be associated with renal cell carcinoma, including ferrochromium workers, rocket-engine test-stand personnel, newspaper web pressmen, printers, and white-collar workers. The excess in kidney cancer (5 versus 1.6 expected) seen in one study of newspaper web pressmen is notable, given the substantial exposure to organic solvents as well as lead in this industry. A study of workers involved in rocket-engine testing found an association between exposure to hydrazine-based fuels and increased relative risk of death from bladder and kidney cancer (RR = 1.8; 95% CI, 0.6–5.1).50 The authors indicate that there was significant exposure to trichloroethylene in the same procedures that utilized the hydrazine. The increased risk in white-collar workers, such as physicians, found in several studies, is thought to be related to increased levels of other risk factors associated with renal cell carcinoma, including obesity, meat and fat consumption, analgesic use, as well as increased diagnostic surveillance.

WILMS’ TUMOR The general epidemiology of Wilms’ tumor has only recently been understood. There appears to be both a genetic and non-genetic form of this cancer. The genetic cases are more

1. Christenson PJ, Craig JP, Bibro MC, et al. Cysts containing renal cell carcinoma in von Hippel-Lindau diseases. J Urol 1982; 128:798-800. 2. McCredie M, Ford JM, Stewart JH. Risk factors for cancer of the renal parenchyma. Int J Cancer 1988; 42:13-16. 3. Sokoloff MH, deKernion JB, Figlin RA, et al. Current management of renal cell carcinoma. CA Cancer J Clin 1996; 46:284-302. 4. Diaz JI, Mora LB, Hakam A. The Mainz classification of renal cell tumors. Cancer Control 1999; 6:571-9. 5. Motzer RJ, Russo P. Systemic therapy for renal cell carcinoma. J Urol 2000; 163:408-17. 6. Motzer RJ, Vogelgang NJ. Chemotherapy for renal cell carcinoma. In: Raghaven D, Scher HI, Leibel SA, et al., eds. Principles and practice of genitourinary oncology. Philadelphia: Lippincott-Raven; 1991:885. 7. Yagoda A, Abi-Rached B, Petrylak D. Chemotherapy for advanced renal-cell carcinoma: 1983-1993. Semin Oncol 1995; 22:42-60. 8. Motzer RJ, Bander NH, Nanus DM. Renal-cell carcinoma. N Engl J Med 1996; 335:865-75. 9. Childs R, Chernoff A, Contentin N, et al. Regression of metastatic renal-cell carcinoma after nonmyeloblative allogenic peripheral-blood stem-cell transplantation. N Engl J Med 2000; 343:750-8. 10. McLaughlin JK, Mandel JS, Blot WJ, et al. A population-based case-control study of renal cell carcinoma. J Natl Cancer Inst 1984; 72:275-84. 11. McCredie M, Stewart JH. Risk factors for kidney cancer in New South Wales. I. Cigarette smoking. Eur J Cancer 1992; 28A:2050-4.

772 Renal Cancer 12. Yu MC, Mack TM, Hanisch R, Cicioni C, Henderson BE. Cigarette smoking, obesity, diuretic use and coffee consumption as risk factors for renal cell carcinoma. J Natl Cancer Inst 1986; 77:351-6. 13. Maclure M, Hankinson S. Analysis of selection bias in a case-control study of renal adenocarcinoma. Epidemiology 1990; 6:441-7. 14. Gamble JF, Pearlman ED, Nicolich MJ. A nested case-control study of kidney cancer among refinery/petrochemical workers. Environ Health Perspect 1996; 104:642-50. 15. McCredie M, Stewart JH. Risk factors for kidney cancer in New South Wales, Australia. II. Urologic disease, hypertension, obesity, and hormonal factors. Cancer Causes Control 1992; 3:323-31. 16. Grove JS, Nomure A, Stemmermann GN. The association for blood pressure with cancer incidence in a prospective study. Am J Epidemiol 1991; 134:942-7. 17. Sandler DP, Smith JC, Weinberg CR, et al. Analgesics use and chronic renal disease. N Engl J Med 1989; 320:1238-43. 18. Walker C, Goldsworthy TL, Wolf DC, Everitt J. Predisposition to renal cell carcinoma due to alteration of a cancer susceptibility gene. Science 1992; 255:1693-5. 19. Sweeney C, Farrow DC, Schwartz SM, et al. Glutathione S-transferase M1, T1, and P1 polymorphisms as risk factors for renal cell carcinoma: a case-control study. Cancer Epidemiol Biomarkers Prev 2000; 9:449-54. 20. Redmond CK, Ciocco A, Lloyd JW, Rush HW. Long-term mortality study of steelworkers. VI – Mortality from malignant neoplasms among coke oven workers. J Occup Med 1972; 14:621-9. 21. Rockette HE, Arena VC. Mortality studies of aluminum reduction plant workers; potroom and carbon departments. J Occup Med 1983; 25:549-57. 22. Sharpe CR, Rochon JE, Adam JM, Suissa S. Case-control study of hydrocarbon exposures in patients with renal cell carcinoma. Can Med Assoc J 1989; 140:1309-18. 23. Lewis RJ, Gamble JF, Jorgensen G. Mortality among three refinery/petrochemical plant cohorts. I. 1970 to 1982 active/terminated workers. J Occup Environ Med 2000; 42:721-9. 24. Lynge E, Andersen A, Nilsson R, et al. Risk of cancer and exposure to gasoline vapors. Am J Epidemiol 1997; 145:449-58. 25. Poole C, Dreyer NA, Satterfield MH, et al. Kidney cancer and hydrocarbon exposures among petroleum refinery workers. Environ Health Perspect 1993; 101(Suppl 6):53-62. 26. Rushton L. A 39-year follow-up of the U.K. oil refinery and distribution center studies: results for kidney cancer and leukemia. Environ Health Perspect 1993; 101(Suppl 6):77-84. 27. Wong O, Raabe G. A critical review of cancer epidemiology in the petroleum industry, with a meta-analysis of a combined databse of more than 350,000 workers. Regul Toxicol Pharmacol 2000; 32:78-98. 28. Gamble JF, Lewis EJ, Jorgensen G. Mortality among three refinery/petrochemical plant cohorts. II. Retirees. J Occup Environ Med 2000; 42:730-6. 29. Parent ME, Hua Y, Siemiatycki J. Occupational risk factors for renal cell carcinoma in Montreal. Am J Ind Med 2000; 38:609-18. 30. Kadamani S, Asal NR, Nelson RY. Occupational hydrocarbon exposure and risk of renal cell carcinoma. Am J Ind Med 1989; 15:131-41. 31. McLaughlin JK, Blot WJ. A critical review of epidemiology studies of trichloroethylene and perchloroethylene and risk of renal-cell cancer. Int Arch Occup Environ Health 1997; 70:222-31. 32. Bruning T, Lammert M, Kempkes M, et al. Influence of polymorphisms of GSTM1 and GSTT1 for risk of renal cell

33.

34.

35.

36. 37.

38. 39. 40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

cancer in workers with long-term high occupational exposure to trichloroethene. Arch Toxicol 1997; 71:596-9. Lash LH, Parker JC, Scott CS. Modes of action of trichloroethylene for kidney tumorigenesis. Environ Health Perspect 2000; 108:225-40. Shalat SL, True LD, Fleming LE, Pace P. Kidney cancer in utility workers exposed to polychlorinated biphenyls (PCB’s). Br J Ind Med 1989; 46:823-4. Marsh GM, Lucas LJ, Yuok AO, Schall LC. Mortality patterns among workers exposed to acrylamide: 1994 follow up. Occup Environ Med 1999; 56:181-90. Kolonel LN. Association of cadmium with renal cancer. Cancer 1976; 37:1782-7. Partanen T, Heikkila P, Hernberg S, et al. Renal cell cancer and occupational exposure to chemical agents. Scand J Work Environ Health 1991; 17:231-9. Selevan SSG, Landrigan PJ, Stern FB, Jones JH. Mortality of lead smelter workers. Am J Epidemiol 1985; 122:673-83. Steenland K, Boffetta P. Lead and cancer in humans: where are we now? Am J Ind Med 2000; 38:295-9. Hobbesland A, Kjuus H, Thelle DS. Study of cancer incidence among 8530 male workers in eight Norwegian plants producing ferrosilicon and silicon metal. Occup Environ Med 1999; 56:625-31. Selikoff IJ, Hommond EC, Seidman HA. Mortality experience of insulation workers in the United States and Canada, 1948-1976. Ann N Y Acad Med 1979; 330:91-116. Sali D, Boffetta P. Kidney cancer and occupational exposure to asbestos: a meta-analysis of occupational cohort studies. Cancer Causes Control 2000; 11:37-47. Dupree-Ellis E, Watkins J, Ingle JN, Phillips J. External radiation exposure and mortality in a cohort of uranium processing workers. Am J Epidemiol 2000; 152:91-5. Fraser P, Carpenter L, Maconochie N, et al. Cancer mortality and morbidity in employees in three United Kingdom Atomic Energy Authority 1946-86. Br J Cancer 1993; 67:615-24. Beral V, Fraser P, Carpenter L, et al. Mortality of employees of the atomic weapons establishment, 1951-82. BMJ 1988; 297:757-70. Cardis E, Gilbert ES, Carpenter L, et al. Effects of low doses and low dose rates of external ionizing radiation: cancer mortality among nuclear industry workers in three countries. Radiat Res 1995; 142:117-32. Carpenter L, Higgins C, Douglas A, et al. Combined analysis of morality in three United Kingdom nuclear industry workforces, 1946-1988. Radiat Res 1994; 138:224-38. Omar RZ, Barber JA, Smith PG. Cancer mortality and morbidity among plutonium workers at the Sellafield plant of British Nuclear Fuels. Br J Cancer 1999; 79:1288-301. Ritz B. Cancer mortality among workers exposed to chemicals during uranium processing. J Occup Environ Med 1999; 41:556-66. Ritz B, Morgenstern H, Froines J, Moncau J. Chemical exposures of rocket-engine test-stand personnel and cancer mortality in a cohort of aerospace workers. J Occup Environ Med 1999; 41:903-10. Bunin GR, Mass CC, Kramer S, Meadows AT. Parental occupation and Wilms’ tumor: results of a case-control study. Cancer Res 1989; 49:725-9. Fear NT, Roman E, Reeves G, Pannett B. Childhood cancer and paternal employment in agriculture: the role of pesticides. Br J Cancer 1998; 77:825-9. Colt JS, Blair A. Parental occupational exposures and risk of childhood cancer. Environ Health Perspect 1998; 106(Suppl 3):909-25. Olshan AF, Breslow NE, Daling JR, et al. Wilms’ tumor and parental occupation. Cancer Res 1990; 50:3212-17.

30.5 Cancer of the Liver and Gastrointestinal Tract Howard Frumkin, Patricia Blackwell The gastrointestinal tract, considered as a whole, is the most common site of cancer incidence worldwide.1 It is also the most common cause of cancer mortality worldwide, and in the US ranks second to cancer of the pulmonary system in this regard.2,3 This chapter considers occupational and environmental contributions to cancers at several sites: the esophagus, the stomach, the colon and rectum, the liver, the gallbladder, and the pancreas. These cancers are summarized in Table 30.5.1. The cancers discussed in this chapter share several common features. Most are adenocarcinomas, with the exception of esophageal cancer, in which both squamous cell cancer and adenocarcinoma are common. Most of these cancers exhibit strikingly variable incidence rates in geographic comparisons and migrant studies, suggesting substantial contributions from environmental factors. Diet, infectious agents, and alcohol and tobacco use generally have been implicated more than chemical or other workplace exposures. Most gastrointestinal cancers do not cause symptoms until they are relatively far advanced, resulting in poor prognoses. The primary treatment is surgical. Unfortunately, with the important exception of colorectal cancer, effective means of early detection are not available.

SPECIFIC DISORDERS Esophageal cancer Etiology An interesting feature of the epidemiology of esophageal cancer is its tremendous geographic variation, ranging over 500-fold from nation to nation and within nations. The highest rates occur in a belt from the Caspian littoral in Turkey and northern Iran, through the southern republics of Kazakhstan, Uzbekistan, and Turkmenistan, to northern and eastern China, as well as in parts of southern and eastern Africa, India, and Sri Lanka. Rates are much lower in most of Europe and North America. In the US, the incidence is higher in blacks than in whites, and is higher in males than in females. Throughout the world, esophageal cancer rates are higher in lower socioeconomic classes.4 This geographic and class variability strongly suggests environmental determinants, and many have been identified. Alcohol consumption and tobacco use are recognized as important causes, especially in developed nations and in Africa. In parts of China with a high incidence of esophageal cancer, domestic fowl also exhibit a high incidence of pharyngoesophageal tumors, suggesting a common environmental cause, perhaps nitrosamine contamination of water, fungal agents, or both. In India, the chewing of betel nuts is associated with esophageal cancer. The use of hot drinks also may be a risk factor. Dietary factors5 may play a role; some populations with a high incidence of esophageal cancer eat high-grain diets low in

vegetables, fruits, and animal products. Fumonisin B(1), a mycotoxin produced by Fusarium verticillioides that contaminates maize and other grains, especially in Africa, has been linked to esophageal cancer risk. Radiation is carcinogenic to the esophagus, and pre-existing conditions such as achalasia, lye injury, and Plummer–Vinson syndrome also are risk factors for esophageal cancer. Human papilloma virus is a suspected cause of esophageal squamous cell carcinoma, although results across studies have been inconsistent.6 Beginning in the early 1980s, a rising incidence of adenocarcinoma of the esophagus was noted, particularly in white males.7 While squamous cell carcinoma remains the predominant histologic type for esophageal cancer worldwide, in some centers adenocarcinoma has come to predominate. Most cases of adenocarcinoma of the esophagus occur in individuals with Barrett’s esophagus (intestinalized mucosa lining the lower esophagus8), related to gastroesophageal reflux. Of possible etiological factors investigated, obesity9 is the best established. Relatively few occupational exposures have been clearly associated with esophageal cancer.10 Table 30.5.2 shows industries, exposures, and occupations for which such observations have been made. However, few firm conclusions can be drawn at present with regard to the observations in Table 30.5.2 for several reasons: most of the observations are unique and have not been corroborated; opposing evidence exists in some cases; and confounders such as smoking, alcohol use, and social class often were not adequately controlled. Interestingly, several studies have shown that farmers have a lower-than-expected rate of esophageal cancer; the significance of this observation is unclear. There is relatively more information on asbestos exposure as a cause of esophageal cancer, but the evidence is equivocal.11 Similarly, considerable evidence is available on the rubber industry, but this, too, is equivocal. Because worker exposures often vary within the same facility, exposure misclassification may have obscured a true association in some studies. Straif and collaborators12 reported increased rates of esophageal cancer in rubber workers exposed to nitrosamines. Other notable observations include an association between esophageal cancer and two exposures, sulfuric acid and carbon black, in the large case–control study database at Institut Armand-Frappier in Quebec,13 an elevation of esophageal cancer among workers with potential exposure to silica dust and chemical solvents or detergents in data from the National Center for Health Statistics,14 and an association between cutting oils and cancer of several gastrointestinal sites, including the esophagus, in several studies.15

Diagnosis, treatment, and prognosis The diagnosis and treatment of esophageal cancer in the workplace do not differ from those of esophageal cancer in

774 Cancer of the Liver and Gastrointestinal Tract Origin

Approximate US incidence (/100,000)

Major cell type

Five-year survival (%)

3.9 6.7 43.9 4.3 1–2 8.8

Squamous/adenocarcinoma Adenocarcinoma Adenocarcinoma Adenocarcinoma, angiosarcoma Adenocarcinoma Adenocarcinoma

14.8 22.6 62.0 5.5 7 micrograms/g creatinine, blood Cd > 10 micrograms/l). Chest x-ray and pulmonary function testing are also recommended for cadmium-exposed workers.

If possible, other materials should be substituted for cadmium. When this is not an option, exposures should be reduced through engineering controls, improved work practices, protective clothing, good housekeeping, and other standard approaches. Workers should be educated about cadmium, so that they can take appropriate steps when exposed to cadmium-containing materials.

References 1. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological profile for cadmium. Atlanta, GA: US Department of Health & Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, 1999. 2. Shaham J, Meltzer A, Ashkenazi R, Ribak J. Biologic monitoring of exposure to cadmium, a human carcinogen, as a result of active and passive smoking. J Occup Environ Med 1996; 38:1220–8. 3. dell’Omo M, Muzi G, Piccinini R, et al. Blood cadmium concentrations in the general population in Umbria, Central Italy. Sci Total Environ 1999; 226:57–64. 4. Cabrera C, Ortega E, Lorenzo M-L, Lopez M. Cadmium contamination of vegetable crops, farmlands, and irrigation waters. Rev Envir Contam Toxicol 1998; 154:55–81. 5. Nogawa K, Kido T. Biologic monitoring of cadmium exposure in itai-itai disease epidemiology. Int Arch Occup Environ Health 1993; 65:S43–S46. 6. Nordberg GF, Jin T, Kong Q, et al. Biologic monitoring of cadmium exposure and renal effects in a population group residing in a polluted area in China. Sci Total Environ 1997; 199:111–4. 7. Townshend RH. A case of acute cadmium pneumonitis: lung function tests during a four-year follow-up. Br J Ind Med 1968; 25:68–71. 8. Roels HA, van Assche FJ, Oversteyns M, de Groof M, Lauwerys RR, Lison D. Reversibility of microproteinuria in cadmium workers with incipient tubular dysfunction after reduction of exposure. Am J Ind Med 1997; 31:645–52. 9. Järup L, Elinder CG. Incidence of renal stones among cadmium exposed battery workers. Br J Ind Med 1993; 50:598–602. 10. Staessen JA, Roels HA, Emelianov D, et al. Environmental exposure to cadmium, forearm bone density, and risk of fractures: prospective population study. Lancet 1999; 353:1140–4. 11. Buchet JP, Lauwerys R, Roels H, et al. Renal effects of cadmium body burden of the general population. Lancet 1990; 336:699–702. 12. Davison AG, Fayers PM, Taylor AJ, et al. Cadmium fume inhalation and emphysema. Lancet 1988; 1(8587):663–7. 13. Armstrong BG, Kazantzis G. Prostatic cancer and chronic respiratory and renal disease in British cadmium workers: a case control study. Br J Ind Med 1985; 42:540–5. 14. Elinder CG, Kjellstrom T, Hogstedt C, Andersson K, Spang G. Cancer mortality of cadmium workers. Br J Ind Med 1985; 42:651–5. 15. Sorahan T. Mortality from lung cancer among a cohort of nickel cadmium battery workers: 1946–84. Br J Ind Med 1987; 44:803–9. 16. Thun MJ, Schnorr TM, Smith AB, Halperin WE. Mortality among a cohort of US cadmium production workers – an update. J Natl Cancer Inst 1985; 71:325–33. 17. McDiarmid MA, Freeman CS, Grossman EA, Martonik J. Follow-up of biologic monitoring results in cadmium workers removed from exposure. Am J Ind Med 1997; 32:261–7.

39.5 Chromium Alfred Franzblau, David H Garabrant

EXPOSURE SETTINGS Metallic chromium was first isolated in 1797, and by the second decade of the 19th century chromium compounds had become important in industrial processes (e.g., potassium dichromate used as a mordant in textile dyeing). Following on soon after the discovery of their industrial utility were early descriptions of toxic effects of exposure to chromium compounds. Chromium and its compounds are widely used in industry. The toxic effects of chromium are related primarily to exposure to the hexavalent [chromium(VI)] compounds, which are called chromates. Chrome alloys such as stainless steel [zero valence, chromium(0)] and trivalent chromium [chromium(III)] compounds have important and widespread industrial applications, but these compounds are generally less toxic than the hexavalent forms. NIOSH has estimated that over 305,000 workers in the United States are potentially exposed to chromium compounds, of which 196,000 are exposed to hexavalent chromium compounds. In the occupational setting, exposure occurs by inhalation or skin contact with chromium-containing mist, dust, or fumes, or via direct contact with aqueous chromium solutions (e.g., in tannery workers). Ingestion of chromium salts or solutions can cause significant toxicity or death due to gastrointestinal irritation and renal failure, but this is not an important route of exposure in the industrial setting. Chromium is derived from chromite ores (e.g., iron dichromate, FeOCr2O3), and exposure can occur among workers employed in the various roasting processes involved in the processing of ore and production of common commercially useful compounds (e.g., sodium or potassium chromate and dichromate, ammonium dichromate, and chromium oxide). Significant exposure also occurs during stainless steel production, stainless steel welding, and among users of chromium compounds. Chrome electroplating operations provide an opportunity for exposure to mist of plating bath solutions, which usually contain chromium trioxide (CrO3), a potent irritant. Important exposure to chromium compounds also occurs in leather tanning (basic trivalent chromic sulfate, Cr2[SO4]3, plus hexavalent forms), production and use of chromium-based paint pigments (yellow pigments based on lead chromate and zinc chromate), metallurgic operations (grinding or welding chrome alloy materials), use of cement (many cements contain trace amounts of various soluble hexavalent chromium compounds), textile dyeing (chromic oxide [Cr2O3] and chromic sulfate), manufacture of refractory products, and glassmaking (chromic oxide). Chrome compounds have produced toxicity among workers in a variety of smaller industries, including lithography, wood preserving, match production, paper pulp manufacture, and crayon making.1 Total chromium exposures in US air are typically below 0.01 micrograms/m3 in rural areas and 0.01–0.03 micrograms/m3 in urban areas. Occasionally, levels above 0.5

micrograms/m3 have been measured. The proportion of total chromium comprising chromium(VI) varies but is often 20–30%. In the US, occupational permissible exposure limits are 1000 micrograms/m3 for chromium metal and insoluble salts, 500 micrograms/m3 for chromium(III) compounds, and 100 micrograms/m3 for chromic acid and chromates [carcinogenic chromium(VI) compounds]. Thus, ambient environmental exposures are typically at least two orders of magnitude below occupational exposures.2

CLINICAL EFFECTS Chromium(III) and chromium(VI) are the oxidation states of chromium most commonly encountered by humans. Water-soluble chromium(III) compounds are not well absorbed into the body and enter cells poorly. Water-soluble chromium(VI) compounds (such as CrO42–) are far more toxic because they are actively transported into cells in place of other structurally similar anions, such as phosphate (PO42–). Inside the cell, chromium(VI) is reduced via reactive intermediates ultimately to chromium(III), which binds tightly to cellular ligands and is largely responsible for toxicity. Chromium(III) is an essential nutrient for humans and is involved in the metabolism of glucose. Average daily consumption of Cr(III) is 60 micrograms/day from food, with intake of 50–200 micrograms/day recommended. Soluble hexavalent chromium compounds tend to be strong oxidizing agents. They also penetrate skin and mucous membranes more effectively than trivalent forms. These properties may explain, in part, the potent irritant properties of soluble hexavalent chromium compounds. Cutaneous exposure to mist, dust, or chromium solutions can result in simple skin irritation. The classic cutaneous chromium lesion is the chrome ulcer, or chrome hole. This lesion was first described in 1827 and may be a relatively painless, although deep, punched-out wound that can penetrate into underlying tissue layers including cartilage, but usually not bone. Secondary bacterial infections are common. On removal of the affected individual from exposure, healing may require months and leave a residual scar. These lesions occur at sites of greatest exposure, such as over bony prominences of the hands and forearms, although they have been described on the lower extremities when work practices have been particularly poor. They also tend to occur at sites of skin trauma, such as incidental nicks or cuts. They are not related to chrome sensitization.3 The mucous membranes also suffer irritant consequences, including erythema of the throat, conjunctival irritation, nasal itching and soreness, edema of the uvula, coryza, epistaxis, ulceration of the nasal mucosa, and nasal septal perforation. With very high exposures, a high prevalence of nasal septal perforations may occur (greater than 50% in some settings) with rapid onset, in some cases less than 1 week, following initiation of employment. Because

Prevention 959 the damaged cartilage is not the most anterior portion, there may not be any apparent external deformity of the nose. Sinusitis and nasal polyps also have been noted among exposed workers, and rare cases of perforation of the tympanic membranes have been described. Inhalation of chromium mist, dust, or fumes can produce acute bronchoconstriction, probably through a direct irritant mechanism. Chronic bronchitis and persistent abnormalities in pulmonary function have not been described following cessation of exposure. Allergic asthma due to chromium is uncommon, with fewer than two dozen cases documented in the published literature. Immediate, late, and dual bronchoconstrictive responses may occur. The immunologic mechanisms that have been reported for asthma include chromium-specific IgE (in one case of an immediate reaction) and a cell-mediated response (in one case of a delayed reaction).4 In addition to their direct irritant effects, chrome compounds (usually dichromates or hexavalent forms) also are potent cutaneous sensitizers.5 Chromium allergic contact dermatitis is the most common occupational dermatosis among male patients. Exposure can be occult owing to the ubiquitous nature of chromium. Eczematous skin lesions usually are found in areas of greatest exposure, such as the hands and arms. Sensitization to chromium tends to persist for years, even after cessation of exposure, based on persistence of positive patch tests among workers who have been sensitized6 (see also Chapter 29.1). A number of hexavalent chromium compounds are established human carcinogens, including calcium chromate, chromium trioxide, lead chromate, strontium chromate, and zinc chromate.7 Epidemiologic studies have documented increased rates of respiratory cancer (lung and nasal, primarily) among workers in the chromate production industry and the chromate pigment industry8 (see Chapter 30.2). Of interest, studies of stainless steel welders and in the ferrochromium alloy industry have given inconclusive results regarding cancer risks, even though exposure to chromium(VI) is appreciable in these settings.9 There is scant evidence that metallic chromium and chromium(III) compounds are carcinogenic in humans or animals and these materials are currently not classifiable as to their carcinogenicity to humans (IARC Group 3).1 Lung cancers associated with exposure to hexavalent chromium compounds have prolonged latencies (delay from onset of exposure), generally lasting more than 20 years. There is no unique or predominant histology. Although some studies have indicated excesses of other cancers (such as gastrointestinal cancers) in the bichromate producing industry and the chromate pigment industry, these results are inconsistent. Chromium(VI) compounds generally are not regarded as causing cancers outside the respiratory tract. Mild impairment of proximal renal tubular function, manifested by an increased excretion of β-2 microglobulin and retinol binding protein, has been documented among active chrome platers. There is no apparent effect on glomerular function, as reflected in the excretion of albumin. Tubular dysfunction appears to be reversible on cessa-

tion of exposure. Acute tubular necrosis with renal failure has been observed in rare cases following massive dermal absorption of chromium (e.g., falling into a vat of hot chromic acid). These observations of low molecular weight proteinuria and acute tubular necrosis raise the possibility that long-term exposure to chromates may be a risk factor for persistent renal injury.

DIAGNOSIS, EVALUATION, AND TREATMENT The pharmacokinetics of chromium in humans are complex. Absorbed hexavalent chromium is mainly excreted in the urine. With ongoing exposure, the half-life in urine ranges from 15 to 41 hours. Detailed kinetic studies suggest a three-compartment model for chromates, with half-lives of about 7 hours, 15–30 days, and 3–5 years. Therefore, measurement of chromium in urine can be used to document recent exposure to chromates, although the concentration of chromium in urine is not correlated with toxicity.10,11 The correlation between chromium(III) exposure and urine chromium levels is poor and urine monitoring is not felt to be useful for these exposures. In non-occupationally exposed populations urinary chromium is usually less than 5 micrograms/g of creatinine.2 Because of a lack of adequate data, blood, hair, and nail levels of chromium are not interpretable at present and should not be relied upon in assessing exposure.12 If not prevented, chrome ulcers should be identified early and workers with such lesions should be removed from exposure until complete healing has occurred. Treatment of other acute cutaneous and mucous membrane effects of chromium exposure also requires reduction of exposure. Patch testing can be used to distinguish between irritant dermatitis and allergic contact dermatitis. Topical steroid preparations may be beneficial in treatment of both entities. Because sensitization tends to be long lasting, affected workers may experience adverse social and economic consequences because of the need to find alternative employment. The clinical approach to and prognosis of patients with lung cancer who have had occupational exposure to chromium compounds are the same as for other patients with lung cancer. There is evidence that the chromium content of pulmonary tissue from chromium-exposed workers with lung cancer is higher than among persons without lung cancer matched for smoking history, and also is higher than those with lung cancer but without any history of occupational chromium exposure. However, the diagnostic value of such measurements in an individual case is unclear. The histologic appearance and cell types of respiratory cancers related to carcinogenic chromium(VI) exposure do not have any defining characteristics.

PREVENTION If alternative materials are not available, then engineering controls that minimize or eliminate chromium exposure must be applied. In addition, prevention of chrome ulcers

960 Chromium and other irritant effects of chromium compounds depends on the education of workers concerning the importance of industrial cleanliness and the use of personal protective equipment. Such measures should be supplemented with regular examinations of the skin and nose to detect early signs of irritant effects. Assessment of pulmonary function and chest radiographs is recommended in workers with inhalational chromium exposure. In some processes, substitution of trivalent chromium for hexavalent chromium, as in electroplating baths, can be beneficial in reducing the incidence of sensitization. Chromium sensitization among cement workers can be prevented by adding ferrous sulfate (FeSO4) immediately before mixing, when most skin contact occurs. The quality of the cement is not impaired, but the hexavalent chromium present in the cement is reduced and precipitated as chromium sulfate, an insoluble trivalent species. This procedure is already in practice in Scandinavia.

References 1. IARC. Chromium, nickel, and welding. Lyon, France: International Agency for Research on Cancer, 1990;49–256.

2. ATSDR. Toxicological profile for chromium. Washington, DC: US Department of Health and Human Services, 2000;1–419. 3. Burrows D. The dichromate problem. Int J Dermatol 1984; 23:215–20. 4. Bernstein IL, Nemery B, Brooks S. Metals. In: Bernstein IL, Chan-Yeung M, Malo J-L, Bernstein DI, eds. Asthma in the workplace, 2nd edn. New York: Marcel Dekker, 1999;501–21. 5. Fregert S. Chromium valencies and cement dermatitis. Br J Dermatol 1981; 105:7–9. 6. Peltonen L, Fraki J. Prevalence of dichromate sensitivity. Contact Derm 1983; 9:190–4. 7. National Toxicology Program. Ninth report on carcinogens. Research Triangle Park, NC: National Institute for Environmental Health Sciences, US Dept of Health and Human Services, 2000. 8. Langard S, Vigander T. Occurrence of lung cancer in workers producing chromium pigments. Br J Ind Med 1983; 40:71–4. 9. Moulin JJ, Wild P, Haguenoer JM, Faucon D, DeGaudemaris R, Mur JM. A mortality study among mild steel and stainless steel welders. Br J Ind Med 1995; 50:234–43. 10. Welinder H, Littorin M, Gullberg B, Skerfving S. Elimination of chromium in urine after stainless steel welding. Scand J Work Environ Health 1983; 9:397–403. 11. Lindberg E, Vesterberg O. Urinary excretion of chromium in chromeplaters after discontinued exposure. Am J Ind Med 1989; 16:485–92. 12. Lauwerys RR, Hoet P. Industrial chemical exposure. Guidelines for biologic monitoring, 2nd edn. Boca Raton, FL: Lewis Publishers, 1993;55–70

39.6 Nickel Alfred Franzblau, David H Garabrant

EXPOSURE First isolated in 1751, nickel is abundant in the earth’s crust (ranking 24th), and has multiple industrial applications. Current major industrial uses include stainless steel making, other alloys and superalloys, electroplating, plus lesser amounts used in cast iron, chemicals, batteries, ceramics and as industrial catalysts.1 It is also a nutritionally important element, although a specific deficiency state has not yet been defined in man. Most occupational nickel exposure occurs via inhalation of fume or dust, or skin contact with salts, alloys, or solutions. Diet is an important source of nickel exposure and is believed to be 100–800 micrograms/day (although less than 5% is typically absorbed), and contributes 80–94% of the total body burden of nickel among non-exposed populations.2 NIOSH has estimated that 250,000 persons in the United States have occupational exposure to inorganic nickel, and OSHA has estimated that 709,000 workers are possibly exposed to nickel and its compounds. Most exposed workers are involved in refining, smelting, machining, casting, welding, battery making, or grinding steel or other nickel alloys. Exposure to nickel salts is important in nickel plating operations. Release of air-borne nickel dusts from nickel refining and smelting operations has resulted in documented exposure to persons living downwind from such plants. Because of widespread use of nickel alloys in jewelry and other common consumer products, cutaneous exposure of the general population is prevalent. A small but significant problem is hypersensitivity of patients ‘exposed’ to implanted nickel-alloy bioprostheses; implanted prosthetic devices may employ alloys containing up to 35% nickel. A small number of workers have potential exposure to nickel carbonyl (Ni[CO]4). Formed by passing carbon monoxide over nickel-containing materials, it is a volatile, colorless liquid that boils at 43 degrees C. Sometimes described as ‘musty’, the gas lacks any strong or penetrating odor. This compound presents a considerable hazard in refining of nickel using the Mond process, gas plating, and also in synthesis of acrylates for paints and plastics. Exposure is usually via inhalation of the gas, although transcutaneous absorption of liquid presumably could be significant.

ABSORPTION, EXCRETION, AND BIOLOGIC MONITORING Absorption of nickel compounds is related to their solubility, with nickel carbonyl being highly soluble, most nickel salts being soluble, and nickel oxides and sulfides being relatively insoluble. Gastrointestinal absorption of soluble nickel compounds is rapid, but only about 1–5% is absorbed. Absorption of metallic nickel and insoluble nickel compounds from the gastrointestinal tract is very small.

Absorption and clearance of inhaled insoluble nickel compounds (for example, nickel compounds in welding fumes) from the lungs is slow, occurring over months. Dermal absorption of nickel and nickel compounds is minimal.3 Absorbed nickel is rapidly excreted in the urine, with a half-life of 20–60 hours. Nickel concentrations in plasma and urine reflect recent exposure to soluble nickel compounds and therefore can be used to monitor such exposures. In contrast, concentrations of nickel in plasma and urine do not reflect exposures to insoluble nickel compounds, thus limiting the utility of biologic monitoring for most industrial nickel exposures. Among non-exposed populations, mean nickel concentrations in serum/plasma are in the range 0.14–0.63 micrograms/l, in blood < 0.05–3.8 micrograms/l, and in urine 0.9–3.2 micrograms/l and 1.5–1.6 micrograms/g creatinine.4 The use of hair and fingernail samples as indices of internal exposure is not reliable and has not been demonstrated to have any clinical utility.5

CLINICAL EFFECTS Acute Nickel carbonyl is the most acutely toxic nickel compound. Acute exposure to nickel carbonyl results in a twophase illness that is dominated by pulmonary involvement. The first phase has almost immediate onset, and consists of multiple non-specific systemic complaints (fever, chills, nausea, lightheadedness, headache, dyspnea, and chest pain). These are easily confused with the ‘flu’ or acute bronchitis, and may resolve after a few hours. The second phase develops after 12–36 hours, but can be delayed up to five days. Acute chemical pneumonitis dominates the clinical picture, with cough, dyspnea, tachycardia, and cyanosis. Encephalopathy, due to cerebral edema and possibly hemorrhage, can also occur. Death may result from respiratory failure. Recovery from acute nickel carbonyl exposure can be prolonged because of slowly resolving pneumonitis. Additional acute responses resulting from exposure to nickel compounds include irritation of mucous membranes, contact dermatitis, and/or asthma in sensitized individuals. Allergic dermatitis can result from exposure to either nickel salts, such as nickel sulfate among nickel platers, or metallic nickel or nickel alloys (see Chapter 29.1). The latter is exemplified by cutaneous reactions to jewelry, buttons, and other clothing items that contain nickel. Up to 15% of women in the general population, who have no history of occupational nickel exposure, are sensitized to nickel, based on patch testing.6 Dentists may also have a high prevalence (16%) of nickel sensitization.7 Industrial ‘nickel itch’ can present in a number of ways, including: itching or burning papular erythema in interdigital areas with spread to exposed areas on the upper extremities; or papular or papulovesicular dermatitis with some

962 Nickel lichenification, similar to atopic dermatitis or neurodermatitis. Eruptions distant from the site of metal contact have also been described. Asthmatic reactions to nickel compounds, usually salts such as nickel sulfate, are unusual, but well documented.8 Reactions can involve immediate, delayed, and dual bronchospastic responses.

Chronic Increased incidence of cancer of the lungs and nasal cavities has been documented among nickel refinery workers in numerous epidemiologic studies from different geographic locations (see Chapter 30.2).9 Other malignancies in humans are not established to be causally associated with exposure to nickel and its compounds. The precise agent(s) responsible for nickel carcinogenesis in man is uncertain, but insoluble dusts (nickel subsulfide – Ni3S2; nickel oxides – NiO and Ni2O3), soluble dusts (nickel sulfate, nitrate, or chloride - NiSO4, NiNO3, NiCl2), nickel carbonyl, nickel pellets/dust, nickel acetate, nickel hydroxide, and nickel fluoride have been documented to be carcinogenic in various animal models. The International Agency for Research on Cancer (IARC) classifies nickel compounds as carcinogenic to humans (Group 1), based on evidence of carcinogenicity of nickel sulfate and the combination of nickel sulfides and oxides encountered in the nickel refining industry. Metallic nickel and nickel alloys have also been evaluated, with inadequate evidence for carcinogenicity in humans for both metallic nickel and nickel alloys, but sufficient evidence in animals for metallic nickel and limited evidence in animals for nickel alloys and nickel carbonyl. Thus, metallic nickel is classified as possibly carcinogenic to humans (Group 2B).9 The US National Toxicology Program (NTP) classifies nickel and certain nickel compounds (nickel acetate, nickel carbonate, nickel carbonyl, nickel hydroxide, nickelocene, nickel oxide, and nickel subsulfide) as reasonably anticipated to be human carcinogens.10 It should be noted that this classification is different from the designation given by NTP to agents that are known to be carcinogenic to humans. Overall, it appears that insoluble nickel compounds (oxides and sulfides) are carcinogenic, while the carcinogenicity of soluble nickel compounds by inhalation cannot be determined because of conflicting data.11 IARC has conducted a separate evaluation of the carcinogenicity of surgical implants and other foreign bodies, many of which contain nickel alloys. Implanted foreign bodies of metallic nickel and certain nickel alloys are classified as possibly carcinogenic to humans (Group 2B), while orthopedic implants of complex composition and dental materials are not classifiable as to their carcinogenicity (Group 3).12 As with most occupational malignancies, nickel-related tumors resemble those of non-occupational origin, but there are some features of nickel-related lung cancers and nasal cancers that are noteworthy. The sites of tumor occurrence appear related to airflow patterns and particle deposition in the nose and tracheobronchial tree.

Premalignant metaplastic changes in the nasal mucosa and nasal cancers typically originate at the anterior tips of the middle turbinates and spread along the lateral nasal walls and sinuses. The histology of sinus cancers among nickel workers is predominantly squamous cell. Nickel-related lung cancers tend to arise in the larger bronchi, and there may be a slight disproportionate increase in squamous cell histology.13 Despite numerous studies showing an increased risk of malignancy among nickel refinery workers, an increased risk of cancer among end-users of nickel compounds or products has not been documented unambiguously. Nevertheless, it is prudent to consider nickel dusts as potential human carcinogens, and to employ appropriate control measures. Repeated or chronic exposure to nickel carbonyl may induce pulmonary fibrosis, even in the absence of a history of acute intoxication and ARDS. Exposure to nickel aerosols (e.g., nickel sulfate – NiSO4) has been associated with chronic rhinitis, sinusitis, perforation of the nasal septum, and anosmia. These effects are irritant reactions, and not immunologically mediated. Workers in electrolytic nickel refineries and those employed in nickel plating operations are at greatest risk. Nickel alloys are frequently employed in various internally implanted medical devices, including orthopedic prostheses, cardiac pacemaker electrodes, replacement cardiac valves, and metal sutures. Such devices can induce cutaneous allergic reactions, including urticaria, eczematous changes, and pemphigoid lesions.

DIAGNOSIS, EVALUATION, AND TREATMENT Nickel carbonyl is the only nickel compound that has caused acute inhalation poisoning. Nickel carbonyl poisoning must be suspected on the basis of the history because prevention of acute poisoning, including pneumonitis, and a favorable outcome depend on early diagnosis and treatment. Sodium diethyldithiocarbamate (dithiocarb, DDC) is an investigational drug that has been an effective antidote for nickel carbonyl poisoning in animal studies when administered parenterally soon after exposure. Although it has been used in several hundred patients, there are no adequately controlled studies to establish its clinical effectiveness in the treatment of human poisoning by nickel carbonyl.14 An adult oral dose of 35–45 mg DDC/kg in divided doses in the first 24 hours, then 400 mg every 8 hours until the urine nickel concentrations are near normal and the patient is symptom free has been recommended.15 Critically ill patients may be treated parenterally with 12.5 mg sodium DDC/kg. Patients with more than 100 micrograms Ni/liter of urine in the first 8 hours postexposure, or in whom severe nickel carbonyl exposure is suspected, are candidates for treatment. Poisoning cases in which there is more than minor irritation of the upper respiratory tract should also undergo evaluation for acute chemical pneumonitis and carbon monoxide exposure.3

Prevention 963 Because of chemical similarity to disulfiram, patients treated with dithiocarb must avoid alcohol so as not to experience an ‘Antabuse’ reaction. Studies provide no support for use of other common metal chelating agents, such as BAL, penicillamine, or EDTA. Otherwise, the therapy of acute nickel carbonyl exposure is supportive. Patch testing with nickel sulfate can be used to distinguish irritant from allergic contact dermatitis due to nickel. Allergic reactions to nickel compounds generally mandate removal from exposure since even limited exposure of sensitized individuals will induce a response (i.e., asthma or contact allergic dermatitis). Enhanced in vitro lymphocyte proliferation using nickel acetate in lymphocytes obtained from sensitized individuals has been reported, but this test is an investigational tool and not widely available. Immediate response to skin prick testing may also be used to identify sensitized individuals. Routine screening of asymptomatic workers with patch tests to identify nickel sensitivity is not indicated. Cutaneous nickel hypersensitivity persists for years after cessation of exposure and ongoing dermatitis may persist because of contact with jewelry, buckles, and other apparel. Treatment of nickel hypersensitivity among persons with implanted prostheses may require removal and replacement with non-nickel containing devices.

PREVENTION Nickel and/or nickel compounds are carcinogens, therefore exposure to air-borne dust and fumes must be minimized. Nickel refinery workers should have regular medical examinations that include inspection of the nasal mucosa and chest radiography. Because of severe acute toxic potential, nickel carbonyl must be contained in enclosed systems to prevent human contact. Minimization of cutaneous exposure should reduce the likelihood of development of cutaneous sensitization or irritant effects.

References 1. ATSDR. Toxicological profile for nickel (update). Washington, DC: US Department of Health and Human Services, 1997;1–262. 2. Von Berg R. Toxicology update: nickel and some nickel compounds. J Appl Toxicol 1997; 17:425–31. 3. Barceloux DG. Nickel. Clin Toxicol 1999; 37:239–58. 4. Herber RFM. Review of trace element concentrations in biologic specimens according to the TRACY protocol. Int Arch Occup Environ Health 1999; 72:279–83. 5. Lauwerys RR, Hoet P. Industrial chemical exposure. Guidelines for biologic monitoring, 2nd edn. Boca Raton, FL: Lewis Publishers, 1993;55–70 6. Menne T, Borgan O, Green A. Nickel allergy and hand dermatitis in a stratified sample of the Danish female population: an epidemiological study including a statistic appendix. Acta Dermatovener 1982; 62:35–41. 7. Wallenhammar L-M, Ortengren U, Andreasson H, et al. Contact allergy and hand eczema in Swedish dentists. Contact Derm 2000; 43:192–9. 8. Malo JL, Cartier A, Doepner M, Nieboer E. Occupational asthma caused by nickel sulfate. J Allergy Clin Immunol 1982; 69:55–9. 9. IARC. Nickel and nickel compounds. Lyon, France: International Agency for Research on Cancer, 1990;257–445. 10. National Toxicology Program. Ninth report on carcinogens. Research Triangle Park, NC: National Institute for Environmental Health Sciences, US Department of Health and Human Services, 2000. 11. Haber LT, Erdreicht L, Diamond GL, Maier AM, Ratney R. Hazard identification and dose response of inhaled nickelsoluble salts. Regul Toxicol Pharmacol 2000; 31:210–30. 12. IARC. Surgical implants and other foreign bodies. Lyon, France: International Agency for Research on Cancer, 1999;1–256. 13. Sunderman FW, Morgan LG, Andersen AA, Ashley DL. Histopathology of sinonasal and lung cancers in nickel refinery workers. Ann Clin Lab Sci 1989; 19:44–50. 14. Bradberry SM, Vale JA. Therapeutic review: do diethyldithiocarbamate and disulfiram have a role in acute nickel carbonyl poisoning? J Toxicol Clin Toxicol 1999; 37:259–64. 15. Sunderman FW. Use of sodium diethyldithiocarbamate in the treatment of nickel carbonyl poisoning. Ann Clin Lab Sci 1990; 20:12–21.

39.7 Fluorine and Related Compounds Stephanie Carter, Noah S Seixas

SOURCES OF EXPOSURE Fluorine is a fairly ubiquitous element, forming 0.03% of the earth’s crust, predominantly as the mineral fluorspar (calcium fluoride), as cryolite (sodium aluminum fluoride), and in phosphate rock.1 Fluorine is the most electronegative of elements and is a powerful oxidizer, being able to form ionic or covalent fluoride compounds with all elements except helium, neon, and argon.2 Fluorine gas (F2) is a pale yellow gas used as a rocket fuel and in chemical processing to produce uranium hexafluoride, fluorocarbons, and other organic and inorganic fluoride compounds. Hydrogen fluoride (anhydrous, or in water as hydrofluoric acid), produced by the addition of sulfuric acid to calcium fluoride, is used as a catalyst in petroleum refining (alkylation), in etching glass, and metal cleaning in the electronics and semiconductor industries. Simple salts of fluoride (including sodium, calcium, copper, stannous, and zinc fluorides) find a wide range of uses in steel making, pesticides, drinking water and toothpaste additives, in electroplating, lumber treatment, glass, enamel, ceramics manufacture, and as lubricants and fluxing agents for welding. Other fluoride compounds have specialized uses in industry: cryolite as a flux in aluminum reduction smelters; magnesium fluosilicate in concrete and waterproofing applications; polychlorofluorocarbons (e.g., Freon and Halon) as aerosol propellants and fire suppressors; and fluoropolymers (e.g., Teflon, Tedlar) have numerous specialized applications. An estimated 183,000 US workers may be occupationally exposed to fluoride,3 while elevated community exposures may occur near aluminum smelters or phosphate fertilizer facilities, and during volcanic eruptions. Lower levels of exposure through toothpaste ingestion, fluoridated water supplies, and diet are almost ubiquitous in industrialized nations.4

ACUTE HEALTH EFFECTS Both fluorine and hydrogen fluoride (HF) gas are severe irritants to the eyes, mucous membranes, and respiratory tract (see Chapters 19.5 and 29.1). Upon contact with skin or mucous membranes, HF quickly dissociates to form hydrofluoric acid. Chemical burns of the skin can also occur from contact with F2 or HF. In the US, at least 1000 cases of exposure to liquid hydrofluoric acid and/or hydrogen fluoride occur each year.5 Fluorine concentrations between 100 and 200 ppm result in burns, which appear similar to a thermal burn, and heal relatively quickly. Skin contact with hydrofluoric acid produces deep burns, possibly contacting bone, which can be exceedingly slow to heal. Prompt medical treatment through water irrigation and neutralization is required to limit the severity of these burns. If the concentration and degree of exposure are sufficient, the fluoride ions can remove enough calcium and/or magnesium to induce cardiac arrhythmias, damage the liver and kidney,

and cause neuromuscular hyperirritability.5 Contact with hydrofluoric acid at concentrations greater than 50% produces immediate burning pain. At lower acid concentrations (especially below 20%), the symptoms may be unnoticed or delayed, thus delaying prompt treatment and allowing deep burns and systemic effects.6 Polymer fume fever is a flu-like acute response to the pyrolysis products of fluoropolymers, mold-release agents, and potentially other fluorocarbons (see Chapter 19.5). Symptoms of polymer fume fever include initial irritation followed 3–6 hours later by fever, chills, and general myalgia.7 The pyrolysis products thought to be responsible for the effects vary depending on the amount of oxygen available and the pyrolysis temperature. Of most importance is carbonyl fluoride, which converts to hydrogen fluoride upon contact with moisture in the respiratory tract. Smokers are at increased risk of polymer fume fever due to smoking fluoropolymer-contaminated cigarettes. Chemical pneumonitis and adult respiratory distress syndrome (ARDS) have also been reported after inhalation exposure from HF-containing consumer products.8

CHRONIC HEALTH EFFECTS The most well-known effect of chronic exposure to fluoride is bony fluorosis, an increased bone density or, in severe cases, skeletal deformities and spinal rigidity. The occurrence of fluorosis was first described in an occupationally exposed cohort in 1932 in which 30 of 78 cryoliteexposed workers had radiologic findings ranging from increased bone density to hyperosteoses.9 Skeletal changes and possibly associated musculoskeletal complaints have continued to be observed among relatively highly exposed working populations. Kaltreider et al10 reported a high prevalence of fluorosis among aluminum smelter workers with more than 10 years of exposure to fluoride levels of 2.4–6.0 mg/m3. A follow-up of a small group (n = 56) of these workers 7–8 years later showed no further increase in bone density.11 Possible and definite fluorosis was observed in 14% and 6.2%, respectively, of 2258 aluminum workers with relatively long exposure histories.12 One study of aluminum workers suggests that musculoskeletal symptoms and increased rates of fracture may occur even in the absence of radiologic evidence of fluorosis.13 Fluorosis occurs as a result of the selective affinity of fluoride for bone and formation of fluorapatite. Despite the increased bone density associated with fluoride exposure, the possible benefits of fluoride in the treatment of osteoporosis are uncertain, with some studies indicating a higher risk of bone fracture, and animal studies indicating lower bone strength associated with fluoride exposure. The benefits of small amounts of fluoride in the prevention of dental caries have been well demonstrated. However, higher fluoride exposures during tooth formation in early childhood results in striation and mottling of teeth, while still higher levels may result in more severe dental

Prevention 965 discoloration and pitting, with decreased tooth strength and an increase in cavities.4 Occupational asthma has frequently been observed among aluminum potroom workers exposed to fluorides.14 No specific causative agent has been identified in relation to asthma in this setting, though fluorides have been suspected. Current hypotheses are focusing on an irritantinduced mechanism with exposures to sulfur dioxide and HF as the primary suspected agents. Other than bony fluorosis, there is no clear evidence of chronic systemic health effects of fluoride exposures. Gastrointestinal symptoms have been observed among groups with accidental high exposures or oral ingestion of fluoride compounds, probably due to irritation or erosion of the gastric mucosa.4 Several epidemiologic studies have considered the effect of various fluoride compound exposures on cancer mortality in occupational and community cohorts. Each of these studies, however, has had important potential confounding exposures that may have explained any increased cancer rates observed. For instance, elevated cancer rates among aluminum smelter or fluoride workers15,16 are more likely associated with carcinogens in coal tar pitch volatiles than fluoride exposures. The International Agency for Research on Cancer (IARC) has reviewed the available literature on fluoride exposure through drinking water, and concluded that there was no evidence of carcinogenicity from epidemiologic studies and insufficient evidence from animal studies.17

PREVENTION Prevention of acute effects of fluorine or HF exposure is accomplished by preventing exposures, including shortterm peak levels, from exceeding established guidelines (Table 39.7.1). The odor threshold for both HF and F2 is approximately 0.1–0.2 ppm, exhibiting reasonable warning properties. In large part, the exposure limits are based

Exposure guideline source ACGIH – TLV NIOSH – REL NIOSH – IDLH OSHA – PEL

Fluoride (mg/m3)

Hydrogen fluoride (ppm)

Fluorine (ppm)

2.5TWA 2.5TWA 500 2.5TWA

3C 3TWA, 6STEL 30 3TWA, 6STEL

1TWA, 2STEL 0.1TWA 25 0.1TWA

TWA: Time weighted average, over 8 hours. STEL: Short term exposure limit, over 15 minutes. C: Ceiling, concentration not to be exceeded. REL: Recommended exposure limit. IDLH: Immediately dangerous to life or health.

Table 39.7.1 Occupational exposure guidelines

Reference (ACGIH 1996)1 (DFG 1998)22 (NIOSH 1975)23

on the prevention of irritation of the eyes and upper respiratory tract, although fluorosis is also prevented at these concentrations.1 The ACGIH1 and OSHA PEL18 for fluorine differ substantially (1 ppm and 0.1 ppm, respectively). This difference is based largely on the importance placed by OSHA on animal studies. ACGIH relied more heavily on more recent workplace studies that suggested that exposures above 0.1 ppm did not result in significant health effects or irritation. Hydrogen fluoride exposure limits are based on HF as a primary irritant of the eyes, skin, and respiratory tract. ACGIH recommends a more conservative ceiling limit (3 ppm) to prevent skin reddening and burning of the eyes and nose seen in human studies. The OSHA PEL and NIOSH REL designate both TWA and STEL limits based mainly on prevention of more severe respiratory irritation and fluorosis. Emergency response planning guidelines developed by the American Industrial Hygiene Association (AIHA) for HF and F2 indicate levels of exposure below which nearly all individuals can be exposed for 1 hour without experiencing mild or transient health effects, developing health effects that may impair ability to protect themselves, or experiencing life-threatening health effects. For HF, these limits are 2, 10, and 50 ppm, respectively; for F2, the limits are 0.5, 5, and 20 ppm.19 Irritation is reported at fluoride particulate concentrations of 5–10 mg/m3. Changes in bone density appear to be prevented by maintaining exposures less than 2.6 mg/m3 (ACGIH). All major US exposure-setting organizations concur with the 2.5 mg/m3 limit for particulate fluoride including major international organizations. However, the German Maximum Allowable Concentration (MAK) includes a short-term level of 12.5 mg/m3 averaged over a period of 30 minutes. Because respiratory protection is commonly used in areas of high fluoride particulate, biologic monitoring may be used to evaluate the body burden at a particular point in time. Table 39.7.2 provides a summary of the suggested biologic monitoring levels for fluoride. Urinary levels are corrected for concentration by adjustment either with creatinine concentration or specific gravity. Creatinine adjustment requires that the creatinine concentration be within specified limits. ACGIH recommends monitoring fluoride in urine and assuring that levels are maintained below 3 mg fluoride per gram creatinine for spot samples taken prior to a work shift, or 10 mg/g Cr at the end of the work shift. Background levels of fluoride in urine, related to food and water intake, rarely rise above 2 mg/l, with most below 1 mg/l. Air-purifying respiratory protection should utilize acid gas cartridges. For exposures lower than the exposure

Units

Pre-shift

Post-shift

Screening creat. (mg/l) or SG adjustment (g/ml)

(mg F/g creatinine) (mg F/g creatinine) (mg F/L)

3 4 4

10 7 7

Cr.: 300–3000 Cr.: 500–2500 SG: 1.024 g/ml

Table 39.7.2 Recommended biologic exposure indices for fluorides

966 Fluorine and Related Compounds Contact location

Water irrigation

Eyes Skin

15 minutes 5 minutes

Lung

NA

Neutralization

2.

1% calcium gluconate 2.5% calcium gluconate or 0.13% iced benzalkonium chloride 2.5% calcium gluconate by nebulizer

3.

From: Segal E. First aid for a unique acid: HF. Chemical Health Safety 1998; Sept/Oct: 25–8.

4.

Table 39.7.3 Guidelines for HF burn treatment5 5.

limits specially impregnated filters may also afford some protection. Use of full-face respirators has the advantage of providing protection from splashes and reducing eye irritation. Work areas where there is a potential for large gas releases of HF or F2 should have fixed-point gas alarms or other exposure indicators, and firewater deluge systems to minimize cloud formation. Additional information on management systems for safe operations and maintenance of HF operations, including methods to identify surface contamination and leaks, is available.20

6. 7. 8.

9. 10.

11.

EVALUATION AND TREATMENT Immediate medical treatment is necessary for skin or significant inhalation exposure to hydrofluoric acid/hydrogen fluoride, with focus on irrigation with water followed by neutralization gels or solutions6 (see Chapters 19.5, 29.1, and 20.1). Brief treatment guidelines are given in Table 39.7.3. Further assessment should include electrolyte levels (Ca, Mg, K), ECG if potential for cardiac arrhythmia is present, and chest x-rays for significant inhalation exposure or liquid contact with face/neck. HF can penetrate the fingernails without exhibiting external evidence, requiring advanced treatment through removal of the nails or calcium gluconate by IV. For further information on emergency treatment of HF burns, see Wilkes 2001.21 Workers chronically exposed to fluorides such as aluminum potroom workers who develop respiratory symptoms should undergo pulmonary evaluation for possible asthma, including spirometry and peak flow measurements (see Chapter 19.2). Evaluation and treatment are similar for other workers with non-sensitizing irritant induced asthma (see Chapter 19.2). Bony fluorosis should be considered in workers with prolonged higher exposures who develop musculoskeletal complaints (see Chapter 23.5). X-ray findings include sclerosis and exostoses. There is no specific treatment of fluorosis beyond removal from further exposure.

References 1. ACGIH. Fluorides. Documentation for the Threshold Limit Values and Biologic Exposure Indices for chemical substances in the work environment. Cincinnati, OH: American

12.

13. 14.

15.

16.

17.

18.

19. 20. 21. 22.

23.

Conference of Governmental Industrial Hygienists, 1996;657–9. Hawley G, Lewis R. Hawley’s condensed chemical dictionary. New York: Van Nostrand Reinhold, 1993. NIOSH. National occupational exposure survey (1980–1983). Cincinnati, OH: US Department of Health and Human Services, National Institute for Occupational Safety and Health, 1989. ATSDR. Draft toxicological profile for fluorine, hydrogen fluoride, and fluorides. Atlanta, GA: Agency for Toxic Substances and Disease Registry, US Department of Health and Human Services, 2001;308. Segal E. First aid for a unique acid: HF. Chemical Health Safety 1998; Sept/Oct:25–8. Bertolini JC. Hydrofluoric acid: a review of toxicity. J Emerg Med 1992; 10:163–8. Shusterman D. Polymer fume fever and other fluorocarbon pyrolysis-related syndromes. Occup Med 1993; 8:519–31. Bennion J, Franzblau A. Chemical pneumonitis following household exposure to hydrofluoric acid. Am J Ind Med 1997; 31:474–8. Moller F, Gudjonsson S. Massive fluorosis of bones and ligaments. Acta Radiol 1932; 13:269–94. Kaltreider N, Elder M, Cralley LV, Colwell MO. Health survey of aluminum workers with special reference to fluoride exposure. J Occup Med 1972; 14:531–41. Dinman B, Elder M, Bonney TB, Bovard PG, Colwell MO. A 15-year retrospective study of fluoride excretion and bony radiopacity among aluminum smelter workers – Part 4. J Occup Med 1976; 18:21–3. Czerwinski E, Nowak J, Dabrowska D, Skolarczyk A, Kita B, Ksiezyk M. Bone and joint pathology in fluoride-exposed workers. Arch Environ Health 1988; 43:340–3. Carnow B, Conibear S. Industrial fluorosis. Fluoride 1981; 14:172–81. Abramson M, Wlodarczyk J, Saunders NA, Hensley MJ. Does aluminum smelting cause lung disease? Am Rev Respir Dis 1989; 139:1042–57. Grandjean P, Olsen JH, Jensen OM, Juel K. Cancer incidence and mortality in workers exposed to fluoride. J Natl Cancer Inst 1992; 84:1903–9. Ronneberg A, Langmark F. Epidemiologic evidence of cancer in aluminum reduction plant workers. Am J Ind Med 1992; 22:573–90. IARC. Inorganic fluorides used in drinking-water and dental preparations. Lyon: International Agency for Research on Cancer, World Health Organization, 1982. OSHA. Code of Federal Regulations. Air Contaminants. Washington, DC: Occupational Safety and Health Administration, 1985;29. AIHA. Emergency response planning guidelines. Fairfax, VA: American Industrial Hygiene Association, 1999. API. Safe operation of hydrofluoric acid alkylation units. Washington, DC: American Petroleum Institute, 1999;48. Wilkes G. Hydrofluoric acid burns. eMedicine 2001; May 7, 2(5). http://www.emedicine.com/emerg/topic804.htm DFG. List of maximum allowable concentrations (MAKs) and biologic tolerance values (BATs) for 1998. Munich, Germany: Commission for the Investigation of Health Hazards of Chemical Compounds in the Work Area, 1998. NIOSH. Criteria for a recommended standard: occupational exposure to inorganic fluorides. Washington, DC: US Department of Health, Education and Welfare, Public Health Service, National Institute for Occupational Safety and Health, 1975;191.

39.8 Lead Jacqueline M Moline, Philip J Landrigan Lead is an ancient metal. Because of its malleability and low melting point, humans have used lead since prehistoric times to make statues, jewelry, water pipes, and drinking vessels.1 The salts of lead have long been valued for their brilliant colors. The quantity of lead used since 1940 surpasses the total used in all previous centuries. This heavy recent use reflects industrial applications as well as the use of lead as a fuel additive. In the mid-1970s, nearly 200,000 tons of lead were consumed annually in gasoline in the United States.2 Virtually all of this lead was emitted into the environment from vehicles in a finely particulate form and caused widespread contamination of air, dust, soil, drinking water and food crops. In the United States and in many western European countries, the use of leaded gasoline has decreased sharply since 1976, and population mean blood lead levels have decreased by 90% as a result.3 Leaded gasoline continues, however, to be used widely in Eastern European and developing nations.4

OCCUPATIONAL LEAD EXPOSURE A wide variety of industrial populations are at risk of occupational exposure to lead (Table 39.8.1). In the United States more than 3 million workers are estimated by NIOSH to be potentially exposed to lead in their work. Occupational exposure to lead is regulated in the United States by the OSHA lead standard, first promulgated in 1978. This standard sets limits on air-borne lead exposure as well as on permissible levels of lead in blood. It has been successful in reducing the most serious industrial exposures to lead. However, this standard is enforced inadequately and occupational exposure to lead remains widespread in the United States. Workers at highest risk are those in smaller industries such as secondary lead smelters and automobile radiator repair shops. A major industrial group that previously had been exempted from the protection of the United States lead standard were construction workers. Major outbreaks of lead poisoning were observed among workers who cut leadpainted steel structures such as bridges with oxyacetylene torches and among renovation workers who remove lead paint from older homes. The heating and burning of lead paint by these workers can create a highly respirable fume consisting of air-borne lead particles less than 1 μm in diameter; inhalation of lead fumes can produce acute lead poisoning within days or even hours. In 1993, the OSHA lead standard was extended to cover the construction and demolition trades. A sharp decline in their exposures to lead and their blood lead levels resulted.5,6 Occupational exposure to organic lead, principally tetraethyl lead (TEL), can occur in the manufacture of this compound for use as a fuel additive as well as in the cleaning of tanks and pipes that have contained TEL. Typically, in TEL manufacturing, exposures to organic and inorganic lead occur simultaneously.

Measurement of occupational exposure Assessment of occupational exposure to inorganic lead requires measurement of lead levels in both air and blood. Air lead levels are measured by standard industrial hygiene techniques. The determination of lead in air provides direct information on sources and levels of exposure, information that is essential for controlling exposures to lead in the workplace. The OSHA standard (1993) for the permissible exposure level of lead in workplace air is 50 micrograms/m3, expressed as an 8-hour time-weighted average.7 As is discussed below, this standard can no longer be considered adequately protective of workers’ health, because lead toxicity has been observed in workers whose lead exposures do not exceed the current OSHA standard, and because fetal lead poisoning can occur in the offspring of female lead workers well below this level.8 Measurement of the blood lead level complements determination of lead levels in air. The finding of an elevated blood lead level identifies those workers who may have had uniquely high exposures that are not detected by air monitoring. For those workers in the United States covered by the OSHA lead standard, anyone found to have a blood lead level of 50 micrograms/dl or above must be removed from the high-exposure job (termed ‘medical removal protection’). A worker is not allowed to return to a high-exposure job until the blood lead level has fallen to below 40 micrograms/dl. To prevent workers with elevated lead levels from being economically penalized during medical removal protection, the OSHA standard requires that these workers must continue to receive their full rate of pay (pay rate retention) while they are working in the less hazardous job.

ENVIRONMENTAL LEAD EXPOSURE In the general population, children are the group most heavily exposed and most susceptible to the toxic effects of lead. The mean blood lead level of children in the United States is about 3 micrograms/dl.9 A decline of 90% in blood lead levels occurred among American children from 1976 to 1997 as a consequence of removal of lead from gasoline. In contrast, lead use in gasoline as well as in industry remains widespread in nations undergoing transition to industrialization; environmental contamination and blood lead levels of workers and residents of communities near polluting industries in these countries have been reported to be dangerously elevated.4 According to the Centers for Disease Control and Prevention (CDC), blood lead levels of 10 micrograms/dl and higher in children indicate excessive exposure.10 Blood lead levels of 10 mg/dl and above are associated with subclinical lead toxicity in children. More recently, consideration has been given to further decreasing the blood lead standard for children in light of recent data showing subclinical toxic-

968 Lead Battery makers Brass workers Bronzers Cable makers/splicers Chemical operators Demolition workers Firing range personnel Foundrymen Glass makers/polishers Gunstock/barrel makers Jewelers Lead burners Painters Pigment makers Pipe cutters Pottery workers Printers (lino/electrotype) Ship burners Solderers Stained glass makers Tetraethyl lead manufacturers Welders Table 39.8.1 Major occupations associated with risk of inorganic lead poisoning

ity at blood levels below 10 mg/dl.10a,10b The CDC estimates that over 400,000 children in the United States have blood lead levels above 10mg/dl. The proportion of children with elevated blood lead levels is particularly high among minority inner-city populations, and among such children, the CDC estimates that as many as 35% of preschool children still have elevated blood lead levels.9 Lead-based paint is the major concentrated, high-dose source of lead for American children. Most cases of acute, symptomatic pediatric lead poisoning are caused by ingestion of dust derived from lead-based paint. Although new applications of interior lead-based paint are banned, an estimated 27.1 million housing units in the United States built before 1950 still contain lead-based paint. Since 1979, the maximum allowable concentration of lead in domestic paint has been regulated by the United States Consumer Product Agency at 0.06 parts per million. Air, dust, soil, drinking water, cosmetics, dietary and herbal supplements, and soiled parental work clothing are additional potential sources of exposure to lead for children and for all members of the population. Children, because of their normal oral exploratory behavior, are at high risk of ingesting lead from dust and soil.11,12 In 2001 the US Environmental Protection Agency (EPA) developed standards for lead in dust and soil, which were now consistent with existing US Department of Housing and Urban Development guidance levels for home dust levels. This was critical, since Lanphear et al found that a strong relationship between interior dust level landing and children’s blood lead levels exists below the levels in the prior guidelines of 100 micrograms/ft2 for floors, 500 micrograms/ft2 for interior window sills, and 800 micrograms/ft2 for window troughs.13 The new dust standard is 40 micrograms/ft2 for floors and 250 micrograms/ft2on window sills; soil levels must be below 400 parts per million (ppm) in play areas and 1200 ppm average for bare soil in the rest of the yard. Current lead in drinking water is due principally to the widespread use of lead solder to join pipe in water

service lines. Acidic drinking water can leach lead from solder and also from lead pipes. An analysis of sources of lead poisoning in urban children found that in infants younger than 1 year of age the principal sources were household renovation with lead paint removal (40%), ingestion of lead paint chips (20%), preparation of infant formula with lead-contaminated drinking water (18%), exposure to lead-contaminated parental work clothing (2%), congenital poisoning (2%), and sources unknown (18%). By contrast, the same study found that sources of lead poisoning for older children (18 through 30 months) were home renovations (4%), paint chip ingestion (87%), and sources unknown (9%).14 Cases of lead poisoning through ingestion of contaminated herbal supplements or infant formula prepared in lead-contaminated vessels have been documented.15 The habit of sniffing leaded gasoline is an important source of lead poisoning for children in certain rural populations.16

Measurement of environmental exposure Determination of lead levels in blood is the principal means of assessing the lead exposure of children. This assay should be conducted in a laboratory that has been licensed by the CDC or by comparable authorities and that participates satisfactorily in quality control and quality assurance programs. Hand-held portable blood lead instruments have been developed for use in the field. They complete the analysis in 2–3 minutes, appear highly reliable, and have been approved for use since 1997.9 Assessment of lead exposure in the general environment requires analysis by standard chemical and physical techniques. Lead levels in paint may be determined by portable, hand-held x-ray fluorescence (XRF) detectors. Determination of lead levels in air is undertaken under the requirements of the EPA’s air lead standard; the community air lead standard in the United States is 1.5 micrograms/m3 of air, computed as a quarterly average. Lead in drinking water is most reliably determined in a first-draw morning sample, when water has had an opportunity to sit in contact overnight with any lead in pipes or in solder. The current EPA safe drinking water standard for lead in drinking water is 0.15 mg/l (15 parts per billion). Lead concentrations in dust and soil are determined by standard chemical techniques.

CLINICAL EFFECTS IN ADULTS Acute inorganic lead toxicity Intense occupational exposure to lead over a brief period of time can cause a syndrome of acute lead poisoning.17 Characteristics of this syndrome are abdominal colic, constipation, fatigue, hemolytic anemia, peripheral neuropathy, and in some cases, alteration of central nervous system function. A gingival lead line is occasionally seen. In profound cases, full-blown acute encephalopathy with coma, convulsions, and papilledema may occur; in milder cases,

Clinical Effects in Adults 969 only fatigue, headache or personality changes may be evident as manifestations of the neurologic toxicity of lead. Mild liver involvement, limited to modest elevations in serum glutamic-oxaloacetic transaminase (SGOT), and musculoskeletal pains and arthritic symptoms are common. Basophilic stippling may be seen in the red blood cells. Usually, these finding occur when the blood lead level is very high, 70 micrograms/dl or more.18,19

Chronic inorganic lead toxicity Chronic inorganic lead intoxication is an insidious illness that is highly prevalent among persons chronically overexposed to lead. Symptoms include arthralgias, myalgias, headache, weakness, depression, loss of libido, impotence, and vague gastrointestinal difficulties. The classic abdominal colic of acute lead poisoning is usually absent. Anemia may occur, accompanied in some cases by an increased reticulocyte count. Thyroid and adrenal function often are depressed. Hypertension may be present. Sperm counts can be depressed in men and fertility reduced in women. Measurement of renal function may reveal impairment of glomerular clearance; decreased tubular urate clearance resulting in hyperuricemia is common. Late effects of inorganic lead intoxication include chronic renal failure, gout, chronic encephalopathy, and hypertension.19,20 The definition of chronic lead intoxication has, in recent years, been expanded by the recognition that lead can cause toxic injury at levels of exposure that only a decade ago were thought to be safe. The term subclinical toxicity denotes this concept that relatively low-dose exposure to lead may cause harmful effects to health in the absence of symptoms and signs that are evident on the standard clinical examination. The underlying premise is that there exists a spectrum of toxicity in which clinically apparent effects have their asymptomatic subclinical counterparts, which are detected through special testing.21,22

The effects of lead on organ system function Hematologic toxicity Anemia is a classic clinical manifestation of acute lead toxicity (see also Chapter 21).17 The severity and prevalence of lead-induced anemia are correlated directly with the blood lead level in both adults and children.23 It is important to note that the presence of anemia is not required for a diagnosis of lead poisoning; at lower levels of excessive lead exposure the hemoglobin concentration is most often within the normal range. Most patients with chronic and late-effects of lead poisoning also have normal blood hemoglobin levels. Lead-induced anemia is the result of multiple mechanisms: impairment of heme biosynthesis, acceleration of red blood cell destruction, and an inhibition of erythopoietin production in the kidney. Lead-induced inhibition of heme biosynthesis is due first to inhibition of the cytoplasmic enzyme, δ-aminole-

vulinic acid dehydratase (ALA-D). This effect is dose related. It is noted initially at blood lead levels of 10–20 micrograms/dl and is virtually complete at levels of 70–90 micrograms/dl. The mitochondrial enzyme ferrochelatase is the next most sensitive enzyme in the heme biosynthetic pathway inhibited by lead. Ferrochelatase catalyzes the transfer of iron from ferritin into protoporphyrin to form heme. Inhibition of ferrochelatase causes increased excretion of coproporphyrin in the urine and accumulation of protoporphyrin (a zinc-requiring enzyme, also called zinc protoporphyrin [ZPP]) in the erythrocytes.24,25 Lead-induced acceleration of red blood cell destruction leading to hemolysis reflects toxicity to the red blood cell membranes and results probably from the inhibition of lead by the enzymes 5’-nucleotidase and ATPase. Erythropoietin, a glycoprotein hormone principally secreted by the proximal renal tubule, regulates red blood production. Chronic exposure to lead appears to inhibit secretion of erythropoietin, and may also act directly on erythrocyte progenitor cells in the bone marrow.26

Neurologic toxicity Neurologic toxicity of lead in adults In the peripheral nervous system, the myelin sheaths surrounding the motor axons are the principal target of lead (see also Chapter 28). Lead-induced pathologic changes in these fibers include segmental demyelination and axonal degeneration. Extensor muscle palsy with wrist drop or ankle drop hasbeen recognized as the classic clinical manifestation of the peripheral motor neurologic toxicity of lead since the time of Hippocrates. Unlike other toxic neuropathies, this motor neuropathy may show left–right asymmetry. Mild to moderate sensory dysfunction also may be evident in lead neuropathy, and in such cases the clinical picture is one of mixed motor–sensory neuropathy, often with asymmetric features. Recent studies of peripheral neurologic function in persons exposed to lead have used electrophysiologic probes to determine whether exposures insufficient to produce clinically evident signs and symptoms can cause covert abnormalities.27 With the development of increasingly sensitive test methodologies, these studies have been able to detect peripheral neurologic toxicity at progressively lower blood lead levels. For example, Seppalainen et al, in a prospective study of new entrants to the lead industry, reported slowing of motor conduction velocity in the ulnar nerve at blood lead levels as low as 30–40 micrograms/dl.28 In the central nervous system, high-dose acute exposure to lead causes encephalopathy characterized clinically by obtundation, coma, and convulsions. Among persons who recover from acute lead encephalopathy, a high proportion are left with significant neuropsychologic deficits, including reduced intelligence, shortened attention span, and instability of mood.23 The question of whether lead causes asymptomatic impairment in central neurologic function in adults at doses insufficient to produce clinical encephalopathy has been the subject of extensive research. Early investigations of this issue have shown a correlation between lead exposure and diminished neuropsychologic performance in a

970 Lead group of asymptomatic adult workers, all of whom had blood lead levels below 70 micrograms/dl.29 The functions most severely impaired were those dependent on visual intelligence and visual-motor co-ordination. An increased prevalence of fatigue and short-term memory loss was observed in lead smelter workers that increased as blood lead levels rose. Other studies also found evidence of depressed mood and other central neurologic deficits in lead workers with blood lead levels in the range 40–70 micrograms/dl.23 Stewart et al reported that past exposure to lead, assessed by elevated bone lead content, was associated with persistent decrements in neurobehavioral test results in a group of organolead workers.30 Additional studies are required to confirm this finding, to elucidate toxic mechanisms, and to ascertain whether or not toxic effects occur at lower blood lead levels. A further implication of the finding that lead may cause insidious injury to the central nervous system is the possibility that some as yet unknown proportion of cases of presenile dementia, motorneuron disease (amyotrophic lateral sclerosi [ALS]),30a and other chronic neurologic and psychiatric conditions in adults may be the result of chronic exposure to lead. Such exposure may result in an accelerated attrition of neurons that becomes clinically evident with age.30 These possible associations are currently the focus of research efforts. Pediatric neurologic toxicity of lead Lead is now recognized to cause subclinical neurologic toxicity in children at much lower levels of exposure than in adults (see also Chapter 28). The earliest suggestion of this toxicity came from studies by Byers and Lord in the 1940s of the survivors of acute lead poisoning. These early investigations demonstrated the presence of permanent neuropsychologic sequelae, including learning disabilities and shortening of attention span as well as highly aggressive behavior in children who had recovered from acute lead encephalopathy.31 The existence of subclinical lead neuropathy in asymptomatic children was then confirmed in a series of crosssectional neuroepidemiologic studies undertaken by Needleman and colleagues in the mid-1970s.32 These landmark investigations found an association between asymptomatic increased exposure to lead and decreased intelligence. Children with elevated body lead burdens, corresponding to blood lead levels of 25–55 micrograms/dl, were found to have mean verbal IQ scores 4.5 points lower than similar children with lower lead burdens. This deficit remained evident when results were adjusted for parental education, childhood illness, socioeconomic status, and other biologic and social variables. Subsequent long-term follow-up studies have shown that these deficits in IQ are associated with learning difficulties, reading problems, an increased rate of failure to graduate from secondary school and an increase in violent behavior. Needleman has noted that a 4.5 point downward shift in mean IQ produces a 50% reduction in the number of gifted children (IQ above 120) in a population, and a 50% increase in the number of retarded children (IQ below 80).32,33

The data from these cross-sectional studies have been further corroborated by data from a series of prospective epidemiologic studies of newborn children, followed from birth through the first several years of life. Each of these studies found that cord blood lead levels at birth as low as 15–20 micrograms/dl were associated with decrements in childhood intelligence.34–37 Lead poisoning has also been associated with decreased hearing. An inverse relationship between blood lead levels and height has also been noted. Most recently, an analysis of data from the US National Health and Nutrition Evaluation Survey (NHANES) suggests that a negative association between lead and neurobehavioral function is evident in children at blood lead levels as low as 5–10 micrograms/dl.38 In recognition of the extreme sensitivity of the developing nervous system to low levels of lead, the CDC has periodically revised the definition of childhood lead poisoning. Currently, the CDC considers a blood lead level of 10 micrograms/dl or higher as evidence of excess lead burden.10 (The previous definition of lead toxicity, prior to 1992, was a blood lead level of 25 micrograms/dl or above.) Children with blood lead levels of 10 micrograms/dl or above must be provided medical follow-up and environmental monitoring, as is described in the section on clinical management. The therapeutic goal is to identify and reduce the child’s environmental exposures to lead, so as to reduce the threat of subclinical neurotoxicity and to prevent re-exposure. In the near future, CDC may further reduce the blood lead standard for children.10a,10b

Renal toxicity The cells lining the proximal tubules of the nephron appear to be the tissue in the kidney that is most highly sensitive to lead (see also Chapter 25).20 In children with acute, high-dose lead poisoning, injury to these cells occasionally results in the development of acute Fanconi syndrome, in which glucose, amino acids, and phosphates are spilled in the urine in large quantities. This syndrome typically is reversible, and it is not known whether or not it predisposes the individual to long-term sequelae. Fanconi syndrome is not commonly seen in adults exposed to lead. At blood lead levels below 25 micrograms/dl, lead inhibits the metabolic activation (hydroxylation) of vitamin D, a transformation that takes place in the cells lining the proximal tubules.39 Also in these cells, at blood lead levels of 40–80 micrograms/dl, lead induces the formation of dense intranuclear inclusion bodies consisting of a lead protein complex.20 These inclusion bodies persist for approximately 1 year. Hyperuremic gout, apparently resulting from increased reabsorption of uric acid by the tubular cells, is a third metabolic consequence of leadinduced renal impairment. Chronic nephropathy, which may progress to kidney failure, is the classic late manifestation of chronic lead toxicity in the kidneys. It appears to result from long-term, relatively high-dose exposure to lead, but dose–response relationships have not been well defined. Individuals with chronic low-level lead exposure have abnormal urinary excretion of

Clinical Effects in Adults 971 n-acetyl glucosaminidase (NAG) compared to controls, suggesting renal tubular dysfunction may occur after chronic low-level lead exposure. Population-based studies, such as the Normative Aging Study, also demonstrated a deleterious effect of lead in kidney function when body burdens of lead were assessed.40–42 The evolution of lead nephropathy usually is silent. The central event appears to be the progressive destruction of tubular cells and their replacement by fibrosis. Clinically obvious manifestations of impairment, such as elevations in BUN and serum creatinine, do not ordinarily become evident until 50–70% of the nephrons have been destroyed. Pathologically, chronic lead nephropathy is characterized by interstitial fibrosis with atrophy and dilation of the tubules and relative sparing of the glomeruli; intranuclear inclusions are infrequent, and the kidneys are small and shrunken. The pathologic picture may be indistinguishable from changes of chronic hypertension, which usually coexists (see later).20 Excess mortality from renal disease has been observed in several epidemiologic studies of lead workers. In each of these investigations, a two- to threefold increase has been noted in deaths from chronic nephritis, and in one of the investigations a positive association was observed between duration of employment in a lead smelter and mortality from chronic nephritis.43,44

Lead and hypertension Hypertension has long been associated with lead toxicity, in both animal and epidemiologic studies. Early in the last century, long-term, high-dose exposure to lead was reported to be associated with an increased incidence of hypertension and cerebrovascular disease.17 With the reduction in lead exposure that has occurred in most industries, these overt associations are now noted less commonly. Also, these exposures may have occurred many years prior to the development of hypertension, making this causal association less obvious. However, population studies have investigated the effect of blood lead, even relatively low blood lead levels, on hypertension. An analysis of the NHANES II survey data revealed that lead was a significant contributor to hypertension in the US population.45 Hypertension has also been associated with elevated bone lead levels.45a In addition, Schwartz et al found left ventricular hypertrophy was correlated with blood lead levels, demonstrating that lead was not merely raising blood pressure but also contributing to end-organ damage.46 An analysis of the data from the Normative Aging Study also found an association between body burdens of lead and hypertension.41 Different mechanisms for lead-induced hypertension and renal failure have been proposed, including an increased responsiveness to α-2 adrenoreceptor stimulation after lead ingestion, and alterations in cAMP-dependent availability of calcium for contractile processes in vascular and cardiac myocells. Other potential mechanisms of action include increases in endothelin-3 concentration, decreases in endothelium-derived relaxing factor, and alterations in renin activity.47,48

From the perspective of disease prevention, it is most important to prevent hypertension in lead workers. Elevations in blood pressure, even at low levels, are associated with significantly increased rates of mortality from heart disease and stroke. A biomedical surveillance program of workers exposed to lead, therefore, should incorporate as a specific aim the prevention of elevated blood pressure. Periodic determination of blood pressure should be a routine component of the biomedical monitoring of workers exposed to lead.

Reproductive toxicity of lead Male lead workers, especially those newly exposed to lead, can have evidence of decreased fertility (see also Chapter 27.2). Decreased sperm counts, abnormal sperm morphology, and decreased sperm motility have been noted in several studies when lead levels were above 40 micrograms/dl.18,49 Animal studies indicate that blood lead concentrations greater than 30–40 micrograms/dl are associated with impairment of spermatogenesis and reduced concentrations of androgens. Changes in endocrine profiles have been noted in men with lead levels over 60 micrograms/dl. Men currently employed in lead trades may have reduced fertility, as measured by increased time to conception in their wives.50 The effects of lead on male reproductive function at lower levels are more uncertain. In women, lead at high doses can also be toxic to reproductive function (see also Chapter 27.2). Clinical reports, most of them from the first half of the 20th century, describe an increased incidence in spontaneous abortion among female lead workers as well as in the wives of male lead workers.51 Low-level maternal lead exposure may increase the risk of spontaneous abortion.52 A difficult problem in the prevention and clinical management of the reproductive toxicity of lead is the finding that lead may cause neurologic damage to the fetus at blood levels as low as 5–10 micrograms/dl, substantially below the current OSHA standard for blood levels of lead in workers. Lead passes virtually unimpeded across the placental barrier, and the neurologic impairment that it produces in the fetus appears to be irreversible.34,35 This problem is further compounded by the fact that under the metabolic stress of pregnancy, lead may be mobilized from bony stores of the mother to enter her blood and thus the blood of the fetus.52a Thus, if a woman who has previously had chronic lead exposure or who has suffered lead poisoning becomes pregnant, there may be mobilization of skeletal lead with resultant fetal exposure. It is not known whether or not exposure of the father to lead before conception results in fetal injury. The prevention of fetal lead encephalopathy requires that blood lead levels of prospective mothers be kept below 10 micrograms/dl, not only during pregnancy but also in the years preceding conception.21 If this principle is followed, exposure of the fetus to currently absorbed lead, as well as to lead released from previously accumulated bony stores, should be prevented. If a woman who has previously had elevated blood lead levels wishes to become pregnant, it may be reasonable to recommend either an x-

Clinical Evaluation 973 the amount of stored lead. For this reason, the blood lead level is of only limited value in assessing dose–response relationships for the long-term consequences of lead exposure such as renal disease, hypertension, and chronic neurologic impairment. Also, the storage and later release of large amounts of lead in bone account for the long persistence of elevated blood lead levels in older retired lead workers who have ceased exposure to lead. K-shell x-ray fluorescence (XRF) analysis is now a wellestablished technique for accurate, non-invasive and relatively rapid assessment of chronic lead exposure. It directly measures levels of lead in bone.58,59 The radiation dose delivered is de minimus (approximately 1/500th of the average dental x-ray effective dose for a tibia lead measurement of an adult, and approximately 1/20th for a 1-year old).58 Zinc protoporphyrin (ZPP) or free erythrocyte protoporphyrin (FEP) levels reflect the toxic effect of lead on the erythrocytic enzyme ferrochelatase. FEP levels begin to rise in adults when blood lead levels exceed 30–40 micrograms/dl. Once elevated, FEP levels tend to stay above background for several months, which is substantially longer than acutely elevated blood lead levels. A combination of normal FEP value and high blood lead level indicates very recent exposure that has not yet had opportunity to poison heme metabolism. The long persistence of elevations in ZPP creates a problem in the use of this test as a screening tool for lead exposure in occupationally exposed adults. Because the ZPP level remains elevated long after the blood lead level has fallen, this test does not discriminate between recent and past exposure. Therefore, it is of only ancillary value in evaluating occupational lead exposure. In the past, the gold standard for determination of total body lead burden was the measurement of urinary lead excretion after intramuscular injection of 1 g of calcium EDTA (given mixed with lidocaine). Several studies have now shown that chelatable lead in general, and EDTAchelatable lead in particular, bear little relation to bone lead stores.30,60 Bone lead measurements using the x-ray fluorescence technique60a have therefore become the de facto standard for assessing body lead burden, because they sample the overwhelmingly largest pool of lead in the body. Chelatable lead remains the clinical gold standard but may be better viewed as an indicator of soft tissue bioavailable lead burden. No other standard clinical test or laboratory finding is either very sensitive or specific for lead poisoning. Clinically evident anemia occurs only in relatively faradvanced cases of lead poisoning; thus, the diagnosis of lead poisoning cannot await the appearance of anemia. Basophilic stippling in the red blood cells is seen in most patients with acute hemolysis and occasionally in patients with chronic hemolytic syndromes; it is not specific to lead poisoning. Lead lines on the gums are uncommon except in cases of severe poisoning and may be difficult to recognize in patients with poor dental hygiene. Complete blood counts and routine blood chemistry determinations should be performed as baseline measure-

ments in asymptomatic workers exposed to lead. Other baseline tests that may be of value include creatinine clearance, BUN, and uric acid. Blood pressure should be measured regularly. Blood lead levels should be obtained on a periodic basis in lead workers. If the whole blood lead level exceeds 40 micrograms/dl, these tests should be repeated monthly; blood lead levels over 50 micrograms/dl require immediate removal of workers from lead-exposure, and they may not return to work in a lead-exposed job until there have been two lead-levels below 40 micrograms/dl. Measurement of the urine lead level is of little value in monitoring populations exposed to inorganic lead; the test is too highly variable to be reliable. Similarly, hair lead analysis is not considered reliable for monitoring purposes.61

Clinical evaluation strategy for children Preschool children in the United States should be screened routinely for lead poisoning according to guidelines developed by the CDC and the American Academy of Pediatrics.62,63 These guidelines replace universal screening for all children, and target screening to geographic areas with older housing stock, and to communities where a substantial proportion of 1 and 2 years olds have elevated blood lead levels. Determination of the blood lead level is the principal technique for assessing increased lead absorption in children. Recent evidence suggests this can be accomplished safely and accurately using a capillary tube sample obtained by needlestick. In some areas, an alternative is measurement on a filter paper. Elevated levels must be confirmed on a venous sample and tested in an accredited laboratory. Guidelines for the interpretation of blood lead levels in children were developed by the CDC. Because recent data suggest that adverse health effects may occur in children at blood lead levels as low as 5–10 micrograms/dl, the CDC guideline of 10 micrograms/dl may need to be revised.10a,10b Currently, the CDC guidelines define a level above 10 micrograms/dl as excessive.10 A multi-tier approach for the evaluation of childhood blood lead levels has been developed by CDC and is reproduced in Table 39.8.2; this table also indicates the followup activities appropriate for each level of exposure.10 Provocative or challenge chelation with EDTA can be used to determine whether a child with an initial blood lead level of 25–44 micrograms/dl will respond to chelation therapy with a brisk lead diuresis. A dose of 500 mg EDTA/m2 is administered intravenously over 1 hour. Urine is then collected for 8 hours, and the test is considered positive if the ratio of lead excreted (in micrograms) to the amount of EDTA infused (in milligrams) is greater than 0.6.64 This test is recommended for use only in centers where large numbers of lead-exposed children are routinely evaluated. Children whose blood lead levels are 45 micrograms/dl and above should not receive a provocative chelation test; instead, they should be referred immediately to a pediatric center for appropriate chelation therapy. The use of XRF to assess bone lead stores is beginning to replace challenge chelation as a tool for assessing body lead burden in children.22

974 Lead BLL* (μg/dl)