Maternal-Fetal Toxicology, Third Edition,: A Clinician's Guide (Medical Toxicology)

ISBN: 0-8247-0378-2 This book is printed on acid-free paper. Headquarters Marcel Dekker, Inc. 270 Madison Avenue, New York, NY 10016 tel: 212-696-9000; fax: 212-685-4540 Eastern Hemisphere Distribution Marcel Dekker AG Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland tel: 41-61-261-8482; fax: 41-61-261-8896 World Wide Web http:/ /www.dekker.com The publisher offers discounts on this book when ordered in bulk quantities. For more information, write to Special Sales/Professional Marketing at the headquarters address above. Copyright  2001 by Marcel Dekker, Inc. All Rights Reserved. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage and retrieval system, without permission in writing from the publisher. Current printing (last digit): 10 9 8 7 6 5 4 3 2 1 PRINTED IN THE UNITED STATES OF AMERICA

Series Introduction

Our generation is characterized by an increased awareness of the effects of xenobiotics and environmental agents on humans. The introduction of unique but toxic medications, the development of novel methods of estimating short- and long-term adverse effects, and tighter regulatory demands have all contributed to new insights in the area of medical toxicology. At present, the main sources of information in the field of medical toxicology are reviews of original research and textbooks. Reviews of research generally address a narrow aspect of one issue, whereas textbook reviews need to generalize, within a limited space, which often results in decreased quality. This new series of books aims at filling a gap between these two sources of information. I have aimed at in-depth reviews of areas of interest in medical toxicology, recruiting leading international scientists to discuss the themes. The planned volumes address issues in acute poisoning, chronic adverse drug reactions, regulatory aspects, and so on. We hope to bring new frontiers of medical toxicology to the clinical and scientific communities in a way that will improve the understanding of the complex relationships between xenobiotics and human health. Gideon Koren

iii

Foreword

Almost all mothers will do everything they can to ensure that their babies have the healthiest possible start in life. Although birth defects or congenital malformations have been reported in the earliest recordings of history, teratology, the study of environmentally induced birth defects, did not develop into a modern science until the middle of the 20th century. Although congenital malformations were considered to be serious medical problems, they were thought to be unpreventable and untreatable. With the discovery in the 1940s that certain environmental factors (iodine deficiency and rubella virus) could cause congenital malformations, many embryologists developed an active interest in substances called teratogens that could induce or increase the incidence of human congenital malformations. The occurrence of the thalidomide tragedy in the late 1950s attracted worldwide interest. The publicity accorded this tragedy increased the awareness of the public and the medical and legal professions that some human birth defects are caused by drugs, chemicals, and radiation. It was not long before women, and some medical practitioners, viewed every drug, chemical, and virus as a reproductive hazard. As revealed in this book, many pregnant women exposed to drugs and chemicals that are known not to be teratogenic believe that they have a high risk of having a malformed baby. For a woman who has been exposed to a possible teratogen, obtaining accurate information about the risk involved, if any, is of utmost concern. Information available from newspapers, magazines, and popular books is often frightening and misunderstood. Most physicians and other health professionals receive insufficient training to answer all the questions pregnant women ask about the effects of chemical, physical, or infectious agents on the developing human embryo or fetus. Teratogen information programs have been established in major health centers to v

vi

Foreword

respond to pregnant women’s inquiries about teratogens, and to advise physicians and other health professionals who are responsible for their care. The third edition of Maternal–Fetal Toxicology: A Clinician’s Guide will be greatly appreciated by all who desire to learn about the teratogenic risks of the many environmental factors that women are exposed to during pregnancy. The main purpose of this book is to present those who counsel women with accurate, up-to-date estimates of the teratogenic risks of exposure to drugs, chemicals, viruses, and radiation during pregnancy. This information is very important to women making a decision to continue or terminate a pregnancy. For many years there has been a need for a practical book on teratogens that can be used in health-care programs for pregnant women. This book describes current teratogen information programs and the process of counseling for teratogenic risk. I strongly recommend this book to all who share my enthusiasm for the developing human and my desire to see every one of them have the best possible chance of developing normally. Keith L. Moore, Ph.D., F.I.A.C. Professor of Anatomy The University of Toronto Toronto, Ontario, Canada

Preface

Approximately one fetus is aborted for every two or three children born in Western countries. Since the thalidomide diaster over four decades ago, medicine has been practiced as if every drug were a potential human teratogen. Women exposed to nonteratogens commonly believe they have a high teratogenic risk. Their physicians often encourage them to terminate their pregnancies, yet only about 30 drugs and chemicals have been proven to be teratogenic. Every year scores of new drugs and hundreds of new chemicals are introduced into the market. In neither case are human reproductive effects known. Furthermore, according to different studies, between 40 and 90% of pregnant women consume one or more medications during gestation. Finally, despite hundreds of scientific studies published yearly on reproductive effects of xenobiotics, little has been done in the past to crystallize a clinical approach to deal with these issues. No single medical specialty is equipped to deal with the complex issues of reproductive toxicology. While geneticists commonly deal with congenital malformations, it is unlikely that they have the experience of pharmacologists in tailoring alternative therapy. Neither group is trained to evaluate such factors as occupational exposures. A multidisciplinary team of pharmacologists, toxicologists, geneticists, obstetricians, neonatologists, occupational and addiction specialists, drug information specialists, psychologists, sonographers, and epidemiologists is needed. In September of 1985 we counseled the first patient in the Motherisk clinic in Toronto. This program was designed to inform, counsel, and follow up pregnant women exposed to drugs, chemicals, or radiation in pregnancy. In order to perform these tasks, new approaches and clinical tools had to be developed by our multidisciplinary team; these are presented in this volume. vii

viii

Preface

Much of the confusion surrounding the counseling process stems from the wellunderstood ‘‘do not use in pregnancy’’ statements commonly found in the Physicians’ Desk Reference, Compendium of Pharmaceuticals and Specialties, or their equivalents. Since many women are inadvertently exposed to medications before finding out they have conceived, the ‘‘do not use in pregnancy’’ statements are easily translated into ‘‘harmful’’ or ‘‘teratogenic.’’ While the exact rate of pregnancy termination due to fears of adverse fetal effects of xenobiotics is not known, there is indirect evidence that this is not uncommon. Similarly disturbing are the many cases where women are exposed to drugs and chemicals known to adversely affect the fetus without being appropriately informed. The goal of this book is to assist the large number of health professionals who are asked by women and their families to provide answers on potential reproductive effects of xenobiotics and radiation. These include general physicians, obstetricians, poison control specialists, geneticists, occupational specialists, pediatricians, pharmacologists, toxicologists, pharmacists, nurses, and others. With increasing public awareness of environmental toxins, it is likely that concerns surrounding reproductive toxicology will increase over the next few decades. We hope that an appropriate clinical approach will help put the issue in its correct perspective, by avoiding both understatements and ambiguity. In the six years that have elapsed since publication of the second edition of Maternal–Fetal Toxicology: A Clinician’s Guide, the field of teratology information has seen an exponential growth, both in terms of quantity and quality of knowledge. The Motherisk Program in Toronto, in collaboration with several American services, has initiated and brought to completion large-scale, prospective studies on the safety of drugs in pregnancy, some of which are presented in this volume. In addition to revising and updating most chapters, we have added several novel aspects, such as systematic reviews and metaanalyses of specific xenobiotics. This revised and expanded edition includes clinical cases at the beginning of most chapters, with answers at the end. It is hoped that this problembased approach will enhance the clinical relevance of this book as well as increase its educational value. Finally, I wish to thank the members of the Motherisk team, my students, fellows, and colleagues for being active in generating new knowledge in the field of maternal– fetal toxicology, and making any edition obsolete within a few years. Gideon Koren

Contents

Series Introduction Foreword Keith L. Moore Preface Contributors 1. Pharmacokinetic Changes During Pregnancy and Their Clinical Relevance Ronen Loebstein, Arieh Lalkin, and Gideon Koren

iii v vii xiii

1

2. Developmental Risk Assessments Gerald F. Chernoff

23

3. Drugs in Pregnancy Gideon Koren, Anne Pastuszak, and Shinya Ito

37

4. Teratogenic Drugs and Chemicals in Humans Irena Nulman, Gordana Atanackovic, and Gideon Koren

57

5. Treatment for Epilepsy in Pregnancy Irena Nulman, Dionne Laslo, and Gideon Koren

73

6. The Safety of Commonly Used Antidepressants in Pregnancy Gideon Koren, Anne Pastuszak, Sheila Jacobson, and Irena Nulman

85

ix

x

Contents

7.

Benzodiazepine Use in Pregnancy and Major Malformations or Oral Cleft: Meta-Analysis of Cohort and Case-Control Studies Lisa R. Dolovich, Antonio Addis, J. M. Re´gis Vaillancourt, J. D. Barry Power, Gideon Koren, and Thomas R. Einarson

105

8.

Drugs and Chemicals Most Commonly Used by Pregnant Women Doreen Matsui, Monica Bologa, Frank Fassos, Michael McGuigan, Michael J. Rieder, and Gideon Koren

115

9.

Periconceptional Folate and Neural Tube Defects: Time for Rethinking Gideon Koren

137

10.

11.

12.

The Effectiveness of Preconceptional Counseling on Women’s Compliance with Folic Acid Supplementation Anne Pastuszak, Dimple Bhatia, Bunmi Okotore, and Gideon Koren Pregnancy Outcome Following Maternal Exposure to Corticosteroids: A Prospective Controlled Cohort Study and a Meta-Analysis of Epidemiological Studies Laura Park-Wyelie, Paolo Mazzotta, Myla E. Moretti, Anne Pastuszak, Lizanne Beique, Laura Hunnisett, Mark H. Friesen, Sheila Jacobson, Sonja Kasapinovic, Debra Chang, Orna Diav-Citrin, C. Laskin, K. Spitzer, Irena Nulman, Thomas R. Einarson, and Gideon Koren Use of the Retinoid Pregnancy Prevention Program in Canada: Patterns of Contraception Use in Women Treated with Isotretinoin and Etretinate Anne Pastuszak, Gideon Koren, and Michael J. Rieder

141

151

169

13.

Drugs and Breast-Feeding Anna Taddio and Shinya Ito

177

14.

Poisoning in Pregnancy Milton Tenenbein

233

15.

Carbon Monoxide Poisoning During Pregnancy Benoit Bailey

257

16.

Direct Drug Toxicity to the Fetus Orna Diav-Citrin and Gideon Koren

269

17.

The Approach to the Mother on Nonmedicinal and Chemical Use in Pregnancy Joyce F. Schneiderman

18.

Maternal and Obstetrical Effects of Prenatal Drug Exposure Raafat Bishai and Gideon Koren

321

335

Contents

19. Neonatal Drug Withdrawal Syndromes James B. Besunder and Jeffrey L. Blumer 20. Current Management of the Neonatal Abstinence Syndrome: A Critical Analysis of the Evidence Jochen G. W. Theis, Peter L. Selby, Yazemir Ikizler, and Gideon Koren 21. Fetal Effects of Cocaine: An Updated Meta-Analysis Antonio Addis, Myla E. Moretti, Fayyazuddin Ahmed Syed, Thomas R. Einarson, and Gideon Koren 22. Pregnancy Outcome and Infant Development Following Gestational Cocaine Use by Social Cocaine Users in Toronto, Canada Karen Graham, Annette Feigenbaum, Irena Nulman, Rosanna Weksberg, Thomas R. Einarson, Stan Ashby, Gideon Koren, and Susan Goldberg 23. Long-Term Neurodevelopmental Risks in Children Exposed In Utero to Cocaine: The Toronto Adoption Study Irena Nulman, Gideon Koren, Joanne Rovet, Rachel Greenbaum, Michael Loebstein, and Thomas R. Einarson 24. Biological Markers of Intrauterine Exposure to Cocaine and Cigarette Smoking Gideon Koren, Julia Klein, Rachel Forman, My-Khanh Phan, and Karen Graham 25. Fetal Alcohol Syndrome: The Central Nervous System Tragedy Irena Nulman, Jonathan Gladstone, Bonnie O’Hayon, and Gideon Koren 26. Moderate Alcohol Consumption During Pregnancy and the Incidence of Fetal Malformations: A Meta-Analysis Dimitris Polygenis, Sean Wharton, Christine Malmberg, Nagwa Sherman, Debbie Kennedy, Gideon Koren, and Thomas R. Einarson

xi

347

373

385

433

447

457

467

495

27. Occupational Exposures Known to Be Human Reproductive Toxins Yedidia Bentur, Eli Zalzstein, and Gideon Koren

507

28. The Common Occupational Exposures Encountered by Pregnant Women Yedidia Bentur and Gideon Koren

529

29. Pregnancy Outcome Following Maternal Organic Solvent Exposure: A Meta-Analysis of Epidemiological Studies Kristen I. McMartin, Merry Chu, Ernest Kopecky, Thomas R. Einarson, and Gideon Koren 30. A Proactive Approach for the Evaluation of Fetal Safety in Chemical Industries Kristen I. McMartin and Gideon Koren

547

555

xii

31.

Contents

The Use of Herbal Medicine in Pregnancy and Lactation: A Clinician’s Guide Michael Gallo, Michael J. Smith, Heather Boon, and Gideon Koren

569

32.

Ionizing and Nonionizing Radiation in Pregnancy Yedidia Bentur

603

33.

Prenatal Diagnosis in Clinical Practice David Chitayat and Kathy Hodgkinson

653

34.

Fetal Malformations Associated with Drugs and Chemicals: Visualization by Sonography Irena Nulman, Benjamin Bar-Oz, Dionne Laslo, Shawn Fried, David Chitayat, and Gideon Koren

673

35.

Maternal Disorders Leading to Increased Reproductive Risk Ruthie Geist and Gideon Koren

697

36.

Teratogen Information Services Around the World Antonio Addis, Lavinia Schu¨ler, Myla E. Moretti, Maurizio Bonati, and Gideon Koren

733

37.

Teratogen Information Services Gideon Koren, Anne Pastuszak, and Myla E. Moretti

747

38.

Motherisk: The Toronto Model for Counseling in Reproductive Toxicology Myla E. Moretti and Gideon Koren

767

The Way Women Perceive Teratogenic Risk: The Decision to Terminate Pregnancy Tommy Ho, Adrienne Einarson, and Gideon Koren

789

Evaluating the Effects of Drugs in Pregnancy: A Guide to Critical Assessment of the Literature Thomas R. Einarson

797

Bendectin/Diclectin for Morning Sickness: A Canadian Follow-Up of an American Tragedy Melanie Ornstein, Adrienne Einarson, and Gideon Koren

807

Index

815

39.

40.

41.

Contributors

Antonio Addis Regional Drug Information Center (C.R.I.F.), Laboratory for Mother and Child Health, Istituto di Ricerche Farmacologiche Mario Negri, Milan, Italy Stan Ashby The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Gordana Atanackovic The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Benoit Bailey, M.D., M.Sc., F.R.C.P.C. Divisions of Clinical Pharmacology, Toxicology and Emergency Medicine, Department of Pediatrics, Hoˆpital Ste-Justine, Universite´ de Montreal, Montreal, Quebec, Canada Benjamin Bar-Oz, M.D. Department of Neonatology, Hadassah University Hospital, Mt. Scopus, Jerusalem, Israel Lizanne Beique, B.Sc.Pharm. The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Yedidia Bentur, M.D. Israel Poison Information Center, Rambam Medical Center, Technion–Israel Institute of Technology, Haifa, Israel xiii

xiv

Contributors

James B. Besunder Metro Heath Medical Center and Case Western Reserve University School of Medicine, Cleveland, Ohio Dimple Bhatia, B.Sc. The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Raafat Bishai, M.D., M.Sc., D.C.H. The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Jeffrey L. Blumer Rainbow Babies and Children’s Hospital and Case Western Reserve University School of Medicine, Cleveland, Ohio Monica Bologa The Hospital for Sick Children and the Upjohn Company of Canada, Toronto, Ontario, Canada Maurizio Bonati Regional Drug Information Center (C.R.I.F.), Laboratory for Mother and Child Health, Istituto di Ricerche Farmacologiche Mario Negri, Milan, Italy Heather Boon, B.Sc.Phm., Ph.D. Department of Health Administration, The University of Toronto, Toronto, Ontario, Canada Debra Chang The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Gerald F. Chernoff, Ph.D. Department of Toxic Substances Control, Human and Ecological Risk Division, California Environmental Protection Agency, Sacramento, California David Chitayat The Toronto Hospital and the Hospital for Sick Children, Toronto, Ontario, Canada Merry Chu, M.Sc. Division of Clinical Pharmacology and Toxicology, The Hospital for Sick Children, Toronto, Ontario, Canada Orna Diav-Citrin, M.D. Hadassah Hospital, The Hebrew University, Jerusalem, Israel Lisa R. Dolovich The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Adrienne Einarson, R.N. The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Thomas R. Einarson, Ph.D. Faculty of Pharmacology, The University of Toronto, Toronto, Ontario, Canada

Contributors

xv

Frank Fassos The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Annette Feigenbaum The Motherisk Program, Division of Clinical Genetics, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Rachel Forman The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Shawn Fried The Motherisk Program, Division of Clinical Pharmacology and Toxicology, The Hospital for Sick Children, Toronto, Ontario, Canada Mark H. Friesen, M.Sc.Pharm. The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Michael Gallo, B.Sc. Division of Clinical Pharmacology and Toxicology, The Motherisk Program, The Hospital for Sick Children, Toronto, Ontario, Canada Ruthie Geist, M.D. The Motherisk Program, Division of Clinical Pharmacology, The Hospital for Sick Children, Toronto, Ontario, Canada Jonathan Gladstone, M.D. Department of Neurology, The University of Toronto, Toronto, Ontario, Canada Susan Goldberg Department of Psychology, The University of Toronto, Toronto, Ontario, Canada Karen Graham McMaster University, Hamilton, Ontario, Canada Rachel Greenbaum Department of Psychology and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Tommy Ho The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Kathy A. Hodgkinson The Newfoundland and Labrador Genetics Program, St. John’s, Newfoundland, Canada Laura Hunnisett The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Yazemir Ikizler The Addiction Research Foundation, Toronto, Ontario, Canada

xvi

Contributors

Shinya Ito, M.D., A.B.C.P. Division of Clinical Pharmacology, The Hospital for Sick Children, Toronto, Ontario, Canada Sheila Jacobson, M.B.B.H., F.R.C.P.C. Department of Pediatrics, The Hospital for Sick Children, Toronto, Ontario, Canada Sonja Kasapinovic The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Debbie Kennedy The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Julia Klein The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Ernest Kopecky, Ph.D. Division of Clinical Pharmacology and Toxicology, The Hospital for Sick Children, Toronto, Ontario, Canada Gideon Koren, M.D., F.R.C.P(C), F.A.C.M.T. The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Arieh Lalkin Tel Aviv University, Kefar Saba, Israel C. Laskin, M.D. The Obstetric Medicine Program, Division of Rheumatology, Toronto Hospital, and the University of Toronto, Toronto, Ontario, Canada Dionne Laslo, M.A. Division of Clinical Pharmacology and Toxicology, The Hospital for Sick Children, Toronto, Ontario, Canada Michael Loebstein The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Ronen Loebstein

Sheba Medical Center, Tel Aviv University, Tel Hashomer, Israel

Christine Malmberg The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Doreen Matsui The FRAME Program, Division of Clinical Pharamacology, The Children’s Hospital of Western Ontario, London, Ontario, Canada Paolo Mazzotta, M.Sc. The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada

Contributors

xvii

Michael McGuignan The Poison Control Center and Department of Pediatrics, The Hospital for Sick Children, Toronto, Ontario, Canada Kristen I. McMartin, M.Sc. Division of Clinical Pharmacology and Toxicology, The Hospital for Sick Children, Toronto, Ontario, Canada Myla E. Moretti, M.Sc. The Motherisk Program, The Hospital for Sick Children, Toronto, Ontario, Canada Irena Nulman, M.D. The Motherisk Program, Division of Clinical Pharmacology, The Hospital for Sick Children, Toronto, Ontario, Canada Bonnie O’Hayon, M.D. Division of Clinical Pharmacology and Toxicology, The University of Toronto, Toronto, Ontario, Canada Bunmi Okotore, M.D. The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Melanie Ornstein Faculty of Medicine, The Hospital for Sick Children, The University of Toronto, Toronto, Ontario, Canada Laura Park-Wyelie, B.Sc.Pharm. The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Anne Pastuszak, Ph.D.(C) The Fetal Centre, The Hospital for Sick Children, Toronto, Ontario, Canada My-Khanh Phan The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Dimitris Polygenis Faculty of Pharmacy, The University of Toronto, Toronto, Ontario, Canada J. D. Barry Power Faculty of Pharmacy, The University of Toronto, Toronto, Ontario, Canada Michael J. Rieder The FRAME Program, Division of Clinical Pharamacology, The Children’s Hospital of Western Ontario, London, Ontario, Canada Joanne Rovet Department of Psychology and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Joyce F. Schneiderman The Addiction Research Foundation and The University of Toronto, Toronto, Ontario, Canada

xviii

Contributors

Lavinia Schu¨ler, M.D., Ph.D. Federal University of Rio Grande do Sul, Porto Alegre, Brazil Peter L. Selby, M.D. Department of Family Medicine, St. Joseph’s Health Centre, Toronto, Ontario, Canada Nagwa Sherman The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Michael J. Smith, M.S.PharmS., N.D. cine, Toronto, Ontario, Canada

The Canadian College of Naturopathic Medi-

K. Spitzer, B.Sc. The Obstetric Medicine Program, Division of Rheumatology, The Toronto Hospital, and the University of Toronto, Toronto, Ontario, Canada Fayyazuddin Ahmed Syed The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Anna Taddio, Ph.D. Department of Pharmacy, The Hospital for Sick Children, Toronto, Ontario, Canada Milton Tenenbein University of Manitoba, Winnipeg, Manitoba, Canada Jochen G. W. Theis, M.D. Division of Clinical Pharmacology and Toxicology, The Hospital for Sick Children, Toronto, Ontario, Canada J. M. Re´gis Vaillancourt The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Sean Wharton The Motherisk Program, Division of Clinical Pharmacology and Toxicology, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Rosanna Weksberg Division of Clinical Genetics, Department of Pediatrics and the Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada Eli Zalzstein

Soroka Medical Center, Beer-Sheva, Israel

1 Pharmacokinetic Changes During Pregnancy and Their Clinical Relevance Ronen Loebstein Tel Aviv University, Tel Hashomer, Israel

Arieh Lalkin Tel Aviv University, Kefar Saba, Israel

Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

INTRODUCTION The dynamic physiological changes that occur in the maternal-placental-fetal unit during pregnancy influence the pharmacokinetic processes of drug absorption, distribution and elimination. Pregnancy-induced maternal physiological changes may affect gastrointestinal function and hence drug absorption rates. Ventilatory changes may influence the pulmonary absorption of inhaled drugs. As the glomerular filtration rate usually increases during pregnancy, renal drug elimination is generally enhanced, whereas hepatic drug metabolism may increase, decrease, or remain unchanged. A mean increase of 8 L in total body water alters drug distribution and results in decreased peak serum concentrations of many drugs. Decreased steady-state concentrations have been documented for many agents as a result of their increased clearance. Pregnancy-related hypoalbuminaemia, leading to decreased protein binding, results in increased free drug fraction. However, as more free drug is available for either hepatic biotransformation or renal excretion, the overall effect is an unaltered free drug concentration. Since the free drug concentration is responsible for drug effects, the above mentioned changes are probably of no clinical relevance. The placental and fetal capacity to metabolize drugs together with physiological factors, such as differences in the acid-base equilibrium of the mother versus the fetus, determine the fetal exposure to the drugs taken by the mother. As most drugs are excreted into the milk by passive diffusion, the drug concentration in milk is directly proportional to the corresponding concentration in maternal plasma.

Reprinted from Clinical Pharmacokinetics 1997; 33(5): 328–342.  Adis International Limited. 1

2

Loebstein et al.

The milk-to-plasma (M: P) ratio, which compares milk with maternal plasma drug concentrations, serves as an index of the extent of drug excretion in the milk. For most drugs the amount ingested by the infant will not attain therapeutic levels. Epidemiological surveys have determined that between one- to two-thirds of all pregnant women will take at least one medication during pregnancy. The most commonly used drugs are antimicrobial agents, followed by antiemetics, tranquilizers, and analgesics (1–5). These data demonstrate that drug use during pregnancy is of public concern, as the need to treat the mother must be balanced against the potential adverse effects to the fetus. As pregnancy advances, the complex physiological unit of mother, placenta, and fetus undergoes considerable dynamic physiological changes (6,7). These changes lead to variations in the processes of absorption, distribution, and elimination of drugs. Although most drug therapy in pregnancy is directed to maternal requirements, it is often carried out without considering the important modifications of drug handling by the pregnant woman. Moreover, as previous studies have determined, most drugs easily cross the placenta (8,9), making the fetus an unwanted or even nonbenefiting drug recipient. On the other hand, with the introduction of modern antenatal diagnostic techniques, the idea of treating the fetus in utero, by giving drugs to the pregnant mother, has emerged. Maternal and fetal drug responses during pregnancy are influenced by two factors: (1) changes in absorption, distribution, and elimination of the drug in the mother, which are dictated by pregnancy-induced physiological changes, and (2) the placental-fetal unit, which affects the amount of drug crossing the placenta, the fraction of drug being metabolized by the placenta, and the distribution and elimination of a drug by the fetus. These factors, together with the characteristics of the specific drug, should be taken into consideration when predicting the time course of the drug concentration in the fetomaternal unit.

PHARMACOKINETICS OF THE MATERNAL-FETAL UNIT Although several pharmacokinetic models have been proposed to describe drug disposition in the maternal-placental-fetal unit (10) generally two principal changes characterize pregnancy with respect to drug kinetics (see Fig. 1): The maternal physiological changes The effects of the placental-fetal compartment Alterations in Drug Kinetics Due To Maternal Changes Drug Absorption Gastrointestinal Absorption. The following factors affect the gastrointestinal absorption of drugs: Drug formulation Food composition Chemical composition The pH of the intestinal secretions Gastric emptying time

Pharmacokinetic Changes During Pregnancy

3

Figure 1 Drug disposition in the maternal-placental-fetal system. The factors affecting the pharmacokinetics and drug effects on mother and fetus are (1) altered maternal absorption; (2) increased maternal unbound drug fraction; (3) increased maternal plasma volume; (4) altered hepatic clearance; (5) increased maternal renal blood flow and glomerular filtration rate; (6) placental transfer; (7) possible placental metabolism; (8) placental blood flow; (9) maternal-fetal blood pH; (10) preferential fetal circulation to the heart and brain; (11) undeveloped fetal blood-brain barrier; (12) immature fetal liver enzyme activity; and (13) increased fetal unbound drug fraction. GI ⫽ gastrointestinal.

Intestinal motility Blood flow An increase in plasma progesterone level during pregnancy is believed to be responsible for the reduction in intestinal motility, resulting in a 30 to 50% increase of gastric and intestinal emptying time (6,11). Clinically, this may be important when one needs to achieve a rapid onset of drug effect. There is a reduction in gastric acid secretions (40% less than in the nonpregnant woman) together with an increase in mucous secretions, both of which lead to an increase in gastric pH and, therefore, in buffering capacity (12). Clinically, this would affect the ionization of weak acids and bases and hence their absorption. Nausea and vomiting, which occur frequently during the first trimester, may be another reason for low plasma drug concentrations. Patients should be advised to take their medication at times when nausea is minimal, usually during the evening. The design of different drug formulations takes into account the physicochemical effects of the intestinal secretions so as to allow a relatively predictable rate and site of drug release and absorption. As these factors change dynamically during pregnancy, they may alter the dissolution and absorption of the drug. Pulmonary Absorption. As cardiac output and tidal volumes are increased in pregnancy, hyperventilation and increased pulmonary blood flow occur (13,14). These alterations favor alveolar uptake and therefore should be considered when administering drugs by inhalation. The rate of induction with inhalational anesthetic agents is not necessarily faster because it depends on both pulmonary equilibrium and tissue distribution kinetics.

4

Loebstein et al.

However, dose requirements for volatile anesthetic drugs are likely to be reduced during pregnancy, as was demonstrated for halothane, isoflurane, and methoxyflurane (15). Drug Distribution The volume of distribution (Vd) of drugs may be altered during pregnancy as a result of plasma volume expansion by 50% (16). The increased cardiac output is characteristically distributed to different organs: renal blood flow increases by 50% at the end of the first trimester (14,17). The uterine blood flow reaches its peak at term (36–42 L/h) with 80% perfusing the placenta and 20% the myometrium (18). The total mean increase in total body water is 8 L, distributed 60% to the placenta, fetus and amniotic fluid and 40% to maternal tissues (16,17). As a result of this volume expansion, a decrease in the peak serum concentration (C max ) of many drugs has been documented. It is expected, therefore, that drugs distributed mainly to water compartments, and thus having a relatively small Vd, will demonstrate the greatest decrease in serum C max . Moreover, decreased steadystate drug concentrations have been documented for different agents as a result of their enhanced clearance. Drug dosage requirements would be expected to be greater to achieve the same therapeutic effect if these effects were not offset by other pharmacokinetic changes of pregnancy discussed later in this review (see below). Protein Binding As the pregnancy advances, the plasma volume expands at a greater rate than the increase in albumin production, creating physiological dilutional hypoalbuminemia (6,7,19). Moreover, steroid and placental hormones occupy protein binding sites, thus decreasing the protein binding of drugs. The overall effect is a decrease in binding capacity for albumin and, therefore, an increase in the unbound (‘‘free’’) drug (19). Since the unbound fraction of the drug is the pharmacologically active drug, pregnant women would be expected to have an increased drug effect. However, as more free drug is available for biotransformation, the net effect for most agents is an unaltered free drug concentration. In contrast to albumin, serum α 1-acid glycoprotein concentrations were found to remain the same as those in nonpregnancy. However, as there is a significant decrease in this glycoprotein within the fetus, the free fractions of basic drugs such as propranolol, lidocaine (lignocaine), and metocurine are higher in the developing child (20). There is no laboratory or clinical evidence showing increased free drug concentrations of these agents in the fetus. Yet clinicians who do not have access to measurements of unbound drug may erroneously interpret lower concentrations of the total drug as indicative of lower free drug. This issue is discussed in the context of phenytoin later in this review. Drug Elimination Hepatic Drug Elimination. The increased secretion of estrogen and progesterone in normal pregnancy affect hepatic drug metabolism in different ways: a higher rate of hepatic metabolism of certain drugs, such as phenytoin, is possibly a result of stimulated hepatic microsomal enzyme activity induced by progesterone. On the other hand, the hepatic elimination of other drugs, such as theophylline and caffeine, is reduced secondary to competitive inhibition of microsomal oxidases by progesterone and estradiol (21,22). Finally, the cholestatic effect of estrogen may interfere with the clearance of drugs, such as rifampicin (rifampin) (23), which are excreted into the biliary system. The above mentioned changes of pregnancy on hepatic physiology may alter drug metabolism, but their extent can hardly be quantified.

Pharmacokinetic Changes During Pregnancy

5

Renal Drug Elimination. The relationship between creatinine clearance and the elimination of drugs excreted mainly by the kidney has been demonstrated for many agents (24). As renal plasma flow increases by 25–50% (17) and glomerular filtration rate (GFR) by 50% (25), drugs excreted primarily unchanged in the urine—such as penicillin, digoxin, and lithium—demonstrate enhanced elimination and lower steady-state serum concentrations. However, these changes are clinically insufficient and therefore require no alteration on the dose of the above mentioned drugs. Drug Compliance In the absence of evidence-based studies, many pharmaceutical manufacturers warn the public to minimize or even avoid drug use during pregnancy. Together with the unrealistically high perception of the teratogenic risk related to drugs (26) and sometimes misinformation regarding unfounded teratogenic potential of well-tolerated medications, it is not surprising that pregnant women tend to comply less than optimally with drug therapy. The investigation of low drug serum concentrations during pregnancy must, therefore, rule out a decrease in compliance in addition to pharmacokinetic changes.

The Placental-Fetal Compartment Effect Several pharmacokinetic models have been proposed in an attempt to characterize drug behavior in the fetoplacental unit (10). Depending on the drug and available experimental data, the body is represented as a system of one compartment or more. When an administered drug readily equilibrates to achieve a fetal: maternal drug concentration of about unity, a single compartment model may be applied. On the other hand, when the fetal tissue is more slowly accessible, thus representing a deep compartment, a two-compartment model may be more appropriate. In this case the fetal :maternal concentration ratio will be lower during drug administration than during the postdistribution phase. Salicylates and diazepam are examples of such a model as they achieve higher concentrations in fetal than in the maternal plasma (27–29). Such considerations may be important in assessing fetal exposure to different agents on the basis of maternal plasma concentrations. In addition, different physiological factors such as placental blood flow and differences in plasma protein binding and/or in acid-base equilibrium of the maternal versus the fetus, also contribute to the understanding of fetal-maternal drug concentration ratios (Fig. 1). Effect of Protein Binding The fetal and newborn plasma proteins appear to bind various drugs (e.g., ampicillin and benzyl-penicillin) with less affinity than maternal plasma proteins (30). For other drugs (e.g., salicylates), a more extensive binding to the fetal than to the maternal, proteins has been documented (27). Maternal plasma albumin gradually decreases during pregnancy while fetal albumin progressively increases. This dynamic process results in different fetal :maternal albumin concentration ratios at different gestational ages. The fetal albumin concentration reaches its peak at term when it equals or even exceeds maternal albumin concentration (31) because the free (unbound) drug concentration is an important determinant of drug movement across the placenta. Since only the unbound drug is capable of crossing the placental barrier, protein-bound drugs, such as digoxin or ampicillin, reach higher total concentrations in the fetus. On the other hand, drugs with high protein binding, such as dicloxacillin, achieve higher maternal than fetal total concentrations.

6

Loebstein et al.

Acid-Base Equilibrium Effect Nonionized highly lipid-soluble molecules penetrate biological membranes more quickly than less lipid soluble ionized molecules. Therefore, the maternal and fetal pHs are important determinants of placental transfer, especially for weakly acidic or basic drugs whose pKa is close to the plasma pH. The fetal plasma pH is usually slightly more acidic than the maternal. Consequently, weak bases will be mainly nonionized and, therefore, able to easily penetrate the placental barrier. However, after crossing the placenta and making contact with the relatively acidic fetal blood, these molecules will become more ionized. This results in an apparent fall in fetal concentrations of the drug and leads to a concentration gradient; therefore a net movement from the maternal to the fetal system (32). This phenomenon is commonly referred to as ion trapping. The magnitude of this ion trapping was demonstrated in a study in which sodium salicylate 20 mg/kg body-weight was intravenously administered to 115 pregnant women at the beginning of labor. Serial blood samples from the mothers and umbilical cords of the newborns were obtained. There was a slow transfer of salicylate from the mothers to the fetuses and an increase in the neonatal :maternal serum concentration ratio to an average of 1.5 after 4 hours (120 mg/L in newborns and 80 mg/L in the mothers) (28). Thus, at least at term, total plasma concentrations of salicylate in the fetus are higher—an important consideration when assessing the magnitude of fetal exposure to salicylate on the basis of the drug concentration of the mother. Fetoplacental Drug Elimination There is some evidence that both the human placenta and the fetus are capable of metabolizing drugs (33–35). All enzymatic processes including phase I (oxidation, dehydrogenation, reduction, hydrolysis, etc.) and phase II (glucuronidation, methylation and acetylation) have been documented in the fetal liver as early as 7–8 weeks postconception (33). However, as most enzymatic processes are immature at that stage, their degree of activity is very low; therefore their contribution to the overall drug-elimination capacity is marginal. On the other hand, the reduced elimination capacity may cause more pronounced and prolonged drug effects in the fetus. The fact that about half of the fetal circulation (umbilical vein) directly reaches the heart and brain, bypassing the liver, may also contribute to this effect. Elimination of the drug from the fetus is primarily by its diffusion back to the maternal compartment. However, as most metabolites are more polar than their parent compounds, they are less likely to cross the placental barrier. This may result in metabolite accumulation in various fetal tissues. As pregnancy progresses, higher amounts of different drugs are excreted into the amniotic fluid, reflecting maturation of the fetal kidney. Rate of Maternal-Fetal Drug Equilibration As the major route of placental drug transfer is by simple diffusion, the net transplacental movement of most xenobiotics is proportional to the concentration gradient of the drug across the placenta and the physicochemical characteristics of the xenobiotic and the placental membrane. As with diffusion across other biological membranes, lipophilic drugs generally cross the placenta more readily than nonlipophilic compounds. Drugs which are nonionized at the physiological pH will diffuse across the placenta more rapidly than more basic or acidic compounds. This distinction is not absolute since some drugs which are ionised at physiological pH, such as valproic acid (valproate sodium) or salicylate, rapidly and efficiently reach the fetus (28).

Pharmacokinetic Changes During Pregnancy

7

For most nonionized lipophilic compounds, the rate-limiting step that determines placental drug transfer is blood flow. Changes in placental blood flow that may be secondary to pathophysiological conditions (e.g. pregnancy-induced hypertension, abruptio placenta) or pharmacological effects (oxytotic drugs, nicotine) can, therefore, affect the rate of placental drug transfer. On the other hand, for more polar drugs with lower rates of placental transfer, the rate-limiting step for placental transfer is diffusion. The rate of maternal-fetal drug equilibration is an important determinant in cases when therapeutic fetal drug concentrations must be achieved as quickly as possible. Clinical examples of such cases are maternal administration of antiarrhythmic agents or antibacterials in order to treat intrauterine fetal arrhythmias or infections. We studied the transplacental pharmacokinetics of digoxin using an in vitro human placental perfusion model (36). After 30 minutes of perfusion with digoxin, the amount of digoxin leaving the maternal circulation and the amount of digoxin appearing in the fetal circulation were constant at a fetomaternal ratio of 0.36 ⫾ 0.04. Based on these data, the time to reach equivalent concentrations on both sides of the placenta, assuming the constant rate of digoxin transfer is maintained, is estimated to be 268 ⫾ 34 minutes. At the end of a 3-hour perfusion study, digoxin concentrations were 3.32 ⫾ 0.43 µg/L in the maternal circulation versus 1.3 ⫾ 0.45 µg/L in the fetal circulation. The transplacental pharmacokinetics of cocaine were studied in pregnant rhesus monkeys (37). Animals were intramuscularly given cocaine hydrochloride 1 mg/kg. The C max of cocaine was reached in the maternal blood within 10 to 20 minutes after administration and cocaine was detected in fetal blood within 5 minutes, reaching C max within 30 to 120 minutes.

PREGNANCY-INDUCED PHARMACOKINETIC CHANGES OF SPECIFIC DRUGS Anticonvulsants Epilepsy is a common neurological disorder, with a prevalence of 1%. In most women with epilepsy, seizures are well controlled during pregnancy: however, if the frequency of the seizures changes, it is usually for the worse (38). The potential problem in the management of these patients arises from altered pharmacokinetics associated with pregnancy. In general, the plasma C max of anticonvulsant drugs falls during pregnancy as a result of a 50% expansion in plasma volume, while steady-state concentrations fall secondary to decrease in protein binding (16) and increased clearance. As most anticonvulsant drugs are acidic or neutral, they are highly bound to serum albumin. During pregnancy, albumin levels fall, with a corresponding fall in the bound drug. The decrease in plasma protein binding leads to more free drug available for biotransformation. Thus, the unbound concentrations (free drug) remain relatively constant, but total plasma concentrations (unbound plus protein-bound) fall. It should be remembered that the constant or even higher free drug concentration may provide better antiepileptic control, as it is the free drug that reaches the brain. For practical reasons, most laboratories measure total plasma concentration (bound plus unbound) rather than the unbound concentration, which is the pharmacologically active component. Phenytoin and valproic acid are both highly protein-bound (approximately 90%) to plasma albumin. Monitoring the total plasma concentrations of these drugs

8

Loebstein et al.

can, therefore, be misleading. Only therapeutic drug monitoring that will measure both protein-bound and unbound drug concentrations can be appropriately interpreted. The increase in GFR (25) and renal plasma flow (17) may theoretically enhance the clearance of renally excreted drugs, such as gabapentin and vigabatrin. However, presently there are no studies available regarding the pharmacokinetics of these agents during human pregnancy. Carbamazepine is regarded by many to be the drug of choice for most forms of epilepsy during pregnancy because of its relatively low teratogenic risk (39). Phenytoin has been associated with the fetal hydantoin syndrome and has neurotoxic potential (40– 42). Both valproic acid and carbamazepine have been shown to cause neural tube defects, and in women undergoing carbamazepine therapy a full workup should be performed to rule out this malformation (43,44). Phenobarbital has been much less commonly prescribed in recent years because of its tendency to produce sedation and impair cognitive function. Altered Pharmacokinetics of Commonly Used Antiepileptic Drugs During Pregnancy Carbamazepine. Carbamazepine has a relatively slow absorption, with 70–80% protein binding to albumin. The main eliminating route is by hepatic metabolism, and there may be a decrease in serum concentrations during the first months of therapy as a consequence of autoinduction of metabolism. It is important for the clinician to recognise that the dosage intervals and sample time are critical in interpreting serum concentrations. Large peak-trough fluctuations can be minimized by using a controlled-release formulation (45). As levels tend to be lower in pregnancy, and bioavailability may be lower than with conventional carbamazepine, higher dosages may be required when using controlledrelease medication (46). The concentration of the primary metabolite (carbamazepine-10, 11-epoxide) was reported to increase during pregnancy, possibly as a result of the increased carbamazepine metabolism and impaired conversion of carbamazepine-10, 11-epoxide to carbamazepine10, 11 trans-diol during pregnancy. This increase may be of importance as the metabolite (10, 11 epoxide) is thought to have comparable pharmacological activity to the parent drug (47). Phenytoin. Phenytoin follows nonlinear pharmacokinetics and has a narrow therapeutic window (48). It is highly bound to protein (90–93%) (49,50) and mainly cleared by saturable hepatic metabolism (Table 1). A significant increase in 8-hydroxylation during pregnancy may be responsible, at least partially, for increased phenytoin clearance during pregnancy and consequently for decreased serum concentrations (58). Generally, a fall in total serum phenytoin concentrations greater than 25% or worsening of seizure control should encourage physicians to increase the dose. The fall in total concentrations by itself may not indicate a fall in free drug concentrations. The decrease in the protein binding of phenytoin is probably a major reason for decreasing total drug concentrations in pregnancy, however, as it is the free drug that is available for the enhanced metabolism there should not be a major effect on free drug concentrations. Dose adjustments should ideally be based on free drug concentration monitoring especially for highly protein-bound drugs, such as phenytoin, which are most likely to be affected by pregnancy.

Pregnancy-Induced Pharmacokinetic Changes for Selected Drugs t 1/2β (min)

Drug Ampicillin Cefuroxime Imipenem Piperacillin Azlocillin Nifedipine Labetolol Sotalol Phenytoin

Pregnant

Nonpregnant

⫾ ⫾ ⫾ ⫾

3.9 5 8 10

⫾ ⫾ ⫾ ⫾

18 16 36 314

69.6 ⫾ 6.1 58 ⫾ 8 41 ⫾ 16 53.7 ⫾ 4.6 72 360 160 ⫽ 480 558 ⫾ 42

52.4 44 36 46.5 65.4 81 102 396 900

Vd(L) Pregnant 32.8 17.8 47.1 67.6 15.4

⫾ ⫾ ⫾ ⫾

2.5 1.9 14.8 11.8

106.4 ⫾ 8.1

CL(mL/min)

Nonpregnant 34.5 16.3 18.9 41.9 24.7

⫾ ⫾ ⫾ ⫾

2.7 2.1 5.8 6.2

87.3 ⫾ 7.2

Pregnant 450 ⫾ 282 ⫾ 973 ⫾ 1538 ⫾ 126.1 266 ⫾ 1704 ⫾ 196 ⫾

31 34 47 362 105 531 24

Protein binding (%)

Nonpregnant 370 ⫾ 198 ⫾ 338 ⫾ 540 ⫾ 195.7 27 ⫾ 1430 109 ⫾

Pregnant

Nonpregnant

Reference

90

51 52 53 54 54 55 56 57 50

30 27 85 75 5 7 86

Pharmacokinetic Changes During Pregnancy

Table 1

Abbreviations: CL ⫽ clearance; t 1/2β ⫽ elimination half-life; Vd ⫽ volume of distribution.

9

10

Loebstein et al.

Valproic Acid. Valproic acid is rapidly absorbed and highly protein bound to plasma albumin (88–92%) (45). The interpretation of its pharmacokinetics is limited by large fluctuations in the concentration-time profile, wide therapeutic index and concentration dependent protein binding (59,60). Analysis of unbound valproic acid is not routine and similarly there is no established therapeutic range. Dose adjustments during pregnancy are best made by clinical observation in combination with changes in concentration measurements. New Antiepileptic Drugs. Vigabatrin, lamotrigine, and gabapentin are available for the treatment of epilepsy, but there is little information regarding their pharmacokinetics and safety during human pregnancy. Antibacterials β-Lactams β-Lactam antibacterials are the oldest class of antibacterials used in the treatment of infections during pregnancy. Benzylpenicillin (penicillin G) is still the most common antibacterial used during pregnancy. Other β-lactam agents are structurally related to penicillin and are likely to be used more often than new compounds because of their proven safety profile. The therapeutic effect of β-lactam antibacterials depends primarily on maintaining adequate plasma concentrations during the entire treatment period. Their activity correlates well with pharmacokinetic characteristics [C max, area under the concentration-time curve (AUC), etc.] and time during which plasma concentration exceeds the minimal inhibitory concentration (MIC) for susceptible bacteria (Table 1) (61). The dosage regimen for βlactam antibacterials should be calculated according to the pharmacokinetic profile of the drug and its MIC values (62). Most bacterial infections are confined to the extravascular compartment. β-Lactams enter the extracellular fluid by passive diffusion (63), and thus the tissue concentrations of hydrophilic β-lactams will be low, despite excellent penetration properties into this compartment. Because of that, underestimation of β-lactam antibacterial concentrations can occur when measuring tissue concentrations at the infection site (64,65). There is good evidence that direct correlation exists between β-lactam plasma concentrations and response to infections at an extravascular site, and thus plasma concentration is a predictor for response. β-Lactams are acidic agents. Their binding to albumin ranges from 2.5–97% for different drugs (61). The total AUC values in extravascular sites and plasma are related to the concentration of the non–protein bound concentrations in the plasma. The free drug concentration in the plasma determines the penetration of drugs into tissue (66). This is usually of great importance only when the degree of binding exceeds 80% (67). Plasma protein binding can be dependent on plasma concentration, in which higher levels reduce the extent of protein binding (68). It is important to remember that β-lactams are water soluble and cross the placenta by passive diffusion along the concentration gradient. As this is a slow process, it takes several hours for fetal and maternal concentrations to reach an equilibrium. In the fetal compartment, water soluble drugs circulate in the fetus and are excreted into the amniotic fluid, recycling again in the fetus by gastrointestinal absorption. Thus, water soluble drugs are eliminated from the amniotic fluid slowly and this fluid can be regarded as a reservoir for hydrophilic drugs (69). For more lipophilic drugs, metabolism is an important means

Pharmacokinetic Changes During Pregnancy

11

of elimination. Both the fetal liver and placenta can metabolize lipophilic drugs to some degree (70,71). The common denominator to most studies conducted in pregnancy is the substantially lower serum concentrations, which in many cases should lead to the consideration of higher doses than those given to a nonpregnant woman. Ampicillin The pharmacokinetic of ampicillin in pregnant women has been studied extensively, and alterations in its pharmacokinetics have been found when compared with data from nonpregnant women (72). These altered pharmacokinetics lead to a lower plasma concentration of the drug in pregnant women than in nonpregnant women (61). In a recent study (51), 22 women were given a single dose of ampicillin precaesarean and then again 6 weeks postcaesarean, thereby using the patients as their own control. Pregnancy significantly increased the elimination rate constant, decreased the AUC by 20%, and increased the total body clearance (51). Another study has shown that maternal serum concentrations decline relatively rapidly after intravenous infusion of ampicillin and recovered in fetal serum 30 minutes after maternal infusion, reaching-equilibrium with maternal serum in 1 hour. The concentration of ampicillin in amniotic fluid continued to rise for up to 8 hours, then slowly declined over the next 19 hours (30). The elimination half-life (t 1/2β ) was reported to be 30 minutes during caesarean section (73), 40 and 58 minutes during delivery (74,75), and 40 minutes and 1.6 hours during the third trimester (Table 2) (75,76). Phenoxymethylpenicillin (Penicillin V) Phenoxymethylpenicillin (penicillin V) is the most commonly used antibacterial during pregnancy. Altered pharmacokinetics has been demonstrated in a study comparing 12 pregnant women during the second and third trimesters with 6 nonpregnant women. This study showed significantly faster elimination rates of the drug from the plasma of pregnant women compared with nonpregnant women. The AUC was lower in pregnant women (441 IU/mL/min) compared with nonpregnant women (887 IU/mL/min) as was C max (3.7 IU/mL in pregnant versus 5.6 µ/mL in nonpregnant women) (82). Ureidopenicillins Azlocillin and piperacillin have been demonstrated to cross the human placenta (83,84). In a study of eight women undergoing caesarean section and six nonpregnant women undergoing gynecological operations, the mean C max values after administration of piperacillin 4 g were significantly lower in the pregnant women than in the nonpregnant women (87.5 versus 172.2 mg/L). The total clearance was faster in pregnant women than in nonTable 2 Fetomaternal Concentration Ratio of Selected Antibacterials Drug Ampicillin Cephalosporins Dicloxacillin Gentamycin Methicillin Benzylpenicillin (penicillin G)

Fetomaternal ratio

References

0.38: 0.87 0.13: 1.0 0.07: 0.27 0.21: 1.0 0.83: 1.43 0.06: 0.7

77 78 79 80,81 79 82

12

Loebstein et al.

pregnant women (92.28 versus 32.4 L/h), and volume of distribution (Vd) was significantly greater (67.6 versus 41.2 L) (84). In another study, piperacillin 3 g every 4 hours and mezlocillin 4 g every 6 hours were evaluated in 6 postpartum patients. The serum half-life and clearance rates for mezlocillin were 82.4 minutes and 12.12 ⫾ 6.3 L/h, respectively, and for piperacillin 32.9 minutes and 27.36 ⫾ 5.28 L/h, respectively (85). Imipenem Imipenem is a broad-spectrum β-lactam used in combination with cilastatin, a competitive inhibitor of dehydropeptidase, in order to minimise the renal metabolism of the drug (86). A single dose of imipenem was administered to 14 pregnant women in early and late pregnancy and 6 nonpregnant women, the mean plasma C max were 14.7 ⫾ 4.9, 14.9 ⫾ 5.2, and 43 ⫾ 28.3 mg/L in early pregnancy, late pregnancy, and nonpregnant women, respectively. The Vd was significantly larger during early and late pregnancy than in nonpregnant women, and clearance from plasma was significantly faster than in nonpregnant women (53). Cephalosporins First-Generation Cephalosporins. The mean serum concentration of cefalexin (87), cefalothin (88), and cefazolin (89), are all considerably lower in pregnant women when compared with nonpregnant women (Table 2). Second-Generation Cephalosporins. Cefuroxime plasma concentrations in pregnant women at the time of labour and delivery were lower than in nonpregnant women; the half-life was significantly shorter, and clearance and recovery in the urine were significantly higher. The drug effectively passes the placenta and is detected in cord plasma soon after drug administration to the mother (52). Third-Generation Cephalosporins. The penetration of ceftriaxone into the fetus at parturition was studied in 17 pregnant women. After intravenous administration of ceftriaxone 2 g, substantial concentrations were achieved in the umbilical cord blood, amniotic fluid and placenta after 1 hour. The t 1/2 β was approximately 6 hours, identical to that of the mother (90). Bourget et al. (91) studied the pharmacokinetics and protein binding of ceftriaxone during pregnancy. It was found that the pharmacological behavior of ceftriaxone in pregnant women during the third trimester is comparable with that in healthy volunteers. Concentrations of total and free drug measured at 24 hours were greater than the MIC for allegedly susceptible organisms, both on day 1 and at steady state. No dosage schedule required particular adjustment, i.e., neither a loading dose nor maintenance dose increased. Lithium Lower serum concentrations of lithium have been reported during pregnancy. Since this antidepressant agent is eliminated almost entirely by the kidney, the increase in glomerular filtration rate (GFR) during pregnancy is consistent with the observed decrease in serum concentrations. Because of lithium’s narrow therapeutic range, a drop in its serum concentrations may result in suboptimal therapy (92). Some lithium is reabsorbed in the proximal renal tubule competitively by the sodium transporter. Pregnant women with low circulating effective volume secondary due either to dehydration (hyperemesis gravidarum, vomiting) or restricted sodium intake (toxemia), may experience higher lithium concentrations as a

Pharmacokinetic Changes During Pregnancy

13

result of enhanced renal absorption. Finally, during puerperium, with the decrease in GFR to prepregnancy serum concentrations, values are expected to rise, potentially to toxic levels (32). Digoxin As with lithium, digoxin is eliminated mainly by renal excretion. When measured at term, as expected, the digoxin serum concentrations which were found in 5 pregnant women was twofold lower compared with concentrations 1 month later (32,93). Therefore, it is reasonable to expect maternal dose requirements to increase during pregnancy and to fall in the puerperium. Antihypertensive Agents With more reassuring studies regarding the tolerability of calcium channel blockers and β adrenergic blockers during pregnancy (94–96), these agents are expected to become more commonly used to control hypertension during pregnancy. The pharmacokinetics of nifedipine were studied during the immediate postpartum period in patients with preeclampsia (Table 1) (55). The terminal t 1/2 β was found to be shorter than that reported for normotensive volunteers or nonpregnant women with hypertension (1.35 ⫾ 0.3 versus 3.4 ⫾ 0.4 hours). A mean apparent oral elimination clearance of 3.3 ⫾ 1.3 L/h/kg was more rapid than that found in normal volunteers, 0.49 ⫾ 0.09 L/h/kg. Based on these data a dosage interval of every 3–4 hours is suggested when rapid release nifedipine is used in a postpartum patient with preeclampsia. Pharmacokinetic studies of oral labetalol in women with pregnancy-induced hypertension in the third trimester have shown the terminal t 1/2 β to be significantly shorter than that reported for normotensive volunteers or nonpregnant patients with hypertension (1.7 ⫾ 0.27 hours versus 6 to 8 hours) suggesting the need for a shorter dosage interval (56). Pethidine (Meperidine) Pethidine (meperidine) is widely used as an analgesic agent during labor. Its disposition was studied in 11 pregnant women after administration of a single 50 mg intravenous dose through 48 hours postinjection (Table 1) (97). Pethidine clearance was significantly higher in pregnant women (0.678 ⫾ 0.1 L/min) compared with previously reported data collected from nonpregnant women (1.06 L/min). The half-lives of the rapid (α) and terminal (β) elimination phases were significantly greater in pethidine-treated pregnant women (2.3 and 13.3 hours, respectively) compared with nonpregnant women (0.1 and 3.6 hours, respectively). The longer β-elimination half-life may have important clinical implications in the sequential dose administration of this agent. DRUG USE DURING LACTATION Breast-feeding patterns have changed considerably over the past several decades. In the United States the proportion of women breast feeding their babies was estimated at 60% in 1984 compared with 30% in the 1950s (98). In the early 1980s, women in Canada also demonstrated an increasing trend toward breast-feeding compared with the 1960s (99). Factors that affect maternal decisions to breast-feed include:

14

Loebstein et al.

Parity Educational level Social class Psychosocial factors Socioeconomic class The use of medications is believed to contribute to a mother’s choice of whether to feed by breast or by bottle (100). Breast milk is produced and secreted in the alveolar cells, from which it is expelled by contractile myoepithelial cells into the duct system. Drugs enter breast milk by either passive diffusion or active transport (101). Since most drugs are excreted into milk by passive diffusion, the drug concentration in milk is directly proportional to the corresponding concentration in maternal plasma (Table 3) (113,114). As breast milk is slightly more acidic than the plasma, drugs that typically pass into milk are weak bases, lipid-soluble, and poorly bound to proteins (115,116). The milk :plasma (M:P) ratio, which compares milk with maternal plasma drug concentrations, serves as an index of the extent of drug excretion in the milk. M: P ratios for selected drugs are given in Table 3. However, when dealing with infant exposure to drugs in breast milk, in addition to the M:P ratio, the infant’s clearance of the drug also has to be taken into account (117). Some general principles can be used by the clinician when managing cases with drug exposure in a breast-fed infant: The drug concentration in breast milk usually does not exceed the maternal plasma concentration. Even when the M:P ratio for a given drug approaches or exceeds 1.0, the amount of drug ingested by the infant is rarely sufficient to attain the therapeutic concentrations. A short exposure to a drug is usually of less concern than a drug given for a long period of time. The amount of drug ingested by the infant can be minimized by feeding the infant just before or at the time of maternal administration. In cases of long-term drug therapy, the effect on the infant of drug exposure while breast-feeding is to lower concentrations compared with those shown in utero. Table 3 Drug Excretion in Human Breast Milk Drug Amoxicillin Atenolol Carbamazepine Cefotaxime Diazepam Digoxin Lithium carbonate Penicillin Phenytoin Propranolol Theophylline Warfarin

M :P ratio

References

0.014–0.043 1.5–6.8 0.6–0.7 0.029–0.16 0.08–0.13 0.6–0.8 0.25–0.77 0.02–0.2 0.12–0.24 0.5 0.57 ⬍0.01

102 103,104 105 102 106 107 108 109 110 103 111 112

Abbreviation: M:P ⫽ milk to plasma.

Pharmacokinetic Changes During Pregnancy

15

The American Academy of Pediatrics and the Canadian Paediatric Society list the following drugs as contraindicated during breast feeding: bromocriptine, cyclophosphamide, cyclosporin, doxorubicin, ergotamine, lithium, and methotrexate. Amiodarone is not recommended while breast feeding as it can cause iodine-induced fetal hypo- or hyperthyroidism (118). Bromocriptine and ergotamine are contraindicated because of their prolactin-suppressing activity (119,120). All antineoplastic agents are ill advised because of their theoretical ability to induce short-term toxicity (myelosuppression) and long-term carcinogenicity (121–124). Lithium freely passes into human breast milk and significant plasma concentrations have been detected in nursing infants (108). In addition, temporary cessation of breast-feeding is advised for some radioactive compounds. Antithyroid Drugs After radioactive iodine, methimazole and propylthiouracil (PTU) are the most commonly used antithyroid medications (125). Since the amount of PTU the infant would ingest is estimated to be less than 1% of the mother’s dose on a body-weight basis (126) compared with 2–12% with methimazole (127), PTU is considered the drug of choice for lactating women with hyperthyroidism. Despite the low-level exposure, there is still a potential risk of thyroid suppression in the infant; therefore clinical and thyroid-stimulating hormone (TSH) monitoring is recommended. No adverse effect in breast-fed infants has been reported so far (128). Anticonvulsants Carbamazepine, phenytoin, and valproic acid are considered compatible with breastfeeding since the exposure levels of the infant to these drugs is less than 10% of the expected exposure if the drug was given directly to the infant (105,110,129). In contrast, phenobarbital, ethosuximide and primidone result in an infant’s exposure levels of 100, 50 and greater than 10% of the mother’s dose on a body-weight basis (130). These high-level exposures are mainly secondary to the low clearance of phenobarbital and ethosuximide in infants. Therefore, clinical and drug concentration monitoring in breast milk and in the plasma of the infant is recommended. ␤-Adrenergic Blockers Of concern among all β-adrenergic blockers are atenolol and sotalol, which cause relatively high exposure levels: 25 and 20%, respectively (103,131,132); mainly due to their low protein binding. This is clinically important in the early neonatal period when GFR is a low secondary to immature renal function (104). Alcohol The M: P ratio of alcohol (ethanol) is about 1.0 (133,134). Together with the premature metabolising capacity of alcohol in the neonatal and infantile periods, heavy maternal alcohol consumption is incompatible with breast feeding. Motor development was claimed to be slower in infants breast-fed by alcohol-drinking mothers (135); however, this is unlikely unless the mother is drinking alcohol on an almost continuous basis. Other factors such as in utero exposure to alcohol, and therefore fetal alcohol effects in these children, need to be considered in the light of the fact that the infants concentrations of alcohol are probably too low to explain the developmental delay.

16

Loebstein et al.

CONCLUSIONS Pregnancy is associated with a plethora of physiological changes that may affect the way the body handles drugs. However, for most agents, the net effect of these changes results in unaltered free drug concentrations and, therefore, unchanged drug effects. Dose requirements are likely to be higher for agents with an enhanced clearance. For more highly protein-bound drugs, pregnancy-related hypoalbuminemia may result in higher free drug concentrations. Dose adjustments for such drugs should ideally be based on free drug concentration monitoring. Although most drugs are excreted into breast milk by passive diffusion, the amount ingested by the infant is rarely sufficient to attain therapeutic concentrations.

REFERENCES 1. Heinonen OP, Slone D, Monson RR, et al. Cardiovascular birth defects and antenatal exposure to female sex hormones. N Engl J Med 1977; 296(2):60–67. 2. Forfal JO, Nelson MM. Epidemiology of drugs taken by pregnant women: drugs that may affect the fetus adversely. Clin Pharmacol Ther 1973; 14:632–642. 3. Brocklebank JC, Ray WA, Federspiel CF, et al. Drug prescribing during pregnancy: a controlled study of Tenessee Medicaid recipients. Am J Obstet Gynecol 1978; 132(3):235–244. 4. Cordero JF, Oakley GP. Drug exposure during pregnancy: some epidemiologic considerations. Clin Obstet Gynecol 1983; 26(2):418–428. 5. Piper JM, Baum C, Kennedy DL. Prescription drug use before and during pregnancy in a Medicaid population. Am J Obstet Gynecol 1987; 157(1):148–156. 6. Hytten FE, Leitch T. The physiology of pregnancy. Oxford: Blackwell, 1971. 7. Quilligan EJ, Kaiser IH. Maternal physiology. In: Danforth DN, ed. Danforth Obstetrics and Gynecology. 3rd ed. New York: Harper & Row, 1982:326–341. 8. Young IM. The placenta: blood flow and transfer. In: Downman CBB, ed. Modern Trends in Physiology. London: Butterworth, 1972:214–244. 9. Mirkin BL, Singh S. Placental transfer of pharmacologically active molecules. In: Mirkin BL, ed. Perinatal Pharmacology and Therapeutics. New York: Academic Press, 1976:1–69. 10. Krauer B, Krauer F. Drug kinetics in pregnancy. In: Gibaldi M, Prescott L, eds. Handbook of Clinical Pharmacokinetics. New York: Adis, 1983:1–17. 11. Parry E, Shields R, Turnbull A. Transit time in the small intestine in pregnancy. J Obstet Gynaecol Br Commonw 1970; 77:900–901. 12. Hume Jr RF, Killam AP. Maternal physiology. In: Danforth DN, ed. Obstetric and Gynecology, 6th ed. Philadelphia: J.B. Lippincot Company, 1990:93–100. 13. Cugell DW, Frank NR, Gaensler EA, et al. Pulmonary function in pregnancy. Am Rev Tuberc 1953; 67:568–597. 14. Kerr MG. Cardiovascular dynamics in pregnancy and labour. Br Med Bull 1968; 24:19–24. 15. Palahniuk RJ, Shmider SM, Eger EI. Pregnancy decreases the requirement for inhaled anesthetic agents. Anesthesiology 1974; 41:82–83. 16. Pizani BBK, Campbell DM, McGillivray T. Plasma volume in normal pregnancy. J Obstet Gynaecol 1973; 80:884–887. 17. Dunihoo DR. Maternal physiology. In: Dunihoo DR, ed. Fundamentals of Gynecology and Obstetrics. Philadelphia: J.B. Lippincott Co., 1992:280–284. 18. Assali NS, Rauramot L, Peltonen T. Measurement of uterine blood flow and uterine metabolism. Am J Obstet Gynecol 1960; 79:86–98. 19. Dean M, Stock B, Patterson RJ, et al. Serum protein binding of drugs during and after pregnancy in humans. Clin Pharmacol Ther 1980; 28:253–260.

Pharmacokinetic Changes During Pregnancy

17

20. Wood M, Alastair JJ, Wood MB. Changes in plasma drug binding and α 1-acid glycoprotein in mother and newborn infant. Clin Pharmacol Ther 1981; 29:522–526. 21. Davis M, Simmons CJ, Dordini B, et al. Induction of hepatic enzymes during normal pregnancy. J Obstet Gynaecol Br Commonw 1973; 80:690–694. 22. Juchau MR, Mirkin DL, Zachariah PK. Interactions of various 19-nor steroids with human placental microsomal cytochrome P-450. Chem Biol Interact 1976; 15(4):337–347. 23. Harrison LI, Gibaldi M. Influence of cholestasis on drug elimination: pharmacokinetics. J Pharm Sci 1976; 65:1246–1348. 24. Dettli L. Elimination kinetics and drug dosage in renal insufficiency patients. Triangle 1975; 14(3–4):117–123. 25. Davidson JM, Hytten FE. Glomerular filtration during and after pregnancy. J Obstet Gynaecol 1974; 81:588–595. 26. Koren G, Bologa M, Pastuszak A. The way women perceive a teratogenic risk: the decision to terminate pregnancy. In: Koren G, ed. Maternal-Fetal Toxicology. 2nd ed. New York: Marcel Dekker Inc., 1994:727–735. 27. Levy G, Procknal JA, Garrettson LK. Distribution of salycilate between neonatal and maternal serum at diffusion equilibrium. Clin Pharmacol Ther 1975; 18(2):210–214. 28. Levy G. Salicylate pharmacokinetics in the human neonate. In: Morselli P, Garattini C, Sereni Y, eds. Basic and Therapeutic Aspects of Perinatal Pharmacology. New York: Raven Press, 1975:319–330. 29. Mandelli M, Morselli PL, Nordio S, et al. Placental transfer of diazepam and its disposition in the newborn. Clin Pharmacol Ther 1975; 17(5):564–572. 30. Bray RE, Boe RW, Johnson WL. Transfer of ampicillin into fetus and amniotic fluid from maternal plasma in late pregnancy. Am J Obstet Gynecol 1966; 96:938–942. 31. Dancis J, Braverman N, Lind J. Plasma protein synthesis in the human fetus and placenta. J Clin Invest 1957; 36:398–404. 32. Koren G. Changes in drug disposition in pregnancy and their clinical implications In: Koren G, ed. Maternal-Fetal Toxicology. 2nd ed. New York: Marcel Dekker Inc., 1994:1–13. 33. Juchau MR, Chao ST, Omiecinski CJ. Drug metabolism by the human fetus. In: Gibaldi M, Prescott L. ed. Handbook of Clinical Pharmacokinetics. New York: Adis, 1983:58–78. 34. Pelkonen O. Drug metabolism in the human fetal liver: relationship to fetal age. Arch Int Pharmacodyn Ther 1973; 202(2):281–287. 35. Pelkonen O, Karki NT. Demonstration of cytochrome P-450 in human foetal liver microsomes in early pregnancy. Acta Pharmacol Toxicol (Copenh) 1971; 30(1):158–160. 36. Derewlany LO, Leeder JS, Kumar R, et al. The transport of digoxin across the perfused human placental lobule. J Pharmacol Exp Ther 1991; 256(3):1107–1111. 37. Binienda Z, Bailey JR, Duhart HM, et al. Transplacental pharmacokinetics and maternal/ fetal plasma concentrations of cocaine in pregnant macaques near term. Drug Metab Dispos 1993; 21(2):364–368. 38. Schmidt D. The effect of pregnancy on the natural history of epilepsy. In: Janz D, Bossi L, Dam M, ed. Epilepsy, Pregnancy and the Child. New York: Raven Press, 1981:3–14. 39. Niebyl JR, Blake DA, Freeman JM, et al. Carbamazepine levels in pregnancy and lactation. Obstet Gynecol 1970; 53:139–140. 40. Brodie MJ. Established anticonvulsants and treatment of refractory epilepsy. Lancet 1990; 336:350–354. 41. Hanson JW, Smith DW. The fetal hydantoin syndrome. J Pediatr 1975; 89:662–668. 42. Howe AM, Lipson AH, Sheffield LJ, et al. Prenatal exposure to phenytoin: facial development and a possible role for vitamin K. Am J Med Genet 1995; 58(3):238–244. 43. Robert E, Guibaud P. Maternal valproic acid and congenital neural tube defects [letter]. Lancet 1982; 2:937. 44. Valproate: a new cause of birth defects: report from Italy and follow-up from France. MMWR 1983; 32:438–439.

18

Loebstein et al.

45. Thomson AH, Brodie MJ. Pharmacokinetics optimisation of anticonvulsant therapy. Clin Pharmacokinet 1992; 23(3):216–230. 46. Yerby MS, Friel PN, Miller DQ. Carbamazepine binding and disposition in pregnancy. Ther Drug Monit 1985; 7:269–273. 47. Thomson T, Almkvist O, Nilsson BY, et al. Carbamazepine-10,11-epoxide in epilepsy: a pilot study. Arch Neurol 1990; 47:888–892. 48. Armijo JA, Cavanda E. Graphic estimation of phenytoin dose in adults and children. Ther Drug Monit 1991; 13:507–510. 49. Brodie MJ. Management of epilepsy during pregnancy and lactation. Lancet 1990; 336:426– 427. 50. Perucca E, Richens A, Ruprah M. Serum protein binding of phenytoin in pregnant women. Proc Br Pharmacol Soc 1981; 11:409P–410P. 51. Chamberlain A, White S, Bawdon R, et al Pharmacokinetics of ampicillin and salmactam in pregnancy. Am J Obstet Gynecol 1993; 168:667–673. 52. Philipson A, Stiernstede G. Pharmacokinetics of cefuroxime in pregnancy. Am J Obstet Gynecol 1982; 142:823–828. 53. Heikkila A, Renkonen OV, Erkkola R. Pharmacokinetics and placental passage of imipenem during pregnancy. Antimicrob Agents Chemother 1992; 36:2652–2655. 54. Voigt R, Schroder S, Peiker G. Pharmacokinetic studies of azlocillin and piperacillin during late pregnancy. Chemotherapy 1985; 31:417–424. 55. Barton JR, Prevost RR, Wilson DA, et al. Nifedipine pharmacokinetics and pharmacodynamics during the immediate postpartum period in patients with pre-eclampsia. Am J Obstet Gynecol 1991; 165:951–954. 56. Rogers RC, Sibai BM, Whybrew WD. Labetolol pharmacokinetics in pregnancy-induced hypertension. Am J Obstet Gynecol 1990; 162(2):362–366. 57. O’Hara MF, Leahey W, Murraghan GA, et al. Pharmacokinetics of sotalol during pregnancy. Eur J Clin Pharmacol 1983; 24(4):521–524. 58. Bologa M, Tang B, Klein J, et al. Pregnancy-induced changes in drug metabolism in epileptic women. J Pharmacol Exp Ther 1991; 257(2):735–740. 59. Henviksen O, Johannessen SI. Clinical and pharmacokinetic observations on sodium valproate: a five year follow-up study of 100 children with epilepsy. Acta Neurol Scand 1982; 65:504–523. 60. Pugh CB, Garnett WR. Current issues in the treatment of epilepsy. Clin Pharm 1991; 10: 335–358. 61. Heikkila A, Erkkola R. Review of β-lactam antibiotic in pregnancy. Clin Pharmacokinet 1994; 27(1):49–62. 62. Schentag JJ, Smith IL, Swanson DJ, et al. Role for individualization with cefmenoxime. Am J Med 1984; 77(suppl 6a):43–50. 63. Bergan T. Pharmacokinetics of tissue penetration of antibiotics. Rev Infect Dis 1981; 3:45– 66. 64. Barza M. Principles of tissue penetrations of antibiotics. J Antimicrob Chemother 1981; 8(suppl. C):7–28. 65. Schentag JJ. Clinical significance of antibiotic tissue penetration. Clin Pharmacokinet 1989; 16(suppl. 1):25–31. 66. Craig WA, Welling PG. Protein binding of antimicrobials: clinical pharmacokinetic and therapeutic implication. Clin Pharmacokinet 1977; 2:252–268. 67. Wise R, Gillet AP, Cadge B, et al. The influence of protein binding upon tissue fluid levels of six beta-lactams. J Infect Dis 1980; 142:77–82. 68. Drusano GL, Standiford HC, Bustamante C, et al. Multiple dose pharmacokinetics of imipenem/cilastatin (MK-787/MK-791). Antimicrob Agents Chemother 1984; 26:715–721. 69. Jones HM, Cummings AJ. A study of the transfer of alfamethyl dopa to the human foetus and nerborn infant. Br J Clin Pharmacol 1978; 6:432–434.

Pharmacokinetic Changes During Pregnancy

19

70. Kyegombe D, Franklin C, Turner P. Drug-metabolising enzymes in the human placenta, there induction and repression. Lancet 1973; 2:405–406. 71. Pelkonen O. Xenibiotic metabolism in the maternal-placental-fetal unit: implications for fetal toxicity. Dev Pharmacol Ther 1984; 1(suppl. 7):11–7. 72. Philipson A. Pharmacokinetics of ampicillin during pregnancy. J Infect Dis 1977; 136:370– 376. 73. Boreus L-O. Placental transfer of ampicillin in man. Acta Pharmacol Toxicol (Copenh) 1971; 3(Suppl):250–254. 74. Bastert G, Wallhauser K, Wernicke K, et al. Pharmacokinetische Untersuchungen zum Ubertritt von Antibiotica in das Fruchtwasser am Ende der Schwangerschaft: I. Teil: Ampicillin. Z Geburtshilfe Perinatol 1973; 178:164–173. 75. Voigt R, Schroder S, Meinhold P, et al. Klinische Unter-such Ungen zum einfluss von Schwangerschaft und geburt auf die Pharmacokinetik von Ampizillin. Zentralbl Gynakol 1978; 100: 701–705. 76. Kubacka R, Johnstone H, Tan H, et al. Intravenous ampicillin pharmacokinetics in the third trimester of pregnancy. Ther Drug Monit 1983; 5:55–60. 77. Adamkin DH, Marshall E, Weiner LB. The placental transfer of ampicillin. Am J Perinatol 1984; 1:310–311. 78. Lang R, Shalit I, Segal J. Maternal and fetal serum and tissue levels of ceftriaxone following preoperative prophylaxis in emergency caesarean section. Chemotherapy 1993; 39:77–81. 79. Depp R, Kind AC, Kirby WMN, et al. Transplacental passage of methicillin and dicloxacillin into the fetus and amniotic fluid. Am J Obstet Gynecol 1970; 107:1054–1057. 80. Daubenfeld O, Modde H, Hirsch HA. Transfer of gentamycin to the foetus and amniotic fluid during a steady state in the mother. Arch Gynecol 1974; 217:233–240. 81. Yoshioka H, Monma T, Matsudo S. Placental transfer of gentamycin. J Pediatr 1972; 80: 121–123. 82. Heikkila A, Erkkola R. The need for adjustment of dosage regimen for penicillin V during pregnancy. Obstet Gynecol 1993; 81(6):919–921. 83. Wasz-Hockert O, Nummi S, Vuopala S, et al. Transplacental passage of azidocillin, ampicillin and penicillin during early and late pregnancy. Scand J Infect Dis 1970; 2:125–130. 84. Heikkila A, Erkkola R. Pharmacokinetics of piperacillin during pregnancy. J Antimicrob Chemother 1991; 28:419–423. 85. Martens MG, Faro S, Feldman S, et al. Pharmacokinetics of the acyclureidopenicillins piperacillin and mezlocillin in the postpartum patient. Antimicrob Agents Chemother 1987; 31: 2015–2017. 86. Begue P, Quinet B, Baron S, et al. Clinical and pharmacokinetic study of imipenem/cilastatin in children and newborn infants. Pathol Biol 1989; 5:485–490. 87. Duval J, Mora M, Chartier M, et al. La cephalexine: son transfer placentaire, Nouv Presse Med 1972; 1:1419–1420. 88. MacAulay MA, Charles D. Placental transfer of cephalotin. Am J Obstet Gynecol 1968; 100: 940–946. 89. Bernard B, Barton L, Abate M, et al. Maternal-fetal transfer of cefazolin in the first twenty weeks of pregnancy. J Infect Dis 1977; 136:377–382. 90. Kafetzis DA, Brater CD, Fanourgakis JE, et al. Ceftriaxone distribution between maternal blood and fetal blood and tissues at parturition and between blood and milk post partum. Antimicrob Agents Chemother 1983; 23:870–873. 91. Bourget D, Fernandez H, Quinquis V, et al. Pharmacokinetics and protein binding of ceftriaxone during pregnancy. Antimicrob Agents Chemother 1993; 37(1):54–59. 92. Schou M, Amidsen A, Steenstrup DR. Lithium and pregnancy; hazards to women given lithium during pregnancy and delivery. BMJ 1973; 2:137–138. 93. Rogers ME, Willerson JT, Goldblatt A, et al. Serum digoxin concentrations in the human fetus, neonate and infant. N Engl J Med 1972; 287:1010–1013.

20

Loebstein et al.

94. Butters L, Kennedy S, Rubon P. Atenolol in essential hypertension during pregnancy. BMJ 1990; 301:587–589. 95. Thorley KJ, McAinsh J, Cruickshank JM. Atenolol in the treatment of pregnancy-induced hypertension. Br J Clin Pharmacol 1981; 12:725–730. 96. Magee LA, Schick B, Donnenfeld AE, et al. The safety of calcium channel blockers in human pregnancy: a prospective, multicenter cohort study. Am J Obstet Gynecol 1996; 174:823– 828. 97. Todd EL, Stafford DT, Bucovaz ET, et al. Pharmacokinetics of meperidine in pregnancy. Int J Gynaecol Obstet 1989; 29(2):143–146. 98. Martinez GA, Krieger FW. 1984 milk-feeding patterns in the United States. Pediatrics 1985; 76:1004–1008. 99. McNally E, Hendricks S, Horowitz I. A look at breast-feeding trends in Canada (1963– 1982). Can J Public Health 1985; 76:101–107. 100. Simopoulos AP, Grave GD. Factors associated with the choice and duration of infant-feeding practice. Pediatrics 1984; 74(4 pt 2):603–614. 101. Berlin CM. The excretion of drug and chemicals in human milk. In: Yaffe SJ, editor. Pediatric Pharmacology: Therapeutic Principles in Practice. New York: Grune & Stratton, 1980:137– 147. 102. Kafetzis DA, Siafas CA, Georgakopoulos PA, et al. Passage of cephalosporins and amoxicillin into the breast milk. Acta Paediatr Scand 1981; 70:285–288. 103. Thorley KJ, McAinsh J. Levels of the beta-blockers atenolol and propranolol in the breast milk of women treated for hypertension in pregnancy. Biopharm Drug Dispos 1983; 4:299– 301. 104. Schmimmel MS, Eldelman AJ, Wilschanski MA, et al. Toxic effects of atenolol consumed during breast feeding. J Pediatr 1989; 114:476–478. 105. Froescher W, Eichelbaum M, Niesen M, et al. Carbamazepine levels in breast milk. Ther Drug Monit 1984; 6:266–271. 106. Wesson DR, Camber S, Harkey M, et al. Diazepam and desmethyldiazepam in breastmilk. J Psychoactive Drugs 1985; 17:55–56. 107. Loughnan PM. Digoxin excretion in human breast milk. J Pediatr 1978; 92:1019–1020. 108. Sykes PA, Quarrie J, Alexander FW. Lithium carbonate and breast feeding. BMJ 1976; 2: 1299. 109. Greene HJ, Burkhart B, Hobby GL, et al. Excretion of penicillin in human milk following parturition. Am J Obstet Gynecol 1946; 51:732–733. 110. Steen B, Rane A, Lonnerholm G, et al. Phenytoin excretion in human breast milk and plsma levels in nursed infants. Ther Drug Monit 1982; 4:331–334. 111. Yurchav AM, Jusco WJ. Theophylline secretion into breast milk. Pediatrics 1976; 57:518– 520. 112. Orme ML’O, Lewis PJ, De Swiet M, et al. May mothers given warfarin breast-feed their infants? BMJ 1977; 1:564–565. 113. Catz CS, Giacoia GP. Drugs and breast milk. Pediatr Clin North Am 1972; 19:151–166. 114. World Health Organization Working Group. Determinants of drug excretion in breast milk. In: Bennett PN, Matheson I, Dukes NMG, et al., eds. Drugs and Human Lactation. Amsterdam: Elsevier, 1988:27–48. 115. Wilson JT. Determinants and consequences of drug excretion in breast milk. Drug Metab Rev 1983; 14:619–652. 116. Atkinson HC, Begg EJ. Prediction of drug concentrations in human milk from plasma protein binding and acid-base characteristics. Br J Clin Pharmacol 1988; 25:495–503. 117. Ito S, Koren G. A novel index for expressing exposure of the infant to drugs in breast milk. Br J Clin Pharmacol 1994; 38:99–102. 118. Braverman LE. Iodine-induced thyroid disease. Acta Med Aus 1990; 17:29–33. 119. Kulski JK, Hartmann PE, Martin JD, et al. Effects of bromocriptine mesylate on the composi-

Pharmacokinetic Changes During Pregnancy

120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135.

21

tion of the mammary secretion in non-breast-feeding women. Obstet Gynecol 1978; 52:38– 42. Katz M, Kroll D, Pak I, et al. Puerperal hypertention, stroke and seizures after suppression of lactation with bromocriptine. Obstet Gynecol 1985; 66:822–824. Wiernik PH, Duncan JH. Cyclophosphamide in human milk. Lancet 1971; 1:912. Amato D, Niblett JS. Neutropenia from cyclophosphamide in breast milk [letter]. Med J Aust 1977; 1:383–384. Fletcher SM, Katz AR, Rogers AJ, et al. The presence of cyclosporine in body tissue and fluids during pregnancy. Am J Kidney Dis 1985; 5:60–63. Egan PC, Costanza ME, Dodion P, et al. Doxorubicin and cisplatin excretion into human milk. Cancer Treat Rep 1985; 69:1387–1389. Tegler L, Lindstrom B. Antithyroid drugs in milk [letter]. Lancet 1980; 2:591. Kampmann JP, Johansen K, Hansen JM, et al. Propylthiouracil in human milk: revision of a dogma. Lancet 1980; 1:736–737. Johansen K, Anderson AN, Kampmann JP, et al. Excretion of methimazole in human milk. Eur J Clin Pharmacol 1982; 23:339–341. Cooper DS. Antithyroid drugs: to breast feed or not to breast feed. Am J Obstet Gynecol 1987; 157:234–235. Von Unruh GE, Froescher W, Hoffman F, et al. Valproic acid in breast milk: how much is really there? Ther Drug Monit 1984; 6:272–276. Kaneko S, Sato T, Suzuki K. The levels of anticonvulsants in breast milk. Br J Clin Pharmacol 1979; 7:624–627. White WB, Andreoli JW, Wong SH, et al. Atenolol in human plasma and breast milk. Obstet Gynecol 1984; 63:42S–44S. Liedholm H, Melander A, Bitzen PO, et al. Accumulation of atenolol and metoprolol in human breast milk. Eur J Clin Pharmacol 1981; 20:229–231. Lawton ME. Alcohol in breast milk. Aust NZJ Obstet Gynaecol 1985; 25:71–73. Kesaniemi YA. Ethanol and acetaldehyde in the milk and peripheral blood of lactating women after ethanol administration. J Obstet Gynaecol Br Commonw 1974; 81:84–86. Little RE, Anderson KW, Ervin CH, et al. Maternal alcohol use during breast feeding and infant mental and motor development at one year. N Engl J Med 1989; 321:425–430.

2 Developmental Risk Assessments Gerald F. Chernoff California Environmental Protection Agency, Sacramento, California

INTRODUCTION For those on the front line communicating risk information to pregnant women, one of the more frustrating questions that can be asked regards the risk of exposures to occupational chemicals and environmental hazards. Whereas information about pharmaceuticals is available from central sources such as the Catalog of Teratogenic Agents (1), information on occupational and environmental exposures is often available only from government agencies such as the U.S. Environmental Protection Agency (2). The information provided by these agencies can be quite limited, consisting of only permissible emission levels (PELs), maximum concentration limits (MCLs), acceptable daily intakes (ADIs), or some other regulatory number derived from a formal risk assessment. What is a risk assessment? What do the regulatory numbers mean? How can they be used in the risk communications process? In this chapter we investigate some of these questions. The intent is not to provide a formula for instant expertise on risk assessments but rather to broaden the reader’s appreciation for how these assessments can be used and misused in counseling concerned individuals. To accomplish this task, we review briefly the basic principles of each of the four steps in the risk assessment process: hazard identification, reference dose determination or dose-response assessment, exposure assessment, and risk characterization (3). After each step has been reviewed, a database of fictional studies on the imaginary chemical AVOID is presented. This database will help illustrate the types of study available for risk assessments and will provide an opportunity to apply the principles of risk assessment to representative data. Finally, questions germane to the evaluation of the AVOID data are listed and the issues that generate the most controversy are discussed briefly.

Originally presented as a workshop at the Fifth International Conference of Teratogen Information Services, March 1992. The material presented in this chapter represents the views of the author and does not necessarily reflect the policy of the Office of Environmental Health Hazard Assessment. 23

24

Chernoff

HAZARD IDENTIFICATION Principles 1.

2.

3.

4.

The purposes of hazard identification are to identify the types of adverse health effect that may be associated with exposure to an agent and to characterize the quality and strength of evidence supporting this identification. For the purpose of this exercise only developmental (teratogenic) adverse effects induced by AVOID are considered. Human studies are generally considered to be the best source of information, since ultimately it is the human population that we are attempting to protect. Unfortunately, for most chemicals of concern, good human studies are lacking; even when data do exist, establishing a causal link between the exposure and developmental endpoint is seldom possible. Human developmental studies can generally be divided into two categories, descriptive and analytical (4). a. Descriptive studies, which are useful for generating hypotheses, include case reports, surveillance systems, and ecological and cluster studies. b. Analytical studies, which test a hypothesis to examine cause–effect relationships, include case control, cohort, and human experimental studies. Studies with experimental animals are often the best source of information for hazard identification. These studies can be either regulatory or experimental. a. Regulatory studies are those that are prescribed or recommended by agencies such as the U.S. Environmental Protection Agency and the Food and Drug Administration; they include single- and multigeneration reproductive studies, continuous breeding studies, and developmental studies with exposure in the embryonic and fetal periods or in the perinatal period (5–10). These studies have the advantage of consistent protocols, with full reporting of all the collected data. Unfortunately, they apply to only a limited spectrum of agents, and they are generally not reported in the open scientific literature. b. Experimental studies are those conducted to investigate a hypothesis of interest to the investigator; as such, they do not conform to any one protocol. The results of these studies are reported in the open literature and often are your only source of information for hazard identification. In evaluating the individual human and animal studies, several important factors should be taken into consideration: a. Exposure parameters such as the route, time, and duration of exposure, and the actual or estimated dose of exposure should be well defined; for animal studies, these doses should be within the realm of expected human exposures. b. The endpoints evaluated should be described, and the methodology used in the evaluation should be appropriate. c. The presence of maternal toxicity should be evaluated (11). d. Appropriate statistical procedures should be used to demonstrate the significance of the adverse effect (12). For human studies, the strength of the association and study power should be discussed (13–15). e. The results should be evaluated for the presence of a dose-response relationship. f. The quality of each paper should be assessed using the foregoing considerations.

Developmental Risk Assessments

25

5. The evaluation of each individual study should lead to one of three conclusions: a. The study provides data indicative of an adverse effect. b. The study provides data indicative of no adverse effect. For a chemical to be placed in this category, it is essential that the study design be adequate to detect an adverse effect if one is present. c. The study provides inconclusive data. The reasons for placing a study in this category are many but usually involve inadequate or inappropriate study designs or incomplete reporting of the data. 6. After the individual studies have been evaluated, it is necessary to assess the total body of evidence. This is usually accomplished by evaluating the animal data separately from the human data. The following factors should be taken into consideration: a. Consistent adverse findings in two or more studies that use different study designs and populations are generally regarded as evidence of a causal relationship. b. The evidence that an agent is a developmental toxicant may be strengthened by demonstrating biological plausibility of a causal relationship. This is an especially important consideration in evaluating a group of epidemiological studies. c. Pharmacokinetic and pharmacogenetic differences that may account for differences between studies should be evaluated. d. Structural relationships and other evidence for chemical similarity may in some cases be useful in drawing inferences of potential developmental toxicity. 7. The final step in hazard identification is a weight of evidence determination. Different agencies use different schemes, which range from the very simple to the very complex (16,17). Most of these schemes can be reduced to three basic categories: a. Sufficient human evidence: Data from human studies provide sufficient evidence for the scientific community to judge that a causal relationship is or is not supported. Supporting animal data may or may not be available. b. Sufficient experimental animal evidence/limited human data: Data from experimental animal studies and/or limited human data that provide sufficient evidence for the scientific community to judge that the potential for developmental toxicity does or does not exist. c. Insufficient evidence: Data are not available or the data that are available are based on human or experimental animal studies that are flawed in design.

Database Eight studies on AVOID, three in humans and five in animal, make up the database available for the hazard identification step. Each study is briefly summarized below. Study 1 There have been a series of anecdotal reports from the union health and safety committee at the formulation and packaging facility, as well as from rural health clinics, suggesting that pregnant workers exposed to AVOID experience a high incidence of spontaneous abortion. These reports have not been published or reviewed in any scientific forum.

26

Chernoff

Table 1 Clinical Symptoms and Newborn Evaluations Reported in Study 2 Case

Clinical symptoms

1 2 3 4

Sweating, Sweating, Sweating, Sweating,

tremors tremors tremors tremors, convulsions

5 6

Sweating, tremors, convulsions Sweating, tremors, coma

Newborn evaluation Normal physical and growth Normal physical and growth Spina bifida with myeloschisis Normal physical; pre- and postnatal growth retardation Fetal alcohol syndrome Normal physical and growth

Study 2 A paper on accidental and purposeful poisoning with AVOID was recently published in a reputable peer-reviewed medical journal. Forty cases were described, and six involved women who were estimated to be in their first trimester of pregnancy at the time of the poisoning. The poisoning was attributed to unsuccessful suicide attempt by AVOID ingestion in five of the cases and to attempted homicide with AVOID-poisoned food in one case. The clinical symptoms recorded at the time of admission to the emergency room for the poisoning and the pediatrician’s evaluation of each newborn in the perinatal period are shown in Table 1. Study 3 An epidemiological study was conducted in a plant where AVOID was formulated and packaged. The study, which was conducted for the plant owners by investigators from Cosmic University, has never been published. The study population consisted of women who had worked in the packaging section of the plant for at least 6 months between 1981 and 1985. Over this 4-year period, the mean combined dose of AVOID from respiration and dermal absorption was estimated to be less than 0.07 mg/kg/day in the packaging department. The highest combined dose during this same period was estimated at 0.26 mg/kg/day. The control group consisted of female clerical staff who worked in plant offices, where it was assumed that there was no exposure to AVOID. After identifying the study and control populations, the investigators sent each woman a questionnaire regarding any pregnancies she may have had while working at the plant. Included were questions on each respondent’s length of time working in the plant, her work location, and the outcome of pregnancy. Response rates to the questionnaire were 37% for the study group and 42% for the controls. Table 2 gives the major results as reported by the investigative team. Study 4 In an unpublished study conducted for the manufacture of AVOID, groups of 20 C57 mated mice were administered AVOID in their diets at doses of 0, 25, or 75 mg/kg/day from days 6–18 of gestation. At the high dose tested, there was a significant decrease in maternal food consumption between days 6 and 12, and several of the females exhibited convulsions. Fetal resorptions and exencephaly were significantly increased. At the middose, there was a nonsignificant increase in cleft palate. Major study results are summarized in Table 3.

Developmental Risk Assessments Table 2 Study 3

27

Summary of Major Results Reported in

Group Respondents Pregnancies Abortions Therapeutic Spontaneous Stillbirths Live births Normal Abnormal Anencephaly Extra finger Heart defect Holoprosencephaly Trisomy 21

Study 56 110

Control 48 86

10 (9%) 10 (9%) 1 (1%) 89 (81%) 85 (96%) 4 (4%) 1 1 1 1 0

1 (1%) 9 (10%) 0 76 (88%) 74 (97%) 2 (3%) 0 0 1 0 1

Study 5 A meeting abstract reported that groups of five C3H mice administered AVOID by intraperitoneal injection at 5 mg/kg/day on days 10, 11, and 12 of gestation exhibited a significant increase in cleft palate (86%) compared to controls. No mention was made of maternal toxicity or the presence of any other malformations in the treated group. Study 6 In an unpublished teratology study conducted for the manufacturer of AVOID, groups of 20 pregnant SD rats were administered AVOID in their diets at doses sufficient to provide 0, 0.2, 2.0, or 20 mg/kg/day from days 6–20 of gestation. At the high dose, several females died between days 10 and 18 of gestation; between days 7 and 15, maternal food consumpTable 3 Summary of Major Results Reported in Study 4 Number of cases per dose (mg/kg/day) group

Mated Pregnant Resorbed litters Live litters Live fetuses Resorbed fetuses Abnormal fetuses Delayed ossification Cleft palate Exencephaly Anophthalmia a

0

25

75

20 15 0 15 125 7 4 4 1 0 0

20 13 0 13 99 6 9 3 7 0 0

20 20 10 a 10 52 a 37 a 42 a 7 0 38 a 2

Significantly different from control ( p ⬍ 0.05).

28

Chernoff

tion and weight gain were significantly depressed, and a statistically significant increase in fetal resorptions and exencephaly was noted. At 2.0 mg/kg/day there was a significant increase in microphthalmia and wavy ribs. Wavy ribs were also noted at 0.2 mg/kg/day, but the incidence did not reach statistical significance. The key results of this study are summarized in Table 4. Study 7 In an unpublished study conducted for the manufacturer, groups of 20 pregnant CR rats were exposed to AVOID by inhalation at doses of 0, 0.16, 0.3, or 0.64 mg/m 3 for 6 hours per day from days 6–20 of gestation. At the high dose there was a slight decrease in maternal food consumption between days 6 and 7 of gestation and a significant increase in microphthalmia. No significant findings were reported at the other doses tested. The key results of this study are summarized in Table 5. Study 8 The final animal study was an unpublished report using groups of 20 NZW rabbits administered AVOID by oral gavage at doses of 0, 0.2, 20, or 200 mg/kg/day from days 7–19 of gestation. Maternal toxicity was observed at the high dose, with a significant decrease in body weight and increase in liver weight. At this dose there was also a statistically significant increase in resorptions and exencephaly. No significant findings were reported at any of the other dose levels tested. Questions to Ask When Evaluating Hazard Identification Data 1. 2. 3.

How do these data conform (or not conform) to the principles used in hazard identification? What is the most sensitive endpoint of potential teratogenicity in the animal and human studies? Should the wavy ribs in rodents be considered relevant to low exposure risks to humans?

Table 4 Summary of Major Results Reported in Study 6 Number of cases per dose (mg/kg/day) group

Mated Pregnant Resorbed litters Live litters Live fetuses Resorbed fetuses Abnormal fetuses Delayed ossification Wavy ribs Exencephaly Microphthalmia a

0

0.2

2.0

20

20 19 1 18 175 10 6 4 4 1 1

20 20 0 20 182 5 13 5 7 0 0

20 18 0 18 179 7 27 a 3 15 a 0 12 a

20 20 6a 6a 32 a 38 a 32 a 32 a 5 29 a 2

Significantly different from controls ( p ⬍ 0.05).

Developmental Risk Assessments Table 5

29

Summary of Major Results Reported in Study 7 Number of cases per dose (mg/m 3 ) group

Mated Pregnant Resorbed litters Live litters Live fetuses Resorbed fetuses Abnormal fetuses Delayed ossification Wavy ribs Microphthalmia a

0

0.16

0.3

0.64

20 20 1 19 193 7 7 3 7 1

20 17 0 17 180 9 10 2 5 4

20 17 2 16 169 4 3 3 0 0

20 20 0 20 198 5 25 a 4 3 21 a

Significantly different from controls ( p ⬍ 0.05).

4. Should the data obtained by IP injection treatment be considered relevant to human exposure? 5. Should the evidence for maternal toxicity at high doses in the animal studies negate considering the adverse effects seen in the fetuses as true developmental effects? 6. From the data given, is there any way to determine whether responses in humans are likely to be similar to those in the experimental animals? 7. Do the data provide sufficient evidence to convince you that AVOID should be considered to be a teratogen? There are no simple or straightforward answers to these questions, and the underlying issues continue to be debated. One area of lively controversy is the role of maternal toxicity in interpreting study results (11,18–20). Can a 20% decrease in maternal weight gain in the early part of organogenesis negate the finding of a high incidence of cleft palate or decreased fetal weight? These issues remain unresolved. Similarly, the importance of skeletal variations such as wavy ribs or partially ossified sternebra remains controversial and as yet unresolved (21–23). At the conclusion of the hazard identification step in the risk assessment process, a choice must be made. It must be decided which of the following conclusions the data support: (1) AVOID is teratogenic in humans; (2) it is highly likely that AVOID is a teratogen in humans; (3) AVOID is a potential human teratogen; or (4) AVOID is not classifiable as to human teratogenicity.

REFERENCE DOSE DETERMINATION Principles 1. The reference dose (RfD) is defined as an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily exposure to the human population (including sensitive subgroups) that is assumed to be without appreciable risk of deleterious developmental effects (24).

30

Chernoff

2. 3.

4.

5.

The RfD is usually derived from the ‘‘no observed adverse effect level’’ (NOAEL) or the ‘‘lowest observed adverse effect level’’ (LOAEL). The most appropriate NOAEL is determined from the body of evidence examined in the hazard identification stage. a. Dose-response data from human studies are generally preferred over animal data provided that the human data are sufficiently quantitative. b. Data on the most sensitive relevant endpoint should be used. To calculate the RfD, the NOAEL is divided by an overall uncertainty factor (UF) ranging from 10 to 10,000 (25). The total UF is calculated by multiplying together appropriate individual UFs of 10: a. UF for intraspecies variability ⫽ 10 b. UF for interspecies variability ⫽ 10 c. UF for different exposure scenarios ⫽ 10 d. UF when using the LOAEL rather than NOAEL ⫽ 10. The RfD may be further reduced by applying a modifying factor greater than zero but less or equal to 10, which may be used to reflect qualitative professional judgments about scientific uncertainties such as the completeness of the overall database and the number of species and animals tested.

Database The data to be used in this section consist of the same eight studies used in the hazardidentification step. Questions to Ask When Evaluating Reference Dose Data 1. 2. 3.

4. 5. 6. 7. 8.

What study is the most appropriate for deriving a no observed adverse effect level (NOAEL) for AVOID? What do you consider to be the appropriate NOAEL? Is the observed NOAEL from the studies a true ‘‘no-effect’’ level? Could it simply reflect the fact that in experiments with relatively small numbers, the failure to observe a statistically significant increase in adverse developmental effects is an artifact of the experimental design, not a true absence of biological effect? What uncertainty factors should be applied to the NOAEL for AVOID? What is the appropriate reference dose (RfD) for AVOID? Does the RfD adequately account for the uncertainties associated with the NOAEL? Is it appropriate to use an RfD derived from a study using one route of exposure for all routes of exposure? Is the RfD a reliable indicator of human risk? Are there any other conditions that should be applied to this number?

Again, there are no simple or straightforward answers to these questions. Of greatest controversy is the use of NOAELs and safety factors (26). It is now generally recognized that the determination of a NOAEL is highly dependent on the sample size and dose levels used. Methods using various mathematical models that are less sensitive to sample size and utilize the full dose-response curve have been proposed as alternatives

Developmental Risk Assessments

31

to the NOAEL/UF approach (27–31). No one method has yet gained favor, and the majority of risk assessments still utilize the NOAEL-derived RfD. After completing the RfD-determination step in the risk-assessment process, you may conclude that the RfD you calculated, and the NOAEL from which it was derived, are sufficient to determine the developmental risks associated with exposure to AVOID. Alternatively, you may wish to conclude that RfDs are of little value and in fact create a false sense of precision in an area involving tremendous uncertainty. A third possible conclusion is that while risks from developmental toxicants cannot be quantified, they should be described in qualitative terms.

EXPOSURE ASSESSMENT Principles 1. An exposure assessment serves to identify the magnitude of human exposure to an agent, the frequency and duration of that exposure, and the routes by which humans are exposed (32). It may be useful to identify the number of exposed people along with other characteristics of the exposed population, such as age, genetic history, and exposure to (other) known teratogens. 2. Exposure may be based on quantitative or qualitative measurements in various media such as air, water, or food. a. Daily intake of individual and combined media exposures should be determined. b. When individuals may be exposed by contact with several media, it is important to consider total intake from all media. c. Daily intake under different conditions of activities in different locations should be determined. 3. Sampling and monitoring of individual exposures is usually conducted by contractors for either the responsible parties or regulatory bodies. 4. Usually only a limited amount of monitoring and only a limited number of samples of various media can be taken for measurement. a. The representativeness of measured values is usually uncertain. b. The degree to which data for a given medium are representative of that medium should be estimated. 5. When data are incomplete or lacking, mathematical models may be used to estimate air and water concentrations. The validity of the model should be checked in context of the exposure scenario. 6. Standard average values and ranges for human intake of various media are available and are generally used unless data on specific agents indicate that such values are inappropriate (33). a. Adults drink approximately 2 L of water per day. b. The average adult inhales 23 m 2 of air per day.

Database Four monitoring studies have been conducted on AVOID: two were conducted in workplace settings: one on air and one on water. A brief description is provided below.

32

Chernoff

Table 6 Exposure Levels Reported in Study 2 Route of exposure Workers Pilots Aerial mix/load Flagger Ground mix/load Ground applications Combined mix/load/applications Gofer

Dermal

Inhalation

0.50 0.02 0.53 0.05 0.02 0.10 0.03

1.13 0.53 1.73 0.85 0.37 1.77 0.70

Study 1 In conjunction with the epidemiology study reported earlier, a 3-day study was conducted in the plant during which time workers packaging AVOID were monitored for exposure via the inhalation and dermal routes. The data collected were used to estimate the mean combined dose, which was calculated to be 0.07 mg/kg/day. The highest dose received by a single worker on any one day was 0.26 mg/kg. Study 2 A study was conducted by a state regulatory agency that used passive dosimetry to monitor field worker exposure to AVOID. From the data collected, the absorbed dosage was calculated. The relative contributions of the inhalation and dermal routes of exposure were back-calculated from the total absorbed dosage, assuming that a certain proportion of AVOID exposure was through inhalation. This proportion was based on measured concentrations of AVOID in the air, standard respiratory rate and retention, and 100% absorption. The calculated absorbed exposure dosage for a person weighing 70 kg expressed micrograms per kilogram per day is shown in Table 6. Study 3 An air-monitoring study was conducted in four rural communities near fields where AVOID was being used. Twenty-four-hour collections were conducted over three 4-day periods in 1987. The mean concentrations (ng/m 3 ) of air-borne AVOID over the 12 sampling days, along with the range of values, were published in a reputable peer-reviewed journal and are shown in Table 7. Study 4 A water-monitoring study was conducted using samples collected from the major river that runs through the agricultural area and past several metropolitan areas downstream. Samples were collected near each of the metropolitan areas, one per month from May Table 7 Sampling Results Reported in Study 3 Location

Mean Range

1

2

3

4

62 2–142

132 ⬍1.4–280

152 76–415

630 145–1720

Developmental Risk Assessments

33

through September 1989. The results of the study, which appear in a government report, indicate that only trace amounts of AVOID were detected. The validity of this study has been challenged by several environmental advocacy groups. Questions to Ask When Evaluating Exposure Data 1. Do the monitoring data adequately describe exposures to AVOID in all possible media? If not, what additional media need to be considered, and why? 2. Should the different types of exposure data be treated the same for purposes of characterizing human risk? 3. Is the mean concentration in the various media the appropriate summary statistic to use to characterize human exposure? Should the upper range or statistical upper confidence limit be used as an alternative? 4. Are the various assumptions about human intake and average exposure to various media valid? 5. Should exposure and risk to workers at the production facility be considered in the same context as exposure and risk to fieldworkers or residents in exposed communities? It should come as no surprise that there are no simple or straightforward answers to these questions. Debate continues on the use of various models for estimating exposures, but generally the issue of greatest concern is the method of extrapolating from one route of exposure to another (34,35). This is typified by the debate over the various routes of exposure following an animal whole-body versus nose-only exposure. With whole-body exposure, it must be decided how much of the agent enters via the respiratory route, how much via the dermal route, and how much via ingestion from grooming behavior and contaminated food. The rates of absorption for these various routes may differ rather dramatically. With nose-only exposures, the majority of the exposure is considered to be respiratory, but dermal adsorption from the nose area and ingestion by swallowing cannot be ruled out. After the exposure-assessment step in the risk-assessment process has been completed, you may conclude that although different exposure estimates are based on different data and assumptions, they are all adequate and sufficient for assessing risks. Alternatively, you may wish to conclude that none of the exposure data are adequate for use in risk assessment and that no quantitative risk assessment should be developed until better information is available. RISK CHARACTERIZATION Principles 1. The purpose of risk characterization, the final step of the risk assessment, is to integrate the information collected and analyzed in the first three steps to characterize the excess risk to humans. 2. An explicit numerical RfD should be included in the characterization. 3. Compare the exposures experienced or expected for different groups of individuals. 4. Estimate the margin of exposure (MOE) for each group by dividing the NOAEL from the critical study used to estimate the RfD by the exposure for each group.

34

Chernoff

5. 6.

Describe risks qualitatively for each population group. Describe the statistical and biological uncertainties in estimating the extent of adverse health effects.

Database The data used in the risk-characterization step are those accumulated in the preceding steps of the risk-assessment process. The risk characterization can be thought of as the conclusion of the risk assessment, summarizing the information from the preceding steps (hazard identification, RfD determination, and exposure assessment). As such, the issues of controversy discussed earlier also apply to this final step.

CONCLUSION Having gained some appreciation of risk assessments, we can now consider how they can be used in the risk-communication process. Paul Peters (personal communications) has observed that there are two types of risk information: ‘‘ready made’’ and ‘‘tailor made.’’ The RfDs developed in the risk assessment process are applicable to all individuals in the population, and as such, represent ‘‘ready made’’ information. They serve as a powerful public health tool in defining the exposure level below which it can be assumed that no adverse developmental effects will occur. In contrast, individual risk counseling requires tailor-made information. Typically, this information is based on interpreting the results of human and animal studies in the context of a pregnant woman’s age, parity, genetic background, and exposure to other chemicals of concern. The ready-made information from the risk assessment can have value beyond the public health perspective. While using an RfD as the sole basis for individual risk counseling is never appropriate, the studies that served as the basis for obtaining the NOAEL on which the RfD is based can serve as the starting material for crafting the tailor-made information needed to communicate risk information to individual pregnant women.

ACKNOWLEDGMENTS I am indebted to the U.S. Environmental Protection Agency’s Workshop on Risk and Decision Making, which served as a model for the workshop on which this chapter is based. I am also grateful to Linda Chernoff and Dr. Paul Peters for their help and insight in planning this project, and to Dr. Gideon Koren for his encouragement to complete the project.

REFERENCES 1. Shepard TH. Catalog of Teratogenic Agents, 6th ed. Baltimore: Johns Hopkins University Press, 1989. 2. U.S. Environmental Protection Agency. Integrated Risk Information Service (IRIS). Online. Washington, DC: Office of Health and Environmental Assessment, 1991.

Developmental Risk Assessments

35

3. National Research Council. Risk Assessment in the Federal Government: Managing the Process. Committee on the Institutional Means for the Assessment of Risks to Public Health. Commission on Life Sciences, National Research Council. Washington, DC: National Academy Press, 1983, pp 17–83. 4. Erickson JD. Epidemiology and developmental toxicology, In: Kimmel CA, Buelke-Sam J, eds. Developmental Toxicology. New York: Raven Press, 1981, pp 289–301. 5. U.S. Environmental Protection Agency. Pesticide assessment guidelines, subdivision F. Hazard evaluation: Human and domestic animals, EPA-540/9-82-025. Washington, DC: Office of Pesticides and Toxic Substances, 1982. Available from NTIS, Springfield, VA. 6. U.S. Environmental Protection Agency. Toxic Substances Control Act test guidelines; final rules. Fed Reg 1985; 50:39426–39428 and 39433–39434. 7. U.S. Environmental Protection Agency. Pesticide Assessment guidelines, subdivision F. Hazard evaluation: Human and domestic animals, EPA 540/09-91–123. Addendum 10: Neurotoxicity, series 81–83. Washington, DC: Office of Pesticides and Toxic Substances. 1991. Available from NTIS, Springfield, VA. 8. Organization for Economic Cooperation and Development. Guidelines for Testing of Chemicals’ Teratogenicity, OECD, 1981. 9. U.S. Food and Drug Administration. Guidelines for reproduction and studies for human use. Rockville, MD: Bureau of Drugs, 1966. 10. U.S. Food and Drug Administration. Advisory Committee on Protocols for Safety Evaluation. Panel on reproduction—Studies in the safety evaluation of additives and pesticide residues. Toxicol Appl Pharmacol 1970; 16:264–296. 11. Kimmel GL, Kimmel CA, Francis EZ, eds. Evaluation of maternal and developmental toxicity. Teratogenesis Carcinog Mutagen 1987; 7:203–338. 12. Kimmel CA, Kimmel GL, Frankos V, eds. Interagency Regulatory Liaison Group Workshop on Reproductive Toxicity Risk Assessment. Environ Health Perspect 1986; 86:193–221. 13. Bloom AD. Guidelines for reproductive studies in exposed human populations. Report of Panel II. In: Bloom AD, ed. Guidelines for Studies of Human Populations Exposed to Mutagenic and Reproductive Hazards. White Plains, NY: March of Dimes Birth Defects Foundation, 1981, pp 37–110. 14. Stein Z, Kline J, Shrout P. Power in surveillance. In: Hemminki K, Sorsa M, Vaninio H, eds. Occupational Hazards and Reproduction. Washington, DC: Hemisphere, 1985, pp 203– 208. 15. Greenland S. Quantitative methods in the review of epidemiologic literature. Epidemiol Rev 1987; 9:1–30. 16. U.S. Environmental Protection Agency. Guidelines for developmental toxicity risk assessment. Fed Reg 1991; 56:63798–63826. 17. California Department of Health Services. Draft guidelines for hazard identification and doseresponse assessment of agents causing developmental and/or reproductive toxicity, Sacramento, CA: Office of Environmental Health Hazard Assessment, 1991. 18. Schardein JL. Approaches to defining the relationship of maternal and developmental toxicity. Teratogenesis Carcinog Mutagen 1987; 7:255–271. 19. Johnson E, Christian M. When is a teratology study not an evaluation of teratogenicity? J Am Coll Toxicol 1984; 3:431–434. 20. Black DL, Marks TA. Role of maternal toxicity in assessing developmental toxicity in animals: a discussion. Regul Toxicol Pharmacol 1992; 16:189–201. 21. Kimmel CA, Wilson JG. Skeletal deviations in rats: malformations or variations? Teratology 1973; 8:309–316. 22. Chernoff N, Rogers JM, Turner CI, Francis BM. Significance of supernumerary ribs in rodent developmental toxicity studies: postnatal persistence in rats and mice. Fundam Appl Toxicol 1991; 17:448–453. 23. Palmer AK. Incidence of sporadic malformations, anomalies and variations in random-bred

36

24. 25. 26. 27. 28. 29. 30. 31. 32. 33.

34.

35.

Chernoff laboratory animals. In Neubert D, Merker HJ, Kwasigroch TE, eds. Methods in Prenatal Toxicology. Stuttgart: Thieme, 1977, pp 52–71. Barnes DG, Dourson M. Reference dose (RfD): Description and use in health risk assessments. Regul Toxicol Pharmacol 1988; 8:471–486. Dourson M, Stara J. Regulatory history and experimental support of uncertainty (safety) factors. Regul Toxicol Pharmacol 1983; 3:224–238. Gaylor DW. Incidence of developmental defects at the no observed adverse effect level (NOAEL). Regul Toxicol Pharmacol 1992; 15:151–160. Crump KS. A new method for determining allowable daily intakes. Fundam Appl Toxicol 1984; 4:854–871. Gaylor DW. Quantitative risk analysis for quantal reproductive and developmental effects. Environ Health Perspect 1989; 79:243–246. Kimmel C, Gaylor D. Issues in qualitative and quantitative risk analysis for developmental toxicology. Risk Anal 1988; 8:15–20. Kodell RL, Howe RB, Chen JJ, Gaylor DW. Mathematical modeling of reproductive and developmental toxic effects for quantitative risk assessment. Risk Anal 1991; 11:583–590. Ryan L. The use of generalized estimating equations for risk assessment in developmental toxicity. Risk Anal 1992; 12:439–447. U.S. Environmental Protection Agency. Guidelines for exposure assessment. Fed Reg 1986; 51:34042–34054. U.S. Environmental Protection Agency. Exposure Factors Handbook, EPA-600/8-89-043. Washington, DC: Office of Health and Environmental Assessment, 1989. Available from NTIS, Springfield, VA. Sachsse K, Zbinden K, Ullman L. Significance of mode of exposure in aerosol inhalation toxicity studies: Head-only versus whole-body exposure. Arch Toxicol Suppl 1980; 4:305– 311. Iwasaki M, Yoshida M, Ikeda T, Tsuda S, Shirasu Y. Comparison of whole-body versus snoutonly exposure in inhalation toxicity of fenthion. Jpn J Vet Sci 1988; 50:23–30.

3 Drugs in Pregnancy Gideon Koren, Anne Pastuszak, and Shinya Ito The Hospital for Sick Children, Toronto, Ontario, Canada

INTRODUCTION Before marketing a new drug, the manufacturer almost never tests the product in pregnant women to determine its effects on the fetus. Consequently, most drugs are not labeled for use during pregnancy. Typically, descriptions of drugs that appear in the Physicians’ Desk Reference and similar sources contain statements such as, ‘‘Use in pregnancy is not recommended unless the potential benefits justify the potential risks to the fetus.’’ Since the risk has been adequately established for only a few drugs, physicians caring for pregnant women have very little information to help them decide whether the potential benefits to the mother outweigh the risks to the fetus. These typical disclaimers, although understandable from the medicolegal standpoint, put large numbers of women and their physicians in difficult situations for several reasons. One is that at least half the pregnancies in North America are unplanned (1), and every year, hundreds of thousands of women therefore expose their fetuses to drugs before they know they are pregnant. Such women often interpret the statement that use during pregnancy is not recommended as meaning that the drug is not safe during pregnancy. There is evidence that this perception of fetal risk causes many women to consider or even seek termination of otherwise wanted pregnancies (2,3). Another reason is that with the recent increase in the age at which women have children, conditions that necessitate long-term drug therapy are diagnosed in larger numbers of women before pregnancy. Furthermore, for pregnant women with certain conditions once believed to be incompatible with pregnancy, such as systemic lupus erythematosus and heart diseases, the outcome of pregnancy has improved dramatically in the past few decades (4). In this article, we review current knowledge of the fetal and neonatal effects of prescription and over-the-counter drugs given to pregnant women, with an emphasis on the approaches used to determine safety and risk. In addition, we review approaches to communicating such information to pregnant women and their families.

From Drug Therapy, A.J.J. Wood, ed. N Engl J Med 1998; 338:1128–1137.

37

38

Koren et al.

HUMAN TERATOGENESIS Teratogenesis is defined as the dysgenesis of fetal organs as evidenced either structurally or functionally (e.g., brain functions) (5). The typical manifestations of teratogenesis are restricted growth or death of the fetus, carcinogenesis, and malformations (6), defined as defects in organ structure or function. These abnormalities vary in severity (e.g., hypospadias that is mild and may be missed, or is severe, necessitating several corrective operations). Major malformations may be life-threatening and require major surgery or may have serious cosmetic or functional effects.

A HISTORICAL PERSPECTIVE Several milestones highlight the problems of drug therapy facing pregnant women, their families, and health professionals.

Thalidomide For decades it was believed that the placenta served as a barrier that protected the fetus from the adverse effects of drugs. The thalidomide disaster drastically changed this perception by demonstrating that fetal exposure to the drug during critical periods of development resulted in severe limb defects and other organ dysgenesis (e.g., kidney and heart defects) (6,7). Despite the high rates of malformations (20 to 30 percent) and their characteristic pattern, the teratogenicity of thalidomide was not suspected for years. The suffering it caused has prompted the belief that every drug has the potential to be a new thalidomide (2,3). Most known human teratogens are associated with much lower rates of malformations, and the syndromes they cause are not always so pathognomonic, making causation more difficult to confirm. Yet 35 years after the recognition of thalidomide-associated embryopathy, fewer than 30 drugs have been proved to be teratogenic in humans when used in clinically effective doses, and even fewer are currently in clinical use (Table 1). Many other commonly used drugs, including salicylates, glucocorticoids, and spermicides, were once thought to be teratogenic but have been shown to be safe in subsequent studies that were larger and better controlled than the initial studies (Table 2).

Bendectin One example of the gap between the perception of teratogenic risk and evidence-based proof of safety is the case of Bendectin. During the late 1950s and the 1960s, this drug, a combination of an antihistamine (doxylamine) and pyridoxine, was the most widely used medication in the United States for nausea and vomiting associated with pregnancy. During the 1970s, many lawsuits claiming that Bendectin was teratogenic were filed against the manufacturer in American courts. Therefore, the drug was withdrawn from the market by its manufacturer in 1982, which left millions of pregnant women without a drug approved by the Food and Drug Administration (FDA) for the treatment of nausea and vomiting. The rate of hospitalization for severe nausea and vomiting during pregnancy

Drugs in Pregnancy Table 1

39

Drugs with Proven Teratogenic Effects in Humans a

Drug Aminopterin b, methotrexate Angiotensin-converting-enzyme inhibitors Anticholinergic drugs Antithyroid drugs (propylthiouracil and methimazole) Carbamazepine Cyclophosphamide Danazol and other androgenic drugs Diethylstilbestrol b Hypoglycemic drugs Lithium Misoprostol Nonsteroidal antiinflammatory drugs Paramethadione b Phenytoin Psychoactive drugs (e.g., barbiturates, opioids, and benzodiazepines) Systemic retinoids (isotretinoin and etretinate) Tetracycline Thalidomide Trimethadione b Valproic acid Warfarin

Teratogenic effect CNS and limb malformations Prolonged renal failure in neonates, decreased skull ossification, renal tubular dysgenesis Neonatal meconium ilcus Fetal and neonatal goiter and hypothyroidism, aplasia cutis (with methimazole) Neural-tube defects CNS malformations, secondary cancer Masculinization of female fetuses Vaginal carcinoma and other genitourinary defects in female and male offspring Neonatal hypoglycemia Ebstein’s anomaly Moebius sequence Constriction of the ductus arteriosus c, necrotizing enterocolitis Facial and CNS defects Growth retardation, CNS deficits Neonatal withdrawal syndrome when drug is taken in late pregnancy CNS, craniofacial, cardiovascular, and other defects Anomalies of teeth and bone Limb-shortening defects, internal-organ defects Facial and CNS defects Neural-tube defects Skeletal and CNS defects, Dandy–Walker syndrome

a

Only drugs that are teratogenic when used at clinically recommended doses are listed. The list includes all drugs proved to affect neonatal morphology or brain development and some of the toxic manifestations predicted on the basis of the pharmacologic actions of the drugs. Data are from Briggs et al. (8). CNS denotes central nervous system. b The drug is not currently in clinical use. c Sulindac probably does not have this effect.

increased by a factor of 2 in both the United States and Canada after Bendectin was withdrawn from the market (Fig. 1). The drug was withdrawn despite a substantial body of evidence that the rate of major malformations among the children of women who had received Bendectin during pregnancy did not differ from the rate in the general population (24,25). Withdrawal of the drug from the American market did not decrease the rate of any specific category of malformation, as would be expected for a truly teratogenic drug estimated to have been used by up to 40 percent of pregnant women at one time (26,27). In Canada, the drug continues to be marketed under the trade name Diclectin. A review committee has advised the Canadian Minister of Health that the drug is safe (27). A recent study revealed that severe nausea and vomiting of pregnancy often lead women to terminate or consider the termination of otherwise wanted pregnancies (28). Other for-

40

Koren et al.

Table 2 Common Drugs Initially Thought To Be Teratogenic But Subsequently Proved Safe Drug

Initial evidence of risk

Diazepam a

Oral clefts a

Oral contraceptives

Birth defects involving the vertebrae, anus, heart, trachea, esophagus, kidney, and limbs (13); masculinizing effects on female fetuses resulting in pseudohermaphroditism (16) Limb defects, tumors, Down’s syndrome, and hypospadias (17) Cleft palate (19) and congenital heart disease Cardiac and limb defects (22,23)

Spermicides Salicylates Bendectin (doxylamine plus pyridoxine) a

Subsequent evidence of safety No increase in risk in large cohort and case-control studies (10–12) No association between first-trimester exposure to oral contraceptives and malformations in general or external genital malformations in two meta-analyses (15,16) No increase in risk in a meta-analysis (18) No increase in risk in large cohort studies (21,21) No increase in risk in two metaanalyses (24,25)

Diazepam taken near term may cause the neonatal withdrawal syndrome or cardiorespiratory instability.

Figure 1 Rates of Hospitalization among Pregnant Women with Severe Nausea and Vomiting and Numbers of Prescriptions for Bendectin and Diclectin in North America, 1979 through 1989. Bendectin was withdrawn from the U.S. market in 1982, whereas Diclectin, the same drug, remained on the market in Canada. Adapted from Neutel and Johansen with the permission of the publisher (26).

Drugs in Pregnancy

41

mulations of doxylamine in combination with pyridoxine are available in other countries (e.g., South Africa, Spain, and Thailand).

Isotretinoin The experience with thalidomide led drug regulators, drug manufacturers, and the medical community to believe that appropriate labeling of teratogenic drugs, with warnings not to take them around the time of conception, would be effective in preventing fetal exposure to the drugs. The naivete´ of this belief became evident after isotretinoin was introduced in North America in the early 1980s for the treatment of acne. For years before its clinical introduction, this drug had been known to cause malformations in animals (29). Despite explicit warning labels, scores of children with retinoid embryopathy were born in the years after the drug was introduced (30). Such warnings are not sufficient, because women taking isotretinoin may not plan their pregnancies, or their birth-control methods may fail. In addition, some women and men are functionally illiterate, and they may not read or understand the content of a drug label (31). The initial experience with isotretinoin led to the development of a more comprehensive program to prevent teratogenesis. The Retinoid Pregnancy Prevention Program includes explicit and detailed printed warnings as well as a line drawing of a malformed child (32), and as part of the program, women are asked to sign a consent form indicating that they agree to use two effective methods of contraception before therapy is started. Since the program was implemented in 1989, a substantial number of fetuses have been exposed to the drug. As many as 30 percent of the women with exposed fetuses did not use any mode of contraception, even though they were cognizant of the high fetal risk (32). Many of these women explained that they did not believe they were fertile, since they had not conceived during periods of months or years when they had not used contraceptive methods (33).

CURRENT TRENDS IN PREVENTING FETAL EXPOSURE TO TERATOGENS The advent of effective injectable hormonal contraceptives has made it possible to minimize the risk of an unplanned pregnancy during therapy with a known teratogen. This approach was first implemented in South America, where sexually active women with cutaneous leprosy were injected with medroxyprogesterone before receiving a prescription for thalidomide (34). Yet numerous new cases of thalidomide-associated embryopathy have been reported in the children of women who continued to take the drug after the period of contraceptive efficacy (three months) or who received the drug from their male partners (34). Because any new drug may be teratogenic, it is important to develop more effective methods to prevent fetal exposure. One such method may be the use of implantable hormonal compounds (e.g., levonorgestrel implants), which can provide long-term, reversible contraception for up to five years. Levonorgestrel implants have documented efficacy in young women in whom oral methods of contraception are likely to fail (35). Implants should be considered by sexually active women who are taking a teratogen medicinally (e.g., phenytoin or warfarin) or as part of a pattern of substance abuse (e.g., alcohol or

42

Koren et al.

cocaine). Furthermore, women taking teratogenic drugs who are not sexually active should be informed of the availability of effective postcoital contraceptives (35,36).

THE PROCESS OF ESTABLISHING RISK OR SAFETY OF DRUGS IN PREGNANCY Every year, many new drugs are approved and marketed. By this stage, several thousand people have usually participated in studies of the drugs, but the majority have been men. Since there are scarcely any data on fetal effects at the time of marketing, data from studies in animals provide the initial guidelines. The Value of Studies in Animals Typically, studies of reproductive toxicology in animals compare the outcome of pregnancy in groups of animals receiving a range of doses of the drug in question during the period of organogenesis with the outcome in untreated (control) animals. The occurrence of thalidomide-associated embryopathy led to the erroneous belief that human teratogenicity could not be predicted on the basis of studies in animals. However, every drug that has since been found to be teratogenic in humans has caused similar teratogenic effects in animals (Table 3), except misoprostol, which causes a morphologic pattern known as the Moebius sequence in humans. In at least one case, that of isotretinoin, the studies in animals probably prevented a disaster similar to that of thalidomide (29). However, there are drugs that have teratogenic effects in animals when administered in high doses that are not teratogenic in humans given clinically relevant doses. For example, high doses of glucocorticoids (19,70–75) or benzodiazepines (76,77) can cause oral clefts in animals, but clinically relevant doses in humans have no such effects (10–12,75). Similarly, salicylates (78–80) cause cardiac malformations in animals but not in humans (20,21). Such discrepancies have led to unwarranted anxiety on the part of women, their families, and physicians and may have contributed to unnecessary terminations of pregnancies (3). Although studies in animals may identify teratogenic effects, it can be difficult to extrapolate these effects to humans. Epidemiologic Studies In addition to studies in animals, a variety of other approaches are used to identify possible drug teratogenicity and to assess the relation between drug exposure and fetal outcome. The first accounts of adverse fetal outcomes after exposure to a marketed drug are usually published in the form of case reports. These reports can be either very useful or useless in establishing teratogenic risk on the basis of relatively simple statistical considerations. If the drug in question is taken by relatively small numbers of women (e.g., isotretinoin (30)) or causes a rare malformation (e.g., ear agenesis (81)), then a small number of cases can establish a strong association. Warfarin (9), diethylstilbestrol (82), and isotretinoin (83) were originally identified as human teratogens on the basis of case reports. If, on the other hand, the drug is taken by many pregnant women (e.g., Bendectin), a small number of case reports of abnormalities may simply reflect the spontaneous occurrence of malformations in the general population, which ranges from 1 to 5 percent, unless there is a characteristic pattern of malformations (as, for example, with alcohol or thalidomide). To

Drugs in Pregnancy

43

date, prenatal exposure to many of the known human teratogens has been associated with characteristic patterns of malformations, and this has become an important tenet in establishing teratogenicity. Epidemiologic studies are typically designed to determine whether mothers who took a specific drug during pregnancy have a larger number of malformed children than mothers who did not (cohort studies) or whether mothers of children with a specific malformation took the drug more often than mothers of children without the malformation (case– control studies). With the international development of teratology-information services (84), a new source of data for prospective observational research has emerged. Pregnant women taking prescription or over-the-counter drugs voluntarily call these centers for risk-assessment counseling, usually during the first trimester. Since the exposure data are recorded prospectively, the probability of recall bias is reduced, and follow-up of exposed pregnancies can extend well beyond parturition. Collaboration among these services can yield the large samples needed to study rare events more effectively (53,61,85). Drug manufacturers may perform postmarketing cohort studies of prospectively reported exposures. Such studies were useful in establishing the safety and risk of Bendectin, isotretinoin, fluoxetine, and acyclovir (30,86). Because most studies of teratogenic risk are limited in size, meta-analyses of studies of similar design are becoming more frequent. A detailed, stepwise methodologic approach to meta-analysis of teratologic studies has been described (24). The appropriate use of this approach depends to a large extent on establishing sound a priori criteria for methodologic quality and ensuring the inclusion of data from all available studies, in order to obviate any publication bias against negative results. Long-term studies are increasingly important, because it is becoming clear that the long-term effects of teratogenic drugs on neurobehavioral development can have a more devastating effect on children and their families than structural anomalies. To date, several drugs have been shown to affect brain development, including carbamazepine, isotretinoin, phenytoin, valproic acid, and warfarin (Table 1). Carbamazepine and valproic acid may cause cognitive brain dysfunction as part of the neural-tube defects they induce. Originally, isotretinoin was found to cause structural abnormalities that affected brain development, but recent studies have suggested that even phenotypically normal children may have abnormal neurodevelopment (87). Warfarin was initially associated with chondrodysplasia punctata and mental retardation and has subsequently been found to cause the DandyWalker brain malformation in an estimated 1 to 2 percent of exposed fetuses (6,69). Common Methodologic Issues No single approach can definitively establish the safety or risk of drugs, because of several underlying difficulties. Sample Size Most congenital malformations occur rarely, and many teratogens, even when known to be associated with an increased risk of a given malformation, do not affect the great majority of exposed fetuses. In fact, very few drugs increase the total malformation rate by a factor of more than two (isotretinoin and thalidomide are two such drugs). If, for example, the risk of major malformations in a given population is 3 percent, then at least 220 pregnancies with the specific exposure and a similar number of control pregnancies

44

Table 3

Teratogenic Effects of Drugs in Animals and Humans a

Drug Angiotensin-converting— enzyme inhibitors Carbamazepine Cocaine

Ethanol

Isotretinoin

Effects in animals

Effects in humans

Stillbirths and increased fetal loss in sheep and rabbits (37) Cleft palate, dilated cerebral ventricles, and growth retardation in mice (39) Dose-dependent decrease in uterine blood flow, fetal hypoxemia, hypertension, and tachycardia in sheep (41); reduced fetal weight, fetal edema, and increased resorption in rats and mice (42) Microcephaly, growth deficiency, and limb anomalies in dogs, chickens, and mice (44–48)

Prolonged renal failure and hypotension in the newborn, decreased skull ossification, hypocalvaria, and renal tubular dysgenesis (38) Neural-tube defects

CNS, head, limb, and cardiovascular defects in rats and rabbits (29)

Growth retardation involving weight, length, and head circumference (48); placental abruption (44,45) and uterine rupture

Koren et al.

Fetal alcohol syndrome: prenatal and postnatal growth deficiency, CNS anomalies (microcephaly, behavioral abnormalities, and mental retardation), characteristic pattern of facial features (short palpebral fissures, hypoplastic philtrum, and flattened maxilla), and major organ-system malformations (49); with age, facial features may become less distinctive, but short stature, microcephaly, and behavioral abnormalities persist (50) Retinoid embryopathy resulting in some or all of the following abnormalities (50): CNS defects (hydrocephalus, optic-nerve blindness, retinal defects, microphthalmia, posterior fossa defects, and cortical and cerebellar defects); craniofacial defects (microtia or anotia, low-set ears, hypertelorism, depressed nasal bridge, microcephaly, micrognathia, and agencies or stenosis of external ear canals); cardiovascular defects (transposition of great vessels, tetralogy of Falloe, and ventricular or atrial septal defects); thymic defects (ectopia and hypoplasia or aplasia); and miscellaneous defects (limb reduction, decreased muscle tone, spontaneous abortion, and behavioral abnormalities)

Phenytoin

Heart defects in rats (51) CNS abnormalities in rats (54); growth retardation, motor disturbances, microencephaly, and brain lesions in rhesus monkeys (55) Cleft palate, micromelia, renal defects, and hydrocephalus in rabbits, mice, and rats (57–59)

Thalidomide b

Limb-shortening defects in rabbits (most sensitive species) (62)

Valproic acid Warfarin b

Exencephaly in hamsters and mice (64,65) Maxillonasal hypoplasia and skeletal anomalies in rats (68)

a b

Ebstein’s anomaly and other heart defects (52,53) Fetal Minamata disease: diffuse neuronal disintegration with gliosis, cerebral palsy, microcephaly, strabismus, blindness, speech disorders, motor impairment, abnormal reflexes, and mental retardation (56) Fetal hydantoin syndrome (60); prenatal and postnatal growth deficiency, motor or mental deficiency, short nose with broad nasal bridge, microcephaly, hypertelorism, strabismus, epicanthus, wide fontanelles, lowset or abnormally formed ears, positional deformities of limbs, hypoplasia of nails and distal phalanges, hypospadias, hernia, webbed neck, low hairline, impaired neurodevelopment and low performance scores on tests of intelligence (61) Limb-shortening defects (63), loss of hearing, abducens paralysis, facial paralysis, anotia, microtia, renal malformations, congenital heart disease Neural-tube defects (66,67) Fetal warfarin syndrome: skeletal defects (nasal hypoplasia and stippled epiphyses), limb hypoplasia (particularly in distal digits), low birth weight (⬍10th percentile), hearing loss, and ophthalmic anomalies (69); CNS defects with exposure after first trimester; dorsal midline dysplasia (agencies of corpus callosum and Dandy–Walker malformations) or ventral midline dysplasia (optic atrophy) (6)

Drugs in Pregnancy

Lithium Methyl mercury

CNS denotes central nervous system. Initial studies in animals failed to show teratogenicity; hence, documentation in humans preceded that in animals.

45

46

Koren et al.

will be required to show a risk that is increased by a factor of 2.5, with a power of 80 percent. Effect of Maternal Diseases Apart from drug therapy, many medical conditions themselves increase fetal risks. For example, pregnant women with hypertension or cancer are more likely to have infants with intrauterine growth retardation, and pregnant women with epilepsy or diabetes mellitus are more likely to have infants with malformations (88). Therefore, any attempt to establish the role of fetal exposure to drugs must also address the contributing and confounding risk of the underlying maternal illness. Recall Bias in Retrospective Studies There is ample evidence of partial memory and bias in the way women recall the drugs they took during pregnancy. For example, women treated with a prescribed drug for a chronic illness tend to recall their treatment better than women who took an over-thecounter drug (89). Women who have given birth to malformed children may be more likely to remember the course of their pregnancies, in the effort to understand what went wrong, than women who have given birth to healthy children, thus giving rise to false positive associations. The initial suggestions that benzodiazepines, spermicides, and Bendectin, for example, were teratogenic were based on retrospective case-control studies subsequently refuted by other, larger studies (Table 2). With improved epidemiologic methods, the reliability of the case-control design has improved. For example, recruiting mothers of infants with a different major malformation as controls may eliminate or at least reduce the problem of differential maternal recall. In a recent study, this approach was used to document the effect of the mothers’ knowledge of the study hypothesis (that folic acid deficiency causes spina bifida) on the information they reported (90). Nonrandomized Observational Studies With prospective observational studies, the treatment decisions have not been made by the investigators collecting the data. As a result, the indications for treatment and concurrent exposures are not standardized. Therefore, in comparisons of treated and untreated pregnant women or pregnant women who received two different drugs, preexisting confounding factors are not randomly distributed between the two groups. For example, in comparing the outcome of pregnancy in women who received carbamazepine and women who received phenytoin, one must address the issue of whether the two groups of women had the same type and severity of seizure disorder. Observational studies of neurobehavioral development require longer follow-up than observational studies of other abnormalities, and interpretation of the results is often complicated by numerous confounding factors. Maternal and paternal IQ, socioeconomic status, and educational levels all affect cognitive development in children (91). Any attempt to address the developmental effects of drugs without controlling for these factors is likely to be futile. Voluntary Reporting The information received by drug manufacturers is often a mix of prospective and retrospective case reports. The quality of the information about exposure is usually poor, and outcome data are sparse because of high rates of loss to follow-up. Most important, women and health professionals who contact manufacturers are likely to report adverse fetal out-

Drugs in Pregnancy

47

comes, not uneventful ones. For example, the pivotal study that described retinoid embryopathy contained two parts: prospectively collected data from a study cohort, with a malformation rate of 36 percent, and data from voluntary retrospective reporting to the manufacturer, with a malformation rate of 80 percent (30). Meta-Analyses A common concern regarding the use of metaanalysis is the inevitable combination of data from studies that are not equivalent in terms of quality and methods. In addition, there is the concern that negative studies (i.e., those that do not reject the null hypothesis) are less likely to be published than positive studies and that an overall positive association may therefore merely reflect unbalanced reporting.

COUNSELING WOMEN ABOUT TERATOGENIC RISKS In one study, women exposed to nonteratogenic drugs who sought counseling estimated, on average, that they had a 25 percent risk of major malformations, which is in the range of the teratogenic risk associated with thalidomide (2). After counseling, this estimate was substantially reduced, thereby preventing numerous terminations of otherwise wanted pregnancies (2,3). The same women correctly estimated the risk of major malformations in the general population (5 percent), indicating that the high risk they assigned to their own pregnancies was not due to a misunderstanding of the concept of base-line risk. What are the sources of this misperception? Numerous lay publications misinform women by assigning risks to drugs not known to be teratogenic in humans (2,3). Women often report that their physicians have encouraged them to terminate otherwise wanted pregnancies just to be on the safe side, suggesting that many physicians are unfamiliar with the current literature on the safety of drugs taken during pregnancy. Physicians counseling women who are pregnant or are planning a pregnancy should make sure that they understand clearly the nature and magnitude of a risk associated with a drug. Women’s attitudes toward voluntary abortion differ. In addition, the same information about the nature and magnitude of a teratogenic risk may prompt different decisions by different women, according to the clinical situation and specific circumstances. For example, women with epilepsy that has been treated effectively with phenytoin since their childhood may be glad to hear that although the drug is teratogenic, the overall risk of malformations is not high (61). In contrast, women who have been treated with phenytoin for a single grand mal seizure and who have normal children born before phenytoin was prescribed may find it unacceptable to continue an unplanned pregnancy after learning about the higher-than-normal risk of adverse effects. During counseling, it is important to ensure that a woman understands the concept of base-line teratogenic risk and the fetal or perinatal risks, if any, associated with her medical condition. For example, a woman with manic depression treated with lithium in the first trimester will need to understand not only the slightly increased risk of fetal cardiac anomalies associated with the drug (less than 1 percent) (53) but also the increased genetic risk of manic depression in her child. The counselor should be sure to communicate the same information to the woman’s physician so that she does not receive conflicting advice. To receive up-to-date, evidence-based information on the safety and risk of drugs during pregnancy, physicians can consult a teratogen-information service. Table 4 lists

48 Table 4 Selected Teratogen-Information Services Canada Motherisk Program, Toronto (416) 813-6780 World Wide Web address: http:/ /www.motherisk.org United States Organization of Teratology Information Services (801) 328-2229 (for referral to nearest service) World Wide Web address: http:/ /orpheus.ucsd.edu/ctis/ California Teratogen Information Service (619) 543-2131 (800) 532-3749 (only in California) District of Columbia Reproductive Toxicology Center (202) 293-5137 Florida Teratogen Information Service (352) 392-3050 (800) 392-3050 (only in Florida) Illinois Teratogen Information Service (312) 908-7441 (800) 252-4847 (only in Illinois) Indiana Teratogen Information Service (317) 274-1071 Massachusetts Teratogen Information Service (781) 466-8474 (800) 322-5014 (only in Massachusetts) Nebraska Teratogen Project (402) 559-5071 New York Teratogen Information Service (716) 874-4747 (ext. 477) (800) 724-2454 (ext. 270) (only in New York) Texas Teratogen Information Service (800) 733-4727 Utah Pregnancy Riskline (801) 328-2229 Vermont Pregnancy Risk Information (802) 658-4310 (800) 531-9800 (only in Vermont)

Koren et al.

Drugs in Pregnancy

49

the World Wide Web addresses and telephone numbers of services in the United States and Canada.

THE FDA CLASSIFICATION OF TERATOGENICITY To guide physicians in the interpretation of the teratogenic risk associated with prescription drugs, the FDA has established a system that classifies drugs on the basis of data from humans and animals, ranging from class A drugs, which are designated as safe for use during pregnancy, to class X drugs, which are contraindicated during pregnancy because of proven teratogenicity. This system has resulted in ambiguous statements that may be difficult to interpret and use for counseling and that can cause anxiety among women. In addition, the classification is often not changed when new data become available. For example, until recently, combined oral contraceptives were classified as class X drugs. Yet metaanalyses revealed that these combinations of estrogen and progestin were not associated with an increased risk of major malformations, in general (15), or genitourinary malformations, in particular (16). Each year, thousands of women become pregnant while taking these contraceptive hormones because of noncompliance or less-than-perfect efficacy, and their fetuses are exposed to the drugs during embryogenesis. In a similar fashion, tricyclic antidepressant drugs are classified as D (‘‘positive evidence of human fetal risk’’), even though no such evidence exists and the available data suggest that these drugs are safe (85). The Teratology Society has proposed that the FDA abandon the current classification system in favor of more meaningful, evidence-based, narrative statements (92). At an FDA hearing held on September 15, 1997, this proposal received public support. Other countries (e.g., Sweden, Australia, the Netherlands, Switzerland, and Denmark) have different classification systems, although all are based on a hierarchy of estimated fetal risk.

DRUGS OF CHOICE IN PREGNANCY Many pregnant women require drug therapy because of pregnancy-induced conditions such as nausea and vomiting, chronic conditions diagnosed before pregnancy, or acute conditions (e.g., those that require surgical treatment with the use of anesthetic agents). Several principles should guide the selection of therapy during pregnancy. Since fetal safety is a major concern, effective drugs that have been in use for long periods are preferable to newer alternatives. Table 5 lists selected drugs considered to be safe on the basis of either single large cohort studies or metaanalyses of several studies. Newer drugs may be more specific or have fewer adverse effects in adults, but their safety for fetuses is less likely to be known. For example, although acetaminophen with or without codeine may not be effective in many patients with migraine, it is widely used during pregnancy. Other, more potent antimigraine drugs are either too new (e.g., sumatriptan) or have known reproductive risks (e.g., ergotamine alkaloids that cause uterine contraction). To minimize the fetal risk, drug doses at the lower end of the therapeutic range should be prescribed during pregnancy. However, because of increased body weight and more rapid clearance of many drugs (e.g., lithium, digoxin, and phenytoin) during late pregnancy, some women may need higher-than-normal doses (94). Pregnant women should be discouraged from taking over-the-counter drugs, and such drugs should not be taken without counseling, since many factors, including the stage

50

Table 5

Selected Drugs that Can Be Used Safely During Pregnancy, According to Condition a

Condition Acne Allergic rhinitis

Constipation

Cough Depression

Diabetes Headache Tension

Topical: erythromycin, clindamycin, benzoyl peroxide Topical: glucocorticoids, cromolyn, decongestants, xylometazoline, oxymetazoline, naphazoline, phenylephrine; systemic: diphenhydramine, dimenhydrinate, tripelennamine, astemizole Docusate sodium, calcium, glycerin, sorbitol, lactulose, mineral oil, magnesium hydroxide Diphenhydramine, codeine, dextromethorphan Tricyclic antidepressant drugs, fluoxetine

Alternative Drugs Systemic: erythromycin, topical tretinoin (vitamin A acid)

Isotretinoin as contraindicated

Bisacodyl, phenolphthalein

Lithium

Insulin (human)

Insulin (beef or pork)

Acetaminophen

Aspirin and nonsteroidal antiinflammatory drugs, benzodiazepines β-adrenergic-receptor antagonists and tricyclic antidepressant drugs (for prophylaxis)

Acetaminophen, codeine, dimenhydrinate

Comments

When lithium is used in first trimester, fetal echocardiography and ultrasonography are recommended because of small risk of cardiovascular defects Hypoglycemic drugs should be avoided Aspirin and nonsteroidal antiinflammatory drugs should be avoided in third trimester Limited experience with ergotamine has not revealed evidence of teratogenicity, but there is concern about potent vasoconstriction and uterine contraction

Koren et al.

Migraine

Drugs of Choice

Labetalol, methyldopa

β-adrenergic-receptor antagonists, prazosin, hydralazine

Hyperthyroidism

Propylthiouracil, methimazole

Mania (and bipolar affective disorder)

Lithium, chlorpromazine, haloperidol

β-adrenergic-receptor antagonists (for symptoms) For depressive episodes: tricyclic antidepressant drugs, fluoxetine, valproic acid

Nausea, vomiting, motion sickness

Diclectin (doxylamine plus pyridoxine)

Peptic ulcer disease

Antacids, magnesium hydroxide, aluminum hydroxide, calcium carbonate, ranitidine Topical: moisturizing creams or lotions, aluminum acetate, zinc oxide cream or ointment, calamine lotion, glucocorticoids; systemic: hydroxyzine, diphenhydramine, glucocorticoids, astemizole Heparin, antifibrinolytic drugs, streptokinase

Pruritus

Thrombophlebitis, deepvein thrombosis a

Angiotensin-converting-enzyme inhibitors should be avoided because of risk of severe neonatal renal insufficiency Surgery may be required; radioactive iodine should be avoided If lithium is used in first trimester, fetal echocardiography and ultrasonography are recommended because of small risk of cardiac anomalies; valproic acid may be given after neural-tube closure is complete

Drugs in Pregnancy

Hypertension

Chlorpromazine, metoclopramide (in third trimester), diphenhydramine, dimenhydrinate, meclizine, cyclizine Sucralfate, bismuth subsalicylate

Topical: local anesthetics

Streptokinase is associated with a risk of bleeding; warfarin should be avoided

Data are from Smith et al. (98).

51

52

Koren et al.

of pregnancy, can influence the risk to the fetus. For example, a nonsteroidal antiinflammatory drug may be taken safely for pain during the first trimester of pregnancy, but there is increasing evidence that some nonsteroidal antiinflammatory drugs constrict or even close the fetal ductus arteriosus during late pregnancy (95).

CONCLUSIONS In addition to the risk associated with fetal exposure to teratogenic drugs, there is a risk associated with misinformation about the teratogenicity of drugs, which can lead to unnecessary abortions or the avoidance of needed therapy. The medical community and drug manufacturers should make a concerted effort to protect women and their unborn babies from both risks.

ACKNOWLEDGMENTS Supported by grants from the Medical Research Council of Canada, the National Health and Welfare Research Program, the Medical Research Council—Pharmaceutical Manufacturers Association of Canada Program, Physicians Services, Novartis, Roche Canada, Duchesnay, and the Motherisk Research Fund.

REFERENCES 1. Better news on population. Lancet 1992; 339:1600. 2. Koren G, Bologa M, Long D, Feldman Y, Shear NH. Perception of teratogenic risk by pregnant women exposed to drugs and chemicals during the first trimester. Am J Obstet Gynecol 1989; 160:1190–1194. 3. Koren G, Pastuszak A. Prevention of unnecessary pregnancy termination by counseling women on drug, chemical, and radiation exposure during the first trimester. Teratology 1990; 41:657–661. 4. Newton ER, ed. Medical problems in pregnancy. Med Clin North Am 1989; 73:517–752. 5. Moore KL. The developing human: clinically oriented embryology. 4th ed. Philadelphia: W.B. Saunders, 1988:131. 6. Schardein JL. Chemically induced birth defects. 2nd ed. rev. New York: Marcel Dekker, 1993. 7. Lenz W, Knapp K. Die Thalidomid-Embryopathic. Dtsch Med Wochenschr 1962; 87:1232– 1242. 8. Briggs GG, Freeman RK, Yaffe SJ. Drugs in pregnancy and lactation. 4th ed. Baltimore: Williams & Wilkins, 1994. 9. Saxen I. Associations between oral clefts and drugs taken during pregnancy. Int J Epidemiol 1975; 4:37–44. 10. Rosenberg L, Mitchell AA, Parsells JL, Pashayan H, Luvik C, Shapiro S. Lack of relation of oral clefts to diazepam use during pregnancy. N Engl J Med 1983; 309:1282–1285. 11. Shiono PH, Mills JL. Oral clefts and diazepam use during pregnancy. N Engl J Med 1984; 311:919–920. 12. Czeizel A. Lack of evidence of teratogenicity of benzodiazepine drugs in Hungary. Reprod Toxicol 1987; 1:183–188. 13. Nora JJ, Nora AH, Blu J, et al. Exogenous progestogen and estrogen implicated in birth defects. JAMA 1978; 240:837–843.

Drugs in Pregnancy

53

14. Schardein JL. Congenital abnormalities and hormones during pregnancy: a clinical review. Teratology 1980; 22:251–270. 15. Bracken MB. Oral contraception and congenital malformations in offspring: a review and meta-analysis of the prospective studies. Obstet Gynecol 1990; 76:552–557. 16. Raman-Wilms L, Tseng AL, Wighardt S, Einarson TR, Koren G. Fetal genital effects of firsttrimester sex hormone exposure: a meta-analysis. Obstet Gynecol 1995; 85:141–149. 17. Jick H, Walker AM, Rothman KJ, et al. Vaginal spermicides and congenital disorders. JAMA 1981; 245:1329–1332. 18. Einarson TR, Koren G, Mattice, D, Schechter-Tsafriri O. Maternal spermicide use and adverse reproductive outcome: a meta-analysis. Am J Obstet Gynecol 1990; 162:655–660. 19. Walker BE. Induction of cleft palate in rats with antiinflammatory drugs. Teratology 1971; 4:39–42. 20. Werler MM, Mitchell AA, Shapiro S. The relation of aspirin use during the first trimester of pregnancy to congenital cardiac defects. N Engl J Med 1989; 321:1639–1642. 21. Slone D, Siskind V, Heinonen OP, Monson RR, Kaufman DW, Shapiro S. Aspirin and congenital malformations. Lancet 1976; 1:1373–1375. 22. Dickson JH. Congenital deformities associated with Bendectin. Can Med Assoc J 1977; 117: 721. 23. Donnai D, Harris R. Unusual fetal malformations after antiemetics in pregnancy. BMJ 1978; 1:691–692. 24. Einarson TR, Leeder JS, Koren G. A method for meta-analysis of epidemiological studies. Drug Intell Clin Pharm 1988; 22:813–824. 25. McKeigue PM, Lamm SH, Linn S, Kutcher JS. Bendectin and birth defects. I. A meta-analysis of the epidemiologic studies. Teratology 1994; 50:27–37. 26. Neutel CI, Johansen HL. Measuring drug effectiveness by default: the case of Bendectin. Can J Public Health 1995; 68:66–70. 27. Ornstein M, Einarson A, Koren G. Bendectin/Diclectin for morning sickness: a Canadian follow-up of an American tragedy. Reprod Toxicol 1995; 9:1–6. 28. Mazzota P, Magee L, Koren G. Therapeutic abortions due to morning sickness: unacceptable combination. Can Fam Physician 1997; 43:1055–1057. 29. Fantel AG, Shepard TH, Newell-Morris LL, Moffett BC. Teratogenic effects of retinoic acid in pigtail monkeys (Macaca nemestrina). I. General features. Teratology 1977; 15:65– 71. 30. Lammer EJ, Chen DT, Hoar RM, et al. Retinoic acid embryopathy. N Engl J Med 1985; 313: 837–841. 31. A nation of illiterates. US News and World Report. May 17, 1982:1. 32. Pastuszak AL, Koren G, Rieder MJ. Use of the Reinoid Pregnancy Prevention Program in Canada: patterns of contraception use in women treated with isotretinoin and etretinate. Reprod Toxicol 1994; 8:63–68. 33. Koren G. The children of neverland: the silent human disaster. Toronto: Kid In Us, 1997. 34. Jones GR. Thalidomide: 35 years on and still deforming. Lancet 1994; 343:1041. 35. Polaneczky M, Slap G, Forke C, Rappaport A, Sondheimer S. The use of levonorgestrel implants (Norplant) for contraception in adolescent mothers. N Engl J Med 1994; 331:1201– 1206. 36. Glasier A. Emergency postcoital contraception. N Engl J Med 1997; 337:1058–1064. 37. Pipkin FB, Turner SR, Symonds EM. Possible risk with captopril in pregnancy: some animal data. Lancet 1980; 1:1256. 38. Rosa FW, Bosco LA, Graham CF, et al. Neonatal anuria with maternal angiotensin-converting enzyme inhibition. Obstet Gynecol 1989; 74:371–374. 39. Sullivan FM, McElhatton PR. A comparison of the teratogenic activity of the antiepileptic drugs carbamazepine, clonazepam, ethosuximide, phenobarbital, phenytoin, and primidone in mice. Toxicol Appl Pharmacol 1977; 40:365–378.

54

Koren et al.

40. Rosa FW. Spina bifida in infants of women treated with carbamazepine during pregnancy. N Engl J Med 1991; 324:674–677. 41. Wood JR Jr, Plessinger MA, Clark KE. Effect of cocaine on uterine blood flow and fetal oxygenation. JAMA 1987; 257:957–961. 42. Fantel AG, Macphail BJ. The teratogenicity of cocaine. Teratology 1982; 26:17–19. 43. Weathers WT, Crane MM, Sauvain KJ, Blackhurst DW. Cocaine use in women from a defined population: prevalence at delivery and effects on growth in infants. Pediatrics 1993; 91:350– 354. 44. Lutiger B, Graham K, Einarson TR, Koren G. Relationship between gestational cocaine use and pregnancy outcome: a meta-analysis. Teratology 1991; 44:405–414. 45. Chasnoff IJ. Cocaine, pregnancy, and the growing child. Curr Probl Pediatr 1992; 22:302– 321. 46. Ellis FW, Pick JR. An animal model of the fetal alcohol syndrome in beagles. Alcohol Clin Exp Res 1980; 4:123–134. 47. Shoemaker WJ, Koda LY, Shoemaker CA, Bloom FE. Ethanol effects in chick embryos: cerebellar Purkinje neurons. Neurobehav Toxicol 1980; 2:239–242. 48. Sulik KK, Johnston MC, Webb MA. Fetal alcohol syndrome: embryogenesis in a mouse model. Science 1981; 214:936–938. 49. Clarren SK. Recognition of fetal alcohol syndrome. JAMA 1981; 245:2436–2439. 50. Streissguth AP, Aase JM, Clarren SK, Randels SP, LaDuc RA, Smith DF. Fetal alcohol syndrome in adolescents and adults. JAMA 1991; 265:1961–1967. 51. Wilby OK, Tesh SA, Ross FW, Tesh JM. Effects of lithium on development in vitro and in vivo in the rat. Teratology 1987; 35:69, abstract. 52. Nora JJ, Nora AH, Toews WH. Lithium, Ebstein’s anomaly, and other congenital heart defects. Lancet 1974; 2:594–595. 53. Jacobson SJ, Jones K, Johnson K, et al. Prospective multicentre study of pregnancy outcome after lithium exposure during first trimester. Lancet 1992; 339:530–533. 54. Tatetsu M. Experimental manifestation of ‘‘congenital Minamata disease.’’ Psychiatr Neurol Jpn 1968; 70:162. 55. Dougherty WJ, Coulston F, Golberg L. Toxicity of methylmercury in pregnant rhesus monkeys. Toxicol Appl Pharmacol 1974; 29:138, abstract. 56. Matsumoto H, Koya G, Takeuchi T. Fetal Minamata disease: a neuropathological study of two cases of intrauterine intoxication by a methyl mercury compound. J Neuropathol Exp Neurol 1965; 24:563–574. 57. McClain RM, Langhoff L. Teratogenicity of diphenylhydantoin in New Zealand white rabbits. Toxicol Appl Pharmacol 1979; 48:A32, abstract. 58. Finnell RH. Phenytoin-induced teratogenesis: a mouse model. Science 1981; 211:483–484. 59. Harbison RD. Studies on the mechanism of teratogenic action and neonatal pharmacology of diphenylhydantoin. (Ph.D. thesis. Iowa City: State University of Iowa, 1969.) 60. Hanson JW, Smith DW. The fetal hydantoin syndrome. J Pediatr 1975; 87:285–290. 61. Scolnik D, Nulman I, Rovet J, et al. Neurodevelopment of children exposed in utero to phenytoin and carbamazepine monotherapy. JAMA 1994; 271:767–770. 62. Fratta ID, Sigg EB, Maiorana K. Teratogenic effects of thalidomide in rabbits, rats, hamsters, and mice. Toxicol Appl Pharmacol 1965; 7:268–286. 63. McBride WG. Thalidomide and congenital abnormalities. Lancet 1961; 2:1358. 64. Nau H. Species differences in pharmacokinetics and drug teratogenesis. Environ Health Perspect 1986; 70:113–129. 65. Finnel RH, Bennett GD, Karras SB, Mohl VK. Common hierarchies of susceptibility to the induction of neural tube defects in mouse embryos by valproic acid and its 4-propyl-4-pentenoic acid metabolite. Teratology 1988; 38:313–320. 66. Jager-Roman E, Deichi A, Jakob S, et al. Fetal growth, major malformations, and minor anomalies in infants born to women receiving valproic acid. J Pediatr 1986; 108:997–1004.

Drugs in Pregnancy

55

67. Lammer EJ, Sever LE, Oakley GP Jr. Teratogen update: valproic acid. Teratology 1987; 35: 465–473. 68. Howe AM, Webster WS. The warfarin embryopathy: a rat model showing maxillonasal hypoplasia and other skeletal disturbances. Teratology 1992; 46:379–390. 69. Hall JG, Pauli RM, Wilson KM. Maternal and fetal sequelae of anticoagulation during pregnancy. Am J Med 1980; 68:122–140. 70. Baxter H, Fraser FC. The production of congenital defects in the offspring of female mice treated with cortisone. McGill Med J 1950; 19:245–249. 71. Fainstat T. Cortisone-induced congenital cleft palate in rabbits. Endocrinology 1954; 55:502– 508. 72. Buresh JJ, Urban TJ. The teratogenic effect of the steroid nucleus in the rat. J Dent Res 1970; 43:548–554. 73. Wilson JG, Fradkin R, Schumacher HJ. Influence of drug pretreatment on the effectiveness of known teratogenic agents. Teratology 1970; 3:210–211, abstract. 74. Pinsky L, DiGeorge AM. Cleft palate in the mouse: a teratogenic index of glucocorticoid potency. Science 1965; 147:402–403. 75. Fraser FC, Sajoo A. Teratogenic potential of corticosteroids in humans. Teratology 1995; 51: 45–46. 76. Shepard TH. Catalog of teratogenic agents. 7th ed. Baltimore: Johns, Hopkins University Press, 1994. 77. Miller RP, Becker BA. Teratogenicity of oral diazepam and diphenylhydantoin in mice. Toxicol Appl Pharmacol 1975; 32:53–61. 78. Klein KL, Scott WJ, Wilson JG. Aspirin-induced teratogenesis: a unique pattern of cell death and subsequent polydactyly in the rat. J Exp Zool 1981; 216:107–112. 79. Wilson JG, Ritter EJ, Scott WJ, Fradkin R. Comparative distribution and embryotoxicity of acetylsalicylic acid in pregnant rats and rhesus monkeys. Toxicol Appl Pharmacol 1977; 41: 67–78. 80. Beall JR, Klein MF. Enhancement of aspirin-induced teratogenicity by food restriction in rats. Toxicol Appl Pharmacol 1977; 39:489–495. 81. Mastroiacovo P, Corchia C, Botto LD, Lanni R, Zampino G, Fusco D. Epidemiology and genetics of microtia-anotia: a registry based study on over one million births. J Med Genet 1995; 52:453–457. 82. Herbst AL, Scully RE. Adenocarcinoma of the vagina in adolescence: a report of 7 cases including 6 clear-cell carcinomas. Cancer 1970; 25:745–757. 83. Rosa FW. Teratogenicity of isotretinoin. Lancet 1983; 2:513. 84. Koren G, Pastuszak A. Teratogen information services. In: Koren G, ed. Maternal-fetal toxicology: a clinician’s guide. 2nd ed. rev. New York: Marcel Dekker, 1994:683–705. 85. Pastuszak A, Schick-Boschetto B, Zuber C, et al. Pregnancy outcome following first-trimester exposure to fluoxetine. JAMA 1993; 269:2246–2248. 86. Goldstein DJ, Corbin LA, Sundell KL. Effects of first-trimester fluoxetine exposure on the newborn. Obstet Gynecol 1997; 89:713–718. 87. Adams J. Neural and behavioral pathology following prenatal exposure to retinoids. In: Koren G, ed. Retinoids in clinical practice: the risk-benefit ratio. New York: Marcel Dekker, 1993: 111–128. 88. Gonen R, Shilalukey K, Magee L, Koren G, Shime J. Maternal disorders leading to increased reproductive risk. In: Koren G, ed. Maternal-fetal toxicology: a clinician’s guide. 2nd ed. rev. New York: Marcel Dekker, 1994:641–682. 89. Feldman Y, Koren G, Mattice K, Shear H, Pellegrini E, MacLeod SM. Determinants of recall and recall bias in studying drug and chemical exposure in pregnancy. Teratology 1989; 40: 37–45. 90. Werler MM, Shapiro S, Mitchell AA. Periconceptional folic acid exposure and risk of occurrent neural tube defects. JAMA 1993; 269:1257–1261.

56

Koren et al.

91. McCall RB. The development of intellectual functioning in infancy and the prediction of later I.Q. In: Osofsky JD, ed. Handbook of infant development. New York: John Wiley, 1979:707– 741. 92. Teratology Society Public Affairs Committee. FDA classification of drugs for teratogenic risk. Teratology 1994; 49:446–447. 93. Smith J, Taddio A, Koren G. Drugs of choice for pregnant women. In: Koren G, ed. Maternalfetal toxicology: a clinician’s guide. 2nd ed. rev. New York: Marcel Dekker, 1994:115–128. 94. Loebstein R, Lalkin A, Koren G. Pregnancy induced pharmacokinetic changes and their clinical relevance. Clin Pharmacokinet 1997; 33:328–343. 95. Theis JGW. Acetylsalicylic acid (ASA) and nonsteroidal anti-inflammatory drugs (NSAIDs) during pregnancy: are they safe? Can Fam Physician 1996; 42:2347–2349.

4 Teratogenic Drugs and Chemicals in Humans Irena Nulman, Gordana Atanackovic, and Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

Clinical Case A 28-year-old epileptic woman, treated for 6 years with valproic acid, reports amenorrhea of 12 weeks. Pregnancy test is positive, and the patient, who had tried unsuccessfully to conceive for several years, is very happy. What would you advise her? INTRODUCTION Since the thalidomide disaster, drugs and chemicals have been scrutinized carefully for their potential human teratogenicity. However, despite valiant efforts in this direction, several objective limitations hinder our ability to detect human teratogens. 1. Most birth defects occur rarely; therefore, even an increased risk posed by a teratogen may not be easily identified. While thalidomide caused more than 20% of major malformations following first-trimester exposure, most suspected teratogens increase the baseline risk of major malformations very slightly (1–3%), even when they significant increasing the risk of a specific pattern. For example, valproic acid increases by 200fold the risk for neural tube defects (NTDs), yet its impact on the overall risk for major malformations is less than 0.5% (1). Consequently, a woman exposed to valproic acid may still have better than a 95% chance of having a healthy baby. As a result, it may be necessary to study a large number of infants exposed in utero to a certain drug to prove or disprove its teratogenic potential. Most studies are limited in their statistical power because their numbers are not large enough. 2. For obvious reasons, pharmaceutical manufacturers warn the public not to use drugs in pregnancy owing to a lack of information about their safety. Consequently, the accumulation of data on a specific drug is often sketchy and uncontrolled. While most manufacturers try to record and follow up voluntary reports of exposures in pregnancy, it has to be recognized that such data are incomplete and may be biased. For example, it is conceivable that disproportionately high numbers of families having malformed children will report their drug exposure to the manufacturer or to the regulatory agencies, whereas families with healthy babies are less likely to do so. This tendency was recently documented in a large study on the teratogenicity of retinoic acid. In the prospective section 57

58

Nulman et al.

of that cohort, 38% of the infants were malformed, whereas 80% of those reporting retrospectively cited major malformations. This means that a large number of families having normal children after first-trimester exposure to retinoic acid did not report voluntarily to either the manufacturer or the Centers for Disease Control (CDC) (2,3). 3. As part of the regulatory process, the teratogenic potential of drugs has to be tested in animals. The failure of animal models to detect the teratogenicity of thalidomide before the human disaster occurred has resulted in a growing feeling that we cannot extrapolate from animal studies to humans. Differences in pharmacokinetics, metabolism, embryology, target-organ sensitivity, and other factors may account for such discrepancies. Yet almost all known human teratogens have been shown to cause similar effects in animals, warfarin (Coumadin) being the exception (4). Moreover, in the case of retinoic acid (Accutane), the most potent human teratogen currently available, animal data clearly prevented a postmarketing disaster similar to that of thalidomide (5). CASE REPORTS Cases associating in utero exposure to a certain drug or chemical with an adverse outcome may be either most helpful or useless, depending on the following statistical considerations. In the case of a drug that is rarely used in pregnancy, a small number of cases showing the same pattern of malformations may be most indicative, since these cases may already exceed manyfold the baseline risk for the occurrence of such malformations. For example, if a new drug has been reported to cause 10 cases of cleft palate out of the total 100 known cases of first-trimester exposure, then it has a 10% risk of causing cleft palate. This calculated risk exceeds by 100-fold the known risk of cleft palate, which is 0.1%. Based on this approach, several human teratogens, including warfarin and retinoic acid, were incriminated long before prospective studies confirmed these associations. At the other extreme, it will be impossible to prove teratogenicity based on case reports when the drug is commonly used (e.g., salicylates) and the malformation is not rare. Salicylates are consumed by thousands of pregnant women every year; therefore, based on statistical chance only, one would expect to find within this group offspring with any described malformation. Bendectin (the combination of pyridoxin and doxylamine) was for decades one of the most widely used antiemetic drugs to cope with pregnancy-associated morning sickness. The drug was wrongly incriminated as causing congenital malformations based on case reports: hundreds of thousands of pregnant women were exposed to this drug, 1– 3% of their offspring would have had major malformations just because of their baseline risk. Subsequently, controlled studies (both case-control and prospective) failed to confirm an association between Bendectin and human teratogenicity; however, the American manufacturer withdrew the drug from the market owing to excessive insurance costs (see Chap. 32). In Canada, this antiemetic is available under the trade name Diclectin, and recently the Canadian Health Protection Branch has specifically labeled it as an antiemetic appropriate for use in pregnancy. EPIDEMIOLOGICAL STUDIES Several approaches are commonly employed in studying the potential reproductive effects of drugs and chemicals.

Teratogenic Drugs and Chemicals in Humans

59

The retrospective study tries to identify women exposed in pregnancy to the drug in question and to evaluate the outcome of the exposed offspring. A variety of methodological problems may complicate the interpretation of such data. It may not be possible to identify and assess all cases, and it is conceivable that cases with an adverse outcome will be overrepresented. For example, the Danish lithium registry is a voluntary reporting system of exposure to the drug in pregnancy. Of its 300 cases reported by 1993, about 10% represented cardiac malformation; however, it is probable that parents of healthy babies born after such exposures had less motivation to report to the registry than those with an adverse outcome (6). Another major disadvantage of such cohorts, owing to the rare occurrence of most congenital anomalies, is the need for very large numbers of individuals exposed to the drug. To overcome this shortcoming, the case-control study focuses on offspring with a specific malformation and tries to assess maternal exposure to the drug in question. The percentage of maternal use is then compared to that of a group of mothers of infants not having the tested malformation. A major problem with all retrospective studies, both cohorts and case-control, is the need to rely on maternal recall of drug exposure (i.e., time and dose). In a recent study, mothers we had interviewed first at the time of their exposure in pregnancy tended not to remember significant parts of this information when questioned again after giving birth (7). Mitchell has significantly improved the reliability of casecontrol studies by developing questionnaires that reduce the effect of maternal recall characteristic of the previously used open-ended questionnaire (8). In prospective studies, the information about exposure and other possible pregnancy risk factors is collected at the time of exposure or soon after it. Although this is likely to be the most accurate approach to assess potential teratogenicity, such studies are lengthy and costly because very large numbers of test subjects are needed to overcome the rareness of most congenital malformations. Whereas full discussion of methodological problems associated with each approach is beyond the scope of this chapter, it is clear that proving the teratogenicity of a specific drug in humans may be a complex process, demanding a high degree of scrutiny. In many cases evidence is accumulated through several different approaches (e.g., case reports, case control, and prospective studies). The combination of increased awareness to human teratogenicity with the abovementioned difficulties in differentiating normal background from slightly increased rates of malformations has led to unjustified incrimination of useful medications that were later found out to be nonteratogenic. These include oral contraceptive hormones, diazepam, Bendectin, and spermicides. Typically, a devastating potential, as reflected above in the case of Bendectin, may result from wrong incrimination of a nonteratogen, as many women may terminate an otherwise wanted pregnancy. Presently, it is felt that several criteria must be met before an agent is incriminated as a human teratogen (9): An abrupt increase in the frequency of a particular defect or association of defects (syndrome) Coincidence of this increase with a known environmental change, such as widespread use of a new drug or sudden exposure to a chemical Known exposure to the environmental change early in pregnancy yielding characteristically defective infants Absence of other factors common to all pregnancies yielding infants with the characteristic defect(s)

Table 1

Drugs and Chemicals Proven to Be Teratogenic in Humans

Alcohol

Relative risk for teratogenicity

Clinical intervention

Fetal alcohol syndrome (FAS): evidence of unique pattern of facial anomalies such as short palpebral fissures, flat midface, long and narrow upper lip, flattened philtrum: evidence of antenatal and/or postnatal growth retardation—low birth weight for gestation age, decelerated weight not due to nutrition Alcohol-related neurodevelopmental disorder (ARND) such as decreased cranial size at birth, brain abnormalities such as microcephaly, partial or complete agenesis of the corpus callosum, cerebellar hypoplasia, neurological hard or soft signs such as impaired fine motor skills, neurosurgery hearing loss. Alcohol-related birth defects (ARBD) such as cardiac, renal, skeletal, ocular, auditory and other)

In alcoholic women consuming above 2 g/kg/day of ethanol over the first trimester: two- to threefold higher risk for congenital malformations (about 10%)

Growth retardation, cleft palate, microphthalmos, hypoplastic ovaries, cloudy corneas, agenesis of kidney, malformations of digits, cardiac defects, multiple other anomalies

Based on case reports, between 10 and 50% of malformations were due to different drugs. It is possible that adverse outcome was overrepresented.

Calculate accurate dose of alcohol. Prospective: Counsel patient to discontinue exposure. If patient is alcoholic, refer her to an addiction center. Counsel patient to use effective contraception. During pregnancy: Alleviate fears in mild or occasional drinkers who may terminate pregnancy based on unrealistic perception of risk. Level 2 ultrasound should be performed to rule out visible malformation. When the diagnosis of heavy drinking in early pregnancy has been made, the attendant fetal risk should be discussed; in some cases termination of pregnancy may be chosen by the patient. Counsel patient to discontinue drinking if the pregnancy is continuing. After delivery: Aggressive follow-up; early diagnoses of alcohol-related abnormalities to prevent secondary disabilities. Level 2 ultrasound to rule out visible malformations Supplement folic acid to women receiving antifolates (e.g., methotrexate)

Ref. 10,11

12,30

4

Nulman et al.

Alkylating agents (busulfan, chlorambucil, cyclophosphamide, mechlorethamine)

Fetal adverse effects

60

Drug/chemical

Benzodiazepines

Hydrocephalus, meningocephalocele, anencephaly, malformed skull, cerebral hypoplasia, growth retardation, eye and ear malformations, malformed nose and cleft palate, malformed extremities and fingers Aminopterin syndrome: Cranial dysostosis, hydrocephalus, hypertelorism, anomalies of external ear, micrognathia, posterior cleft palate Increased risk for major malformations or oral cleft alone

Based on case reports 7–75% of patients were malformed. It is possible that adverse outcome was overrepresented.

Level 2 ultrasound to rule out visible malformations

4

Meta-analysis of case-control studies showed an association between major malformation and/or oral cleft alone (odds ratios 3.01; 95% CI, 1.32–6.84) NTDs estimated at 1% with carbamazepine

Level 2 ultrasound to rule out visible malformations.

31,32

Periconceptional folate; maternal and/or amniotic α-fetoprotein; ultrasound to rule out NTD. Measure maternal carboxyhemoglobin levels. Treat with 100% oxygen for 5 h after maternal carboxyhemoglobin returns to normal because fetal equilibration takes longer. If hyperbaric chamber available, should be used, as elimination t1/2 of CO is more rapid. Fetal monitoring by an obstetrician; sonographic follow-up. Level 2 ultrasound for measurement of fetal head size.

14,41

Increased risk for neural tube defects (NTDs)

Carbon monoxide

Cerebral atrophy, mental retardation, microcephaly, convulsions, spastic disorders, intrauterine or postnatal death

Based on case reports, when the mother is severely poisoned, there is a high risk for neurological sequelae; there is no risk in mild, accidental exposures

Cocaine

Small head circumference, low birth rate, intracranial bleeding

A dose-response relationship was found between cocaine use and perinatal outcome.

15,16

40

61

Carbamazepine

Teratogenic Drugs and Chemicals in Humans

Antimetabolic agents (aminopterin azauridine, cytarabine, 5-FU, 6-MP, methotrexate)

Table 1

Continued Fetal adverse effects Cleft palate Reduction in birth weight and head circumference, dose-response relationship

Warfarin

Fetal warfarin syndrome: nasal hypoplasia, chondrodysplasia punctata, brachydactyly, skull defects, abnormal ears, malformed eyes, CNS malformations, microcephaly, hydrocephalus, skeletal deformities, mental retardation, optic atrophy, spasticity, Dandy Walker malformations Female offspring: clear cell vaginal or cervical adenocarcinoma in young female adults exposed in utero (before 18th week): irregular menses (oligomenorrhea), reduced pregnancy rates, increased rate of preterm deliveries, increased perinatal mortality and spontaneous abortion Male offspring: cysts of epididymis, cryptorchidism, hypogonadism, diminished spermatogenesis Lower scores in developmental tests

Diethylstilbestrol (DES)

Lead

Clinical intervention

Odds Ratio (OR)-2.5(CI 1.5–4.4) metaanalysis. Antenatal exposure at 24 weeks or later causes up to a 9% reduction (p ⬍ 0.014) in birth weight and up to a 4% reduction (p ⬍ 0.0024) in head circumference. 16% of exposed fetuses have malformations; another 3% have hemorrhages; 8% are stillborn.

Level 2 ultrasound to rule out visible malformations.

35,36

Prospective: switch to heparin for the first trimester. Deliver by cesarean section. Women should be followed up in a high-risk perinatal unit.

17

Exposure before 18 weeks of gestation: ⱕ1.4\1000 of exposed female with carcinoma. Congenital morphological changes in vaginal epithelium in 39% of exposures

Diagnosis: direct observation of mucosa and Shiller’s test. Treatment: mechanical excitement or destruction in relatively confined area. Surgery and radiotherapy for diffused tumor.

18

Higher risk when maternal lead is above 10 µg/dL.

Maternal lead levels ⬎ 10 µg/ dL: investigate for possible source of contamination. Levels ⬎ 25 µg/dL: consider chelation.

19

Ref.

Nulman et al.

Corticosteroids

Relative risk for teratogenicity

62

Drug/chemical

Possibly higher risk for Ebstein’s anomaly; no detectable higher risk for other malformations

Methyl mercury, mercuric sulfide

Microcephaly, eye malformations, cerebral palsy, mental retardation, malocclusion of teeth

Women of affected babies consumed 9–27 ppm mercury; greater risk when ingested at 6–8 gestational months. Relative risk was not elucidated, but 13 of 220 babies born in Minamata, Japan, at the time of contamination had severe disease.

Misoprostol

Spontaneous abortions, fetal death. Mo¨bius syndrome (paralysis of sixth and seventh cranial nerves) Increased risk for major malformations without a single pattern, increased risk for miscarriage

RR 3.15; CI 1.20–8.27

Organic solvents

PCBs

6

20,21

37,38

OR 25.7: CI 8.30–89.4

13% (1.8–99.5) OR–1.64 (CI 1.16–2.30) for major malformations OR–1.25 (0.99–1.58) for spontaneous abortions 4% (6 of 159) to 20% (8 of 39)

Minimize exposure, avoid toxicity by protection and ventilation.

33,34

These figures, which are from cases poisoned by high consumption of PCB-contaminated rice oil, cannot be extrapolated to cases in which maternal poisoning has not been verified. Women working near PCBs (e.g., hydroelectric facilities) should use effective protection.

22

63

Stillbirth Signs at birth: white eye discharge, 30% (32 of 108); teeth present, 8.7% (11 of 127); irritated/swollen gums, 11% (11 of 99); hyperpigmentation (‘‘cola’’ staining), 42.5% (54 of 127); deformed/small nails, 24.6% (30 of 122); acne, 12.8% (16 of 125).

Women who need lithium should continue therapy, with sonographic follow-up. Patients may need higher doses because of increased clearance rate. Good correlation between mercury concentrations in maternal hair follicles and neurological outcome of the fetus. Hair mercury content above 50 ppm was used successfully as a cut point for termination. In acute poisoning, the fetus is 4– 10 times more sensitive than the adult to methylmercury toxicity. Increase the awareness of teratogenicity of the drug.

Teratogenic Drugs and Chemicals in Humans

Lithium carbonate

64

Table 1

Continued Fetal adverse effects

Penicillamine Phenytoin

Subsequent history: bronchitis or pneumonia, 27.2% (30 of 124); chipped or broken teeth, 35.5% (38 of 107); hair loss, 12.2% (14 of 115); acne scars, 9.6% (11 of 115); generalized itching, 27.8% (32 of 1150). Developmental: children do not meet milestones; have lower scores than unexposed controls; show evidence of CNS damage. Skin hyperlastosis Fetal hydantoin syndrome: low nasal bridge, inner epicanthal folds, ptosis, strabismus, hypertelorism, low-set or abnormal ears, wide mouth, large fontanel, anomalies and hypoplasia of distal phalanges and nails, skeletal abnormalities, microcephaly and mental retardation, growth deficiency, neuroblastoma, cardiac defects, cleft palate and/or lip

Relative risk for teratogenicity

Few case reports; risk unknown 5–10% of typical syndrome; about 30% of partial picture. Relative risk of 7 for offspring IQ ⱕ84 (see Chap. 4)

Clinical intervention

Neurologist should consider changing to other medications. Keep phenytoin concentrations at lower effective levels. Level 2 ultrasound to rule out visible malformations. Vitamin K to neonate. Epilepsy itself increases teratogenic risk.

Ref.

23 24,25

Nulman et al.

Drug/chemical

Trimethadione

Thalidomide

Tetracycline

Valproic acid

Spontaneous abortions; deformities of cranium, ears, face, heart, limbs, liver; hydrocephalus, microcephalus, heart defects Cognitive defects even without dysmorphology Fetal trimethadione syndrome: intrauterine growth retardation, cardiac anomalies, microcephaly, cleft palate and lip, abnormal ears, dysmorphic face, mental retardation, tracheoesophageal fistula, postnatal death Limb phocomelia, amelia, hypoplasia, congenital heart defects, renal malformations, cryptorchidism, abducens paralysis, deafness, microtia, anotia Yellow, gray-brown, or brown staining of deciduous teeth, destruction of enamel

Lumbosacral spina bifida with meningomyelocele; CNS defects, microcephaly, cardiac defects, cognitive impairment

For isoretinoin: 38% risk: 80% of malformations are CNS.

Based on case reports: 83% risk, 32% infantile or neonatal death.

There is a risk of about 20% risk when exposure to drug occurs in days 34–50 of gestation.

From 4 months of gestation and on, teratogenicity occurs in 50% of fetuses exposed to tetracycline and in 12.5% to oxytetracycline. There is a 1–2% risk of neural tube defects.

Treated women should have an effective method of contraception. Pregnancy termination. If diagnosed too late, sonographic follow-up to rule out confirmed malformations. No need for this antiepileptic to date.

2

Thalidomide is an effective drug for some forms of leprosy. Treated women should have an effective mode of contraception. If there is exposure between 14 and 16 weeks of gestation, there is no known risk.

27

Level 2 ultrasound and maternal α-fetoproteins or amniocentesis to rule out neural tube defects. Epilepsy itself increases teratogenic risk.

26

Teratogenic Drugs and Chemicals in Humans

Systemic retinoids (isotretinoin, etretinate)

28

1,39,41

65

66

Nulman et al.

COUNSELING WOMEN ABOUT KNOWN TERATOGENS Table 1 presents details of drugs and chemicals known to be teratogenic in humans, with a major reference for each. When a clinician confronts an exposure of a pregnant patient to a known teratogen, it is important to convey the available information to the family in a way that will prevent both understatements and ambiguity. An accurate estimate of the risk for an adverse outcome should be provided, because pregnant women tend to have an unrealistically high perception teratogenic risk even when exposed to nonteratogens (see Chap. 29). In the Motherisk Program, we find that the same estimated risk may be unacceptably high for some families and reasonable for others. For example, epileptic women who are well controlled with phenytoin and have failed to have their epilepsy controlled with other anticonvulsants are often reluctant to change their medication in pregnancy. Conversely, we recently consulted the mother of three healthy children who was treated briefly with phenytoin following a single seizure. When it became apparent that she was pregnant again, an electroencephalogram was taken. The results were normal, and it was planned to discontinue the drug. For this patient, the teratogenic risk of phenytoin was perceived as unacceptable. Not included in Table 1 are scores of drugs that cause direct fetal toxicity consistent with their pharmacological effects. These are detailed in Chapter 13. Proven human teratogens are by no means a homogeneous group of compounds; however, they can be divided by several criteria into subgroups. Obsolete Drugs Diethylstilbestrol and trimethadione are currently unlikely to create a problem in pregnancy because they are not used clinically. Thalidomide, which was banned after the disaster three decades ago, is an important drug for some forms of leprosy. It is presently used in South America for leprosy in women who receive injectable forms of contraceptive hormone. However, there have recently been reports of a new wave of malformed children due to inappropriate use. Retinoic acid, which bears a rate of teratogenicity similar to that of thalidomide, is widely used, mostly for treatment of cystic acne in adolescents and young adults, who are the most likely group to fail contraceptives. In fact, the U.S. Food and Drug Administration (FDA) is reevaluating conflicting reports on the number of pregnancies and birth defects associated with retinoic acid in order to decide the future of this drug in the American market. Existence of Alternative Therapy Several teratogenic drugs may have value as alternative therapies in pregnancy; however, each case is characterized by unique problems. Although there are alternatives to lithium carbonate (e.g., tricyclics) for manic-depressive disorders, some patients may not respond favorably to replacement drugs. Moreover, recent evidence suggests lithium to be safe during pregnancy (6). The same argument is valid for phenytoin and valproic acid, and in each case the physician caring for the woman planning pregnancy should evaluate other alternatives. Heparin, which does not cross the placenta, may substitute for warfarin during the first trimester; the former must be injected, however, and compliance may become a major problem.

Teratogenic Drugs and Chemicals in Humans

67

From a pharmacological standpoint, in very few instances is there no alternative drug for human teratogens: retinoic acid, however, appears to be very efficacious in complicated types of acne, and no other compound shares the same mechanisms of action. This is a strong argument in favor of not removing the drug from the market despite its known teratogenic risk. Thousands of patients would thus be deprived of an irreplaceable therapy. Clearly, this drug is completely contraindicated in pregnancy. Alkylating agents and antimetabolites (azathioprine, chlorambucil, etc.) represent a specific therapy that may need to be continued uninterrupted. The main teratogenic effects of these drugs are associated with first-trimester exposure; current analysis done by us in Toronto reveals that in most cases, when cancer is diagnosed early in pregnancy, the women choose to terminate the pregnancy. However, to date there is increasing use of some of these agents in collagen diseases and nephritis as well as after organ transplants; it is likely that the number of women seeking prospective advice on these drugs will increase. Magnitude of the Public Health Issue Alcohol is undoubtedly the most common human teratogen. Because 10 million Americans are alcoholics, large numbers of fetuses are exposed to the amount of alcohol associated with fetal alcohol syndrome (FAS). It has been estimated that one baby in every 2500 live births has FAS, which means 1600 new cases in the United States per year (10). The combined rate of FAS and alcohol-related neurodevelopmental disorder is estimated to be at least 9.1 per 1000, nearly one in every 100 live births (25). Clearly the number of consumers of this teratogen is several orders of magnitude larger than that for any other teratogenic compound. With increasing public awareness of the adverse fetal outcome associated with alcohol, many women and families fear the potential adverse effects of alcohol consumed before conception was realized, even when much smaller amounts than those associated with FAS are involved. Although there is some preliminary evidence of a dose-response relationship to alcohol teratogenicity in humans (11), even two drinks a day during embryogenesis has not been associated with increased morphological or developmental risks (12). The problem of verifying the degree of drinking is major. Recently, for example, we documented that women who have had an adverse outcome of pregnancy tend to decrease the amount of alcohol reported postnatally compared to their initial report during pregnancy (7). The commonly used anticonvulsants are established human teratogens. Congenital heart defects, cleft lip and/or palate, genitourinary defects, and cognitive impairment were associated with use of anticonvulsant medication (41). Interaction of valproic acid and carbamazepine with folic acid may lead to neural tube defects (2% and 1% respectively). Valproic acid was also found to be associated with cognitive and behavioral impairment in school-age children (39). Phenytoin is undoubtedly another drug that, through its common use, creates a public health issue. It has been estimated by Hanson et al. (22) that between 5 and 10% of fetuses exposed in utero to the drug will exhibit the full picture of fetal hydantoin syndrome (FHS). Since 0.5% of pregnant women are epileptic and about half of them are treated with phenytoin, the rate of FHS should be somewhere between 0.019 and 0.025% of births (23). This means that with an annual birth rate of 4 million in the United States, between 500 and 100 newborns every year suffer from this serious syndrome. The risk for major malformations is increasing with the number of anticonvulsive medications used. Lindhout et al. reported a 5% risk of birth defects in

68

Nulman et al.

children of women with epilepsy who took two drugs concomitantly, 10% in those who took three drugs, and more than 20% when four drugs were used (42). However, proper seizure control is the primary goal in treating women with epilepsy. Patients should understand the risks associated with uncontrolled seizures as well as the potential for teratogenicity of the anticonvulsive medications. Retinoic acid, evolving as a commonly used drug for acne in young adults, has the potential of becoming a similar public health issue if not a worse one. However, no peerreviewed data have been published on the number of pregnancies occurring while women are being treated with Accutane. Unlike phenytoin, which may be essential during pregnancy, the use of retinoic acid is absolutely contraindicated in pregnancy, and treatment of acne can be postponed without risk to the mother. Benzodiazepines are commonly used for anxiety, insomnia, drug withdrawal, and epilepsy. Bergman et al. found that 2% of pregnant women in the United States receive benzodiazepines during pregnancy (26). Antepartum exposure to benzodiazepines has been found to be associated with an increased risk for fetal cleft lip and/or cleft palate in some studies but not others. These contradictory results have led to a meta-analysis that found a small but significantly increased risk for major malformation or oral cleft alone according to data from the case-controlled studies (27). Until more research is reported, level 2 ultrasonography should be used to rule out visible forms of cleft lip. Many women of childbearing age are occupationally exposed to organic solvents. The most common female-dominated occupations are the health care professions and the clothing and textile industries. In the occupational setting, exposure to a multitude of solvents usually occurs. Because exposure usually involves more than one agent and different circumstances, adequate human epidemiological studies are difficult to interpret. The counseling of women who have been exposed to organic solvents is problematic because it is difficult to estimate the airborne or blood levels, the predominant chemicals and/or their by-products, the odor threshold, and the circumstances of exposure. Smelling organic solvents is not indicative of a significant exposure, as the olfactory nerve can detect levels as low as several parts per million, which is not necessarily associated with toxicity. Many organic solvents are teratogenic and embryotoxic in laboratory animals. The various malformations described include hydrocephaly, anencephaly, skeletal defects, cardiovascular abnormalities, blood changes, and neurodevelopmental deficits. In some of these studies, exposure levels were high enough to cause maternal toxicity, which biologically may be a source of fetal toxicity as well. Because many human studies are subject to recall and response bias, not always controlled for confounders, and not powerful, they are not confirmative enough. A metaanalysis concluded that maternal occupational exposure to organic solvents is associated with a tendency toward an increased risk for spontaneous abortions (n ⫽ 2899 patients; OR 1.25, CI 0.99–1.58) and major malformations (n ⫽ 7036 patients; OR 1.64, CI 1.16– 2.30) (28). Another prospective controlled study found that occupational exposure to organic solvents during pregnancy is associated with an increased risk for major malformations (RR 13; CI 1.8–99.5) in women who experienced symptoms of toxicity during exposure. Symptomatic exposure appears to predict higher fetal risk for malformations. More of these exposed women had previous miscarriages while working with organic solvents than did the comparison women who were not exposed (46.2% versus 19.2%) (29). Corticosteroids are used widely as agents for a variety of conditions. They are known to cross the placenta (both animal and human) and to exert teratogenic effects, inducing cleft palate in animals. Prospective follow-up studies failed to show an association between

Teratogenic Drugs and Chemicals in Humans

69

first-trimester exposure to corticosteroids and major birth defects. However, a meta-analysis of epidemiological studies pulled out clustering of cleft palate cases in the exposed group compared to controls. Those findings suggest a lower but yet significant teratogenic potential of corticosteroids (35). Moreover, French et al. in a prospective study followed up 477 infants who were exposed to corticosteroids at 24 weeks’ gestation or later (due to high risk of premature birth). This study showed that repeated corticosteroid courses were associated with reductions in birth weight and head circumference. At the age of 3 years, though, growth and severe disability outcomes did not appear to be related to increasing number of corticosteroid courses (36). Taking into consideration that there is evidence-based increased risk for cleft palate and birth weight, corticosteroids should not be the drugs of first choice in pregnancy. Since the risk is not as great as that associated with some other known human teratogens, those pregnant women who have an inadvertent exposure to corticosteroids should be reassured. Misoprostol, a prostaglandin E1 analog, has been indicated for use in preventing gastrointestinal lesions induced by nonsteroidal anti-inflammatory drugs and in treating duodenal and gastric ulcers. However, misoprostol has been abused because of its known side effect of inducing abortion. In Brazil, where elective abortions are illegal and misoprostol is available over the counter, it has been estimated that 10% of all pregnant women use it as abortifacient. Although a prospective observational cohort study on this population did not find an association with an increased risk for major malformations, it did find that the rate of miscarriages and fetal death were significantly higher among fetuses exposed to misoprostol in utero (37). In addition, an analytical case-control study suggested that gestational misoprostol use is associated with an increased risk for Mo¨bius sequence (38). Since the drug has been used on purpose to abort an unplanned or unwanted pregnancy, there is little to be done to protect the unborn. Cocaine use has increased rapidly over the last two decades for recreational purposes. It is now estimated that more than 10 million Americans currently use cocaine. As a rapid increase in use has occurred in women of childbearing age, much concern has been expressed about the potential effects of the drug during pregnancy. Animal studies show that cocaine affects fetal growth and brain development. The human studies confirmed that fetuses exposed to cocaine in utero have increased rates of stillbirth, smaller head circumference, lower birthweight, and increased rates of intracranial bleeding. Significant depression of interactive behavior and a less effective organizational response to environmental stimuli—suggestive of effects on neurologic integrity—were found in children of preschool age (40). The exposures’ effects on fetal development should be clearly outlined in counseling woman of reproductive age on cocaine. Environmental Contamination The common denominator of methyl mercury, carbon monoxide, and polychlorinated biphenyls (PCBs) is that their human teratogenic effect has been shown only following maternal exposure to excessive amounts. Extrapolation from these exposures to the background amounts of carbon monoxide or PCBs in the environment is not justified. Yet because the lower part of the dose-response curves has not been described, it is possible that moderate exposure may have clinical implications: heavy smokers, for example, have a carboxyhemoglobin level of 10% and even more; such levels have been shown to be associated with lower birth weights and, according to preliminary reports, with a less favorable developmental outcome (29).

70

Nulman et al.

Environmental lead may differ from the compounds above; recent studies suggest adverse developmental effects even with levels within the subtoxic range (above 10 µg/ dL but below 25 µg/dL) (19). Answer Valproic acid causes neural tube defects (NTDs) in about 2% of first-trimester exposures. Maternal α-fetoproteins and ultrasound at 16 weeks of gestation are accepted as the screening test. Positive cases are tested by determination of amniotic fluid α-fetoproteins. With these diagnostic means, NTDs are detected in almost 100% of cases.

REFERENCES 1. Fabro S, Brown NA, Scialli AR. Valproic acid and birth defects. Reprod Toxicol 1983; 2:9– 11. 2. Lammer EJ, Chen DT, Hoar RM, et al. Retinoic acid embryopathy. N Engl J Med 1985; 313: 837–841. 3. Koren G. Retinoic acid embryopathy. N Engl J Med 1986; 315:262. 4. Schardein J. Chemically Induced Birth Defects. New York: Marcel Dekker, 1985. 5. Rosa FW, Wilk AL, Kelsey FO. Vitamin A congeners. In: Sever JL, Brent RL, eds. Teratogen Update. New York: Alan R Liss, 1986, pp 61–70. 6. Jacobson SJ, Jones K, Johnson K, et al. A prospective multicenter study of pregnancy outcome following lithium exposure during the first-trimester of pregnancy. Lancet 1992; 339:530– 533. 7. Feldman Y, Koren G, Mattice D, et al. Determinants of recall and recall bias in studying drug effects in pregnancy. Teratology 1989; 40:37–46. 8. Mitchell AE, Cottler LB, Shapiro S. Effect of questionnaire design on recall of drug exposure in pregnancy. Am J Epidemiol 1986; 123:670–676. 9. Wilson JG. Environmental and Birth Defects. New York: Academic Press, 1973, p 58. 10. Rosett HZ, Weiner L. Alcohol and the Fetus. New York: Oxford University Press, 1984. 11. Graham JM, Hanson JW, Darby BL, et al. Independent dysmorphology evaluation at birth and 4 years of age for children exposed to varying amounts of alcohol in utero. Pediatrics 1988; 81:772–778. 12. Mills JL, Graubard BI. Is moderate drinking during pregnancy associated with an increased risk of malformations? Pediatrics 1987; 80:309–314. 13. Emerson DJ. Congenital malformations due to attempted abortion with aminopterin. Am J Obstet Gynecol 1982; 84:356–357. 14. Rosa FW. Spina bifida in infants of women treated with carbamazepine during pregnancy. N Engl J Med 1991; 324:674–676. 15. Longo LD. The biological effects of carbon monoxide on the pregnant woman, fetus and newborn infant. Am J Obstet Gynecol 1977; 129:69–103. 16. Koren G, Sharav T, Pastusz-ak A, et al. A multicenter prospective study of reproductive outcome following carbon monoxide poisoning in pregnancy. Reprod Toxicol 1991; 5:397–403. 17. Iturbe-Alessio J, Fonesca MDC, Mutchiniko Santos MA, et al. Risks of anticoagulant therapy in pregnant women with artificial heart valve. N Engl J Med 1986; 315:1390–1393. 18. Herbst AL, Uffelder H, Posbanzer DC. Adenocarcinoma of the vagina: association of maternal stilbestrol therapy with tumor appearance in young girls. N Engl J Med 1971; 284:878– 881. 19. Bellinger D, Levitou A, Waternaux C, et al. Longitudinal analysis of prenatal and postnatal lead exposure and early cognitive development. N Engl J Med 1987; 316:1037–1043.

Teratogenic Drugs and Chemicals in Humans

71

20. Amin-Zakil, Majeed MA, Greenwood MR, et al. Methylmercury poisoning in the Iraqi suckling infant: a longitudinal study over 5 years. J Appl Toxicol 1981; 1:210–214. 21. Harada M. Congenital Minamata disease: intrauterine methyl mercury poisoning. Teratology 1978; 18:285–288. 22. Rogan WJ, Gladen BC, Kun-Long H, et al. Congenital poisoning by polychlorinated biphenyls and their contaminants Taiwan. Science 1988; 241:334–336. 23. Rosa FW. Teratogen update: penicillamine. Teratology 1986; 22:127–131. 24. Ehrenbard LT, Chaganti RSK. Cancer in the fetal hydantoin syndrome. Lancet 1981; 2:97. 25. Hansen JW. Fetal hydantoin syndrome. Teratology 1976; 13:185–188. 26. Goldman AS, Zachai EH, Yaffe SJ. Fetal trimethadione syndrome. Teratology 1978; 17:103– 106. 27. Newman CGH. Clinical aspects of thalidomide embryopathy—a continuing preoccupation. Teratology 1985; 32:133–144. 28. Cohlan SQ. Tetracycline staining of the teeth. Teratology 1977; 15:127–130. 29. Sexton MJ, Fox NL, Hebel JR. The effects of neonatal exposure to tobacco on behavioral outcomes in three-year-old children. Teratology 1988; 37:491. 30. Sampson PD, Streissguth AP, Bookstein FL, et al. Incidence of fetal alcohol syndrome and prevalence of alcohol-related neurodevelopmental disorder. Teratology 1997; 56:317–326. 31. Bergman U, Rosa FW, Baum C, et al. Effects of exposure to benzodiazepine during fetal life. Lancet 1992; 340:694–696. 32. Dolovich LR, Addis A, Vaillancourt JMR, et al. Benzodiazepine use in pregnancy and major malformations or oral cleft: metaanalysis of cohort and case-control studies. BMJ 1998; 317: 839–843. 33. McMartin KI, Chu M, Kopecky E, et al. Pregnancy outcome following maternal organic solvent exposure: a meta-analysis of epidemiologic studies. Am J Industr Med 1998; 34:288– 292. 34. Khattak S, Moghtader G, McMartin K, et al. Pregnancy outcome following gestational exposure to organic solvents: a prospective controlled study. JAMA 1999; 281:1106–1109. 35. Diav-Citrin O, Park L, Pastuszak A, et al. Pregnancy outcome following maternal exposure to corticosteroids: a prospective controlled cohort study and a meta-analysis of epidemiological studies. Teratology 1998; 57:188. 36. French NP, Hagan R, Evans SE, et al. Repeated antenatal corticosteroids: size at birth and subsequent development. Am J Obstet Gynecol 1999; 180(1):114–121. 37. Schuller L, Pastuszak AL, Sanseverino MT, et al. Pregnancy outcome after abortion attempt with misoprostol. Teratology 1997; 55:36. 38. Pastuszak AL, Schuller L, Coelho KA, et al. Misoprostol use during pregnancy is associated with an increased risk for Moebius sequence. Teratology 1997; 55:36. 39. Koch S, Jager-Roman E, Losche G, et al. Antiepileptic drug treatment in pregnancy: drug side effects in the neonate and neurological outcome. Acta Paediatr 1996; 85(6):739–746. 40. Koren G, Nulman I, Rovet J, et al. Long-term neurodevelopmental risks in children exposed in utero to cocaine. Ann NY Acad Sci 1998; 846:306–313. 41. Delgado-Escueta AV, Janz D. Consensus guidelines: preconception counseling, management, and care of the pregnant woman with epilepsy. Neurology 1992; 42(4 suppl 5):149–160. 42. Lindhout D, Hoppener RJ, Meinardi H. Teratogenicity of antiepileptic drug combinations with special emphasis on epoxidation (of carbamazepine). Epilepsia 1984; 25(1):77–83.

5 Treatment for Epilepsy in Pregnancy Irena Nulman, Dionne Laslo, and Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

INTRODUCTION Pregnant women with epilepsy represent 0.5% of all pregnancies. Proper seizure control is the primary goal in treating these women. The commonly used antiepileptic drugs (AEDs) are established human teratogens. Factors such as epilepsy, AED-induced teratogenicity, the patient’s genetic predisposition, and the severity of her convulsive disorder may contribute to the adverse outcome of an epileptic woman’s pregnancy. The interaction of an AED with folic acid and vitamin K metabolism may lead to an increased risk for neural tube defect (NTD) and early neonatal bleeding. Psychological, hormonal, and pharmacokinetic changes in pregnancy may escalate seizure activity. Preconceptional counseling should include patient education as to the risks of uncontrolled seizures and the possible teratogenicity of AEDs. Genetic counseling should be performed if both parents have epilepsy or the disease is inherited. Seizure control should be achieved at least 6 months prior to conception and, if clinically possible, by the lowest effective dose of a single AED according to the type of epilepsy. The new AEDs are not recommended in pregnancy and require further research to prove their safety to humans. In addition, 5 mg/day of folic acid should be administered 3 months preconceptionally and during the first trimester to prevent malformations induced by inadequate levels of folic acid. Antenatal management should include an assessment of the patient for AED-associated birth defects through detailed ultrasound examination and levels of maternal serum α-fetoproteins. Therapeutic drug monitoring should be performed monthly or as clinically indicated. If phenobarbital, carbamazepine, or phenytoin is administered, maternal vitamin K supplementation should begin 4 weeks before the expected date of delivery. In order to prevent convulsions during labor, proper seizure control should be achieved during the third trimester. A benzodiazepine or phenytoin is found to be effective for seizure cessation during labor and delivery. Vitamin K should be administered to the newborn immediately after birth. The neonate should be examined carefully for epilepsy and AED-associated dysmorphology. Advising the patient on postpartum management regarding contraception and breast-feeding will help maximize the best possible outcome for the newborn and the mother. With proper preconceptional, antenatal, and postpartum management, up to 95% of these pregnancies have been reported to show favorable results.

73

74

Nulman et al.

Table 1 Congenital Malformations in Humans Caused by Antiepileptic Drugs Malformations Congenital heart defects Cleft lip and/or palate NTD a Genitourinary defects Cognitive impairment Dysmorphic syndrome a

Phenytoin

Valproic acid

Carbamazepine

Phenobarbital

⫹ ⫹ ⫺ ⫹ ⫹ ⫹

⫹ ⫹ ⫹ ⫹ ⫹ ⫹

⫺ ⫺ ⫹ ⫹ ⫾ ⫹

⫹ ⫹ ⫺ ⫹ ⫾ ⫹

NTD, neural tube defect; ⫾, inconclusive evidence.

Epilepsy, although not common among pregnant women, is the most common neurological disorder during gestation. Increased public awareness of the progress in diagnosis and management of epilepsy have enabled many epileptic women to bear children and manage careers (1,2). Pregnant women with epilepsy constitute 0.5% of all pregnancies, and medical professionals should be aware that—with appropriate selection of treatment and prudent preconceptional, antenatal, and postpartum management—up to 95% of these pregnancies have been reported to have favorable results (3). The evaluation of the risk and safety of pregnancy in such women is complicated by a variety of factors. Women with epilepsy are at risk for menstrual abnormalities, reproductive endocrine disorders, and reduced fertility (4). Polycystic ovaries as well as hypo- or hypergonadotrophic hypogonadism are more common in women with epilepsy than in general population. The investigators relate those abnormalities to the effect of seizures on the hypothalamic-pituitary-gonadal axis (5) and to the possible side effect of AEDs (6,7). Even when not exposed in utero to medications, infants of mothers with epilepsy have higher rates of major and minor malformations (8,9) when compared to the general population. To add to the complexity, several commonly used AEDs are established human teratogens. A large number of studies in humans is supported by results acquired from animal investigations showing a consistent increased risk for congenital malformation following AED exposure in utero. Table 1 presents congenital malformations in humans caused by AEDs for which there is consensus among scientists. No agreement has been reached among experts about which of the commonly used AEDs is safest for the unborn baby. Research indicates that the incidence of major and minor malformations is influenced by the number of anticonvulsant medications, the dose of the drug, timing of use during gestation, pharmacokinetics, and differences in metabolism (8,10). Some investigators hypothesize that anticonvulsant-induced teratogenicity occurs in genetically predisposed individuals (11). Other researchers have reported that epilepsy per se—the type and severity of the seizure disorder—also contributes to the dysmorphology (12). Considering the complex interaction between genetic and environmental factors, it is difficult to attribute congenital abnormalities in children of epileptic mothers to any single factor (13), especially in view of the lack of large enough studies with sufficient power to control for such determinants. MECHANISMS AND CLINICAL IMPLICATIONS OF TERATOGENICITY Several mechanisms have been proposed for AED-induced teratology. Some anticonvuls ant medications (e.g., phenytoin) form intermediate oxide metabolites that are known to

Treatment for Epilepsy in Pregnancy

75

be embryotoxic. Free active oxide radicals have been shown to bind to proteins and nucleic acids and may interfere with DNA and RNA synthesis. Critical amounts of free radicals may increase the risk of perinatal death, intrauterine growth retardation, and malformations (14). Scavenging enzymes capable of conjugating these free radicals to inactive substances may prevent fetal damage, and variabilities in such enzymes may explain variable fetal outcome. Unstable intermediates can also be metabolized to nonreactive hydrodiones by epoxide hydrolase. It has been proposed that fetuses with low levels of free radical– scavenging enzymes and low activity of epoxide hydrolase are at increased risk to develop malformations associated with phenytoin. Polytherapy may lead to excessive amounts of unstable epoxides, such as arene oxides, and inhibit epoxide metabolism, especially in fetuses with a genetic defect in fetal epoxide hydrolase activity (15). Evidence for increased risk for major malformations with an increase number of AEDs comes from several epidemiological studies. Lindhout et al. (16) reported a 5% risk for birth defects in children of epileptic women who took two drugs concomitantly, 10% in those who took three drugs, and more than 20% when four drugs were used. The authors reported a malformation risk of only 3% (versus 2% as in untreated epileptic women or in general population) when the seizures were controlled with only one AED. The occurrence of specific malformations depends on the timing of exposure during embryogenesis (10,17). Neural tube defects (NTDs) occur before closure of the neural tube, between days 21 and 28 after the first day of the last menstrual period (LMP). Cleft lip occurs with exposure before day 35 and cleft palate before day 70, whereas congenital heart defects occur with exposure before day 42 post-LMP (18). Exposure after the first trimester should not affect rates of dysmorphology except for the toxic effects on the brain, which develops throughout pregnancy (19). The most commonly observed major malformations are heart defects, orofacial clefting, genitourinary malformations, and NTDs (13). Minor anomalies that were described as a part of the fetal hydantoin syndrome (20) were later hypothesized to be associated with maternal epilepsy and not necessarily with the AEDs (9,21,22). We have recently detected different patterns of minor anomalies to be caused by phenytoin, carbamazepine, and untreated epilepsy (9). Because the vast majority of NTDs can be ruled out by maternal or amniotic fluid α-fetoproteins combined with ultrasound, these examinations should be routinely performed in women taking carbamazepine or valproic acid. At present some practicing neurologists consider carbamazepine to be the drug of choice in pregnancy (18). Carbamazepine poses a risk of 1% for the NTDs (23), compared to 2% for valproic acid (24). It does not cause acne, hirsutism, or facial coarsening, as does phenytoin. While the fetal hydantoin syndrome has been consistently associated with cognitive damage to some children (20), studies with carbamazepine do not suggest similar damage (25), although the literature is not consistent (26,27) and further studies at the preschool and school-age levels are needed. Koch et al. have suggested that valproic acid–exposed children had higher rates of neurological dysfunction. Serum concentrations of valproic acid at birth correlated with the degree of neonatal hyperexcitability and the degree of dysfunction when the children were reexamined at the age of 6 years (28). In a recent consensus opinion, it was suggested that the anticonvulsive medication that is most effective for the type of epilepsy and seizure control for a given patient should be used (18). Women in need of polytherapy are likely to suffer from more severe forms of epilepsy and drug-drug interaction, thus further increasing the risk for adverse pregnancy outcome (10).

76

Nulman et al.

Another mechanism that has been implicated in AED-mediated teratogenicity is folate deficiency (10). Up to a 90% reduction in serum folate levels was reported in patients treated with phenytoin, carbamazepine, and barbiturates. Conversely, valproic acid did not reduce levels of folate directly but interfered with its metabolism (29,30). Folate supplementation was found to be effective in preventing several malformations, in particular NTDs. The Medical Research Council study (31) reported a 70% reduction of NTD recurrence among pregnant women who were supplemented with 4 mg of folic acid before conception and during gestation. The current recommendations call for 5 mg of folic acid per day in women being treated with valproic acid or carbamazepine (2,32). Although there are no data at present to document the efficacy of this regimen in women with epilepsy, the potential benefit of such approach outweighs the theoretical risk of high levels of folate (33). Vitamin B 12 levels should be assessed before folic acid supplementation is begun so as to avoid neurological symptoms of vitamin B 12 deficiency (34). The genetic influence of the maternal epileptic syndrome and seizure effects on fetal development in regards of congenital malformation are not well defined. A control group with a large enough sample size of unmedicated women with epilepsy was not assessed by any research group. Even if collected, such a control group might represent less severe forms of epilepsy with low frequency of seizures. The effect of seizure activity during pregnancy on fetal well-being has been investigated by several groups (35). Isolated, short seizures are generally not believed to have an adverse effect on the fetus. Spontaneous abortions, injury to the mother and fetus, fetal hypoxia, bradycardia, and antenatal death were reported (3,36,37) with repeated tonicclonic seizures, complex partial seizures, and status epilepticus. There is ample evidence that physiological changes during pregnancy may affect the duration and frequency of seizures. Seizure rates have been reported to increase in 17–37% of women with epilepsy (2,36). When seizures recur in a well-controlled woman during pregnancy, they most often appear during the first and second trimesters (33). Poor patient compliance, sleep disturbances, nausea, vomiting, and decreased levels of free (unbound) drugs are the main risk factors believed to decrease seizure control (18,38). Increases in estrogen levels during pregnancy may reduce seizure threshold levels (39). Progesterone reduces intestinal motility, thus interfering with the secretion of mucus as well as gastric pH, and it may affect drug absorption (38). Changes in seizure frequency during pregnancy may also stem from fluid and sodium retention, hyperventilation, and emotional and psychological problems (40). A potential problem in the management of seizures may also arise from altered pharmacokinetics of AEDs associated with pregnancy.

PREGNANCY-INDUCED PHARMACOKINETIC CHANGES OF ANTIEPILEPTIC DRUGS The plasma concentration of anticonvulsant drugs tends to fall during pregnancy as a result of a 50% expansion in plasma volume, decreased protein binding (41), increased clearance rate, and a tendency toward reduced patient compliance due to fears of teratogenicity. As most anticonvulsant drugs are acidic or neutral, they are highly bound to serum albumin. During late pregnancy, albumin levels fall, with a corresponding decrease in the fraction of bound drug, which causes total (unbound and protein-bound) plasma concentration to fall. The decrease in plasma protein binding makes more free drug available for

Treatment for Epilepsy in Pregnancy

77

biotransformation and clearance. Although the free AED levels fall much less than the total levels, they often do decline significantly during pregnancy (18,42). Monitoring the total plasma concentrations of AEDs can, therefore, be misleading. In complex clinical cases, therapeutic drug monitoring that can measure both protein-bound and unbound drug concentrations can be helpful. The measurement of free drug concentrations of highly bound drugs should be considered in cases where seizure control is not achieved. It should be remembered that the constant or even higher free drug concentration may provide appropriate antiepileptic control, as it is the free drug that reaches the brain. The increase in glomerular filtration rate (GFR) (43) and renal plasma flow (44) may theoretically enhance the clearance of renally excreted drugs such as gabapentin and vigabatrin. However, there are presently no studies available regarding the pharmacokinetics of these agents during human pregnancy. Carbamazepine Carbamazepine has a relatively slow absorption, with 70–80% protein binding to albumin. Hepatic metabolism is the main route of elimination, and there may be a decrease in serum concentrations during the first months of therapy as a consequence of autoinduction of metabolism. Dosage intervals and sample times are critical in interpreting serum concentrations, and large peak-trough fluctuations can be minimized by using a controlled-release formulation (45). As drug levels tend to be lower in pregnancy and bioavailability may be lower than with conventional carbamazepine, higher dosages may be required when a controlled-release medication is used (46). The concentration of the active metabolite (carbamazepine-10,11-epoxide) was reported to increase during pregnancy, possibly as a result of the increased carbamazepine metabolism and impaired conversion of carbamazepine-10,11-epoxide to carbamazepine10, 11 trans-diol. This increase is of potential importance, as the metabolite (10, 11epoxide) is believed to have pharmacological activity comparable to that of the parent drug (47). Phenytoin Phenytoin follows nonlinear pharmacokinetics and has a narrow therapeutic window (48). It is highly bound to protein (90–93%) (49,50) and cleared mainly by saturable hepatic metabolism. A substantial increase in 8-hydroxylation during pregnancy may be responsible, at least partially, for its increased clearance rate and consequently decreased serum concentration (51). Generally, a fall in total serum phenytoin concentration may cause lack of seizure control and may require increases in dose. However, as indicated above, the total concentration by itself may not indicate a fall in free drug concentrations. The decrease in the protein binding of phenytoin may be an important mechanism for decreasing total drug concentrations in pregnancy, as it is the free drug that is available for the enhanced metabolism. Valproic Acid Valproic acid is rapidly absorbed and highly protein-bound to plasma albumin (88–92%) (52). The interpretation of its pharmacokinetics is limited by large fluctuations in the concentration-time profile, wide therapeutic index, and concentration-dependent protein

78

Nulman et al.

binding (53,54). Analysis of unbound valproic acid is not routine; similarly, there is no established therapeutic range. Dose adjustments during pregnancy are best made by clinical observation in combination with therapeutic drug monitoring. Divided doses are preferred to avoid high levels of valproic acid in serum (18). Phenobarbital Phenobarbital has been prescribed less frequently during the last few years because of its tendency to produce sedation and impaired cognitive function. It has a high oral bioavailability (90%) and is only 50% protein-bound. Like phenytoin and carbamazepine, it induces hepatic microsomal oxidative enzymes and may affect the therapeutic efficacy of other drugs. Neonates exposed prenatally to phenobarbital should be monitored for withdrawal symptoms for 2–6 weeks starting at day 7 of life because of the long elimination halflife of phenobarbital (100 hours).

NEW ANTIEPILEPTIC DRUGS The new AEDs that are not highly protein-bound (topiramate, felbamate, oxcarbazepine) or non–protein bound (gabapentin, vigabatrin) are eliminated from the body through renal clearance (vigabatrin, gabapentin), which has no effect on the cytochrome P-450 enzyme system (gabapentine, lamotrigine, vigabatrin), which have no antifolate effects, and which have no arene oxide metabolites and if given in monotherapy may be considered for use for women with epilepsy, but there is little information regarding their pharmacokinetics and safety during pregnancy. Animal studies have been extremely helpful in elucidating the mechanisms of adverse effects and teratogenicity of new AEDs. They are very informative in testing hypotheses and assessing nutritional and environmental factors that may interfere with or modify normal embryonic or fetal development. Although animal studies help to clarify pharmacokinetic changes in pregnancy and to define the risk factors associated with teratogenicity (55), they may be imperfect predictors of human teratogenicity. Animals may show increased sensitivity because of higher dose ranges than used in humans (tiagabine) (56) or species-specific effect (topiramate), but today reports regarding animal reproductive toxicology of new AEDs appear to be promising. Although results of animal studies regarding the teratogenic effects of the new AEDs are encouraging, it is too early tell whether such data will apply to humans. The Lamotrigine Pregnancy Registry report (September 1, 1992 through March 31, 1998) contains a description of all prenatal exposure to lamotrigine voluntarily and prospectively reported to the registry. There were no (CI, 0–12.6) birth defects in 34 neonates after first-trimester exposure to lamotrigine monotherapy and 5.6% (95% CI, 2.3–12.3) birth defects in 107 neonates after combined monotherapy and polytherapy exposure in prospective reports. In prospective reports of all trimesters of exposure combined, there were no birth defects in 37 pregnancy outcomes. The registry findings of 5.6% of birth defects in the monotherapy and polytherapy groups are not higher when compared with that expected in pregnant women with epilepsy (8,9). There was no consistent pattern of malformations among defects reported. The Lamotrigine Pregnancy Registry Advisory Committee concluded that the number of exposed pregnancy outcomes represents a sample

Treatment for Epilepsy in Pregnancy

79

of insufficient size for reaching definitive conclusions regarding safety of lamotrigine in pregnancy.

EFFECT OF ANTIEPILEPTIC DRUGS ON VITAMIN K The association between maternal anticonvulsant therapy and neonatal hemorrhage was reported 40 years ago (57) and supported by a number of subsequent studies and reports (58). These hemorrhages, which typically occur during the first 24 hours after birth (in contrast to the classic neonatal bleeding which occurs on day 2 or 3 after delivery (59), may be severe, involving the skin as well as the brain, pleural, and peritoneal cavities. The mechanism and origin of these hemorrhages is not fully understood, but vitamin K deficiency was observed in neonates exposed in utero to enzyme-inducing AEDs such as carbamazepine, phenytoin, phenobarbital, and primidone. These medications readily cross the placenta and induce liver enzymatic pathways, resulting in increased degradation of vitamin K, and they produce proteins that are induced by vitamin K absence (PIVKA). These proteins are present when vitamin K is absent in neonates exposed in utero to AEDs. The decarboxylated form of prothrombin, PIVKA II, is the most sensitive marker for vitamin K deficiency (60) and was proven informative in studies of neonatal vitamin K levels. Although vitamin K does not easily cross the placenta from the maternal to the fetal circulation, prenatal supplementation of vitamin K results in valuable effects in preventing neonatal bleeding (61). The consensus guidelines (10) regarding women receiving enzyme-inducing AEDs call for antenatal maternal vitamin K supplementation at 20 mg orally throughout the last 4 weeks of gestation and 1 mg of vitamin K parenterally to the neonate immediately after delivery. If PIVKA are found in cord blood specimens, fresh frozen plasma should be given at a dose of 20 mL/kg over a period of 1 to 2 hours (10). While some experts still debate the value of maternal antenatal supplementation during the last month of pregnancy, general agreement has been reached regarding the necessity of parenteral neonatal prophylaxis (2).

PRECONCEPTIONAL COUNSELING Women should be counseled about the potential risk of increased seizure activity during pregnancy so as to make sure that they do not avoid taking their medication. Poor compliance, resulting in increased seizure activity, has tangible risks. Preconceptional counseling should optimally begin at least 3 month before conception to allow for adequate supplementation of folic acid. Adequate patient education regarding the increased incidence of major malformations (62) and possible adverse effects of AEDs to the fetal central nervous system (25,63) should be achieved. Genetic counseling should be offered if both partners have epilepsy or the epileptic disorder is inherited. Gradual drug discontinuation (over at least 3 months) should be considered if the patient has been seizure-free for 2 or more years. An epileptic woman who is planning a pregnancy should be encouraged to quit smoking, maintain good nutrition, and get enough sleep (18). If treatment with anticonvulsant medications cannot be avoided, proper seizure control should be achieved by the lowest effective dose of the single AED that best controls seizures in the given patient or, alternatively, pregnancy should be delayed until seizure

80

Nulman et al.

control is reached. It should be remembered that polytherapy and/or higher daily doses of AEDs are associated with higher rates of congenital malformation. If valproic acid is indicated, divided doses are preferred to avoid high levels of valproic acid in plasma (18,64). The combination of valproic acid, carbamazepine, and phenobarbital has been reported to be more teratogenic than other combinations (33). Folate supplementation at 5 mg/day should start 3 months before conception and continue until the end of the first trimester. If possible, serum folate levels should be monitored to confirm sufficient supplementation.

ANTENATAL MANAGEMENT More than 50% of all pregnancies are unplanned (65); if the pregnancy is diagnosed while the woman is seizure-free, there is no proven benefit to changing the patient’s drug because any morphological teratogenic effects will be irreversible by 10 weeks of gestation (18). The patient should be advised regarding appropriate prenatal diagnosis for AED-associated abnormalities. For example, targeted fetal ultrasound examination at 18 weeks can diagnose up to 95% of fetuses with open NTD (66,67) as well as other anomalies (68). Early transvaginal ultrasonography at 11–13 weeks to assess for NTD is presently possible (33). Detailed sonographic imaging of the fetal heart at 18–20 weeks, followed by fetal echocardiography, which can identify up to 85% of cardiac defects. Imaging of the fetal face for cleft lip at 18–20 weeks may be performed, but the sensitivity of this assessment has yet to be established (33). Amniocentesis with amniotic fluid α-fetoprotein and acetylcholinesterase may support blood serum findings caused by structural malformations. Epoxide hydrolase activity in amniocytes has been suggested to predict those fetuses at risk for phenytoin-mediated anomalies; however, this test has not yet been confirmed for routine clinical use. Vitamin K supplementation of the mother is recommended, as detailed above. Optimally, therapeutic drug monitoring should be performed every 1–2 months or more frequently if seizure control is not achieved.

LABOR, DELIVERY, AND BIRTH Tonic-clonic seizures occur during labor or after delivery in 1–2% of women with epilepsy (18). Monitoring plasma AED levels during the third trimester and regular administration of medication(s) are essential to prevent seizures due to inappropriately low serum concentrations (3). Convulsive seizures at the time of labor and delivery are commonly treated with intravenous administration of benzodiazepines or phenytoin. Intravenous administration of phenytoin should be given under cardiac monitoring to detect possible dysrhythmias (66). Emergency cesarean section is often performed in the case of status epilepticus or when there are repeated tonic-clonic, psychomotor, or absence seizures (37). Obstetric intervention in the form of induction of labor, mechanical rupture of membranes, forceps delivery, or cesarean section is more common among women with epilepsy, as are such obstetric complications as vaginal bleeding, anemia, and preeclampsia (3).

Treatment for Epilepsy in Pregnancy

81

MANAGEMENT DURING THE PUERPERIUM Vitamin K should be administered to the newborn immediately after birth. The neonate should be examined carefully for AEDs and epilepsy-associated dysmorphology. If the neonate was exposed to phenobarbital or primidone, observation for withdrawal symptoms should be performed during the first 7 months of life. Maternal anticonvulsant drug levels should be carefully maintained with appropriate changes in dosage, bearing in mind that the decreased clearance rate postnatally may be associated with toxicity. The patient should be counseled regarding postpartum contraception. The use of oral contraceptive agents is not contraindicated in women with epilepsy (67). For patients treated with hepatic enzyme–induced anticonvulsants, a higher dose of estrogen (50 µg) in combination with an oral contraceptive pill will be needed to compensate for the higher rate of clearance of the hormone. The issues of breast-feeding should also be discussed with the mother, as most women with epilepsy can breast-feed. Phenytoin and valproic acid are highly proteinbound; therefore, only low levels of these drugs are present in breast milk. Carbamazepine and phenobarbital are present in higher concentrations. Because of the sedative effect of phenobarbital, breast-feeding by women taking this drug is not recommended. When the mother is taking phenobarbital and breast-feeding, the infant must be monitored for the risk of lethargy and poor suck.

CONCLUSIONS Proper seizure control is the primary goal in treating women with epilepsy. Patients should understand the risks associated with uncontrolled seizures as well as the teratogenicity of the anticonvulsive medication in question. If AEDs cannot be avoided, a first-line drug for the seizure type should be used at the lowest effective dose. Judicious preconceptional, antenatal, and postpartum management leads to favorable maternal and neonatal outcome in the vast majority of patients.

REFERENCES 1. Dansky LV, Andermann E, Andermann F. Marriage and fertility in epileptic patients. Epilepsia 1980; 21(3):261–271. 2. Byrne B. Epilepsy and pregnancy. Ir Med J 1997; 90:173–174. 3. Yerby MS. Pregnancy and epilepsy. Epilepsia 1991; 32(suppl 6):S51–S59. 4. Medeiros YS, Calixto JB. Inhibitory effect of diphenylhydantoin on myometrium from pregnant women in vitro: A comparative study with nicardipine and trifluoperazine. Pharmacol Res 1990; 22(5):597–603. 5. Nappi C, Meo R, Di Carlo C, et al. Reduced fertility and neuroendocrine dysfunction in women with epilepsy. Gynecol Endocrinol 1994; 8:133–145. 6. Isojarvi JI, Laatikainen TJ, Pakarinen AJ, et al. Menstrual disorders in women with epilepsy receiving carbamazepine. Epilepsia 1995; 36:676–681. 7. Isojarvi JI, Laatikainen TJ, Pakarinen AJ, et al. Polycystic ovaries and hyperandrogenism in women taking valproate for epilepsy. N Engl J Med 1993; 329:1383–1388. 8. Kaneko S. Antiepileptic drug therapy and reproductive consequences: functional and morphological effects. Reprod Toxicol 1991; 5(3):179–198.

82

Nulman et al.

9. Nulman I, Scolnik D, Chitayat D, et al. Findings in children exposed in utero to phenytoin and carbamazepine monotherapy: independent effects of epilepsy and medications. Am J Med Genet 1997; 68(1):18–24. 10. Kaneko S, Otani K, Fukushima Y, et al. Teratogenicity of antiepileptic drugs: analysis of possible risk factors. Epilepsia 1988; 29:459–467. 11. Janz D. On major malformations and minor anomalies in the offspring of parents with epilepsy: review of the literature. In: Janz D, Dam M, Richeus A, et al., eds. Epilepsy, Pregnancy, and the Child. New York: Raven Press, 1982:211–222. 12. Majewski F, Steger M, Richter B, et al. The teratogenicity of hydantoins and barbiturates in humans, with considerations on the etiology of malformations and cerebral disturbances in the children of epileptic parents. Int J Biol Res Pregnancy 1981; 2(1):37–45. 13. Dansky LV, Finnell RH. Parental epilepsy, anticonvulsant drugs, and reproductive outcome: epidemiologic and experimental findings spanning three decades. 2: Human studies. Reprod Toxicol 1991; 5(4):301–335. 14. Finnel RM, Buehler BA, Kerr BM, et al. Clinical and experimental studies linking oxidative metabolism to phenytoin-induced teratogenesis. Neurology 1992; 42(4 suppl 5):25–31. 15. Buehler BA, Delimout D, Van Waes M, et al. Prenatal prediction of risk of the fetal hydantoin syndrome. N Engl J Med 1990; 322(22):1567–1572. 16. Lindhout D, Hoppener RJ, Meinardi H. Teratogenicity of antiepileptic drug combinations with special emphasis on epoxidation (of carbamazepine). Epilepsia 1984; 25(1):77–83. 17. Sulik KK, Johnston MC, Daft PA, et al. Fetal alcohol syndrome and DiGeorge anomaly: critical ethanol exposure periods for craniofacial malformations as illustrated in an animal model. Am J Med Genet 1986; (suppl 2):97–112. 18. Delgado-Escueta AV, Janz D. Consensus guidelines: preconception counseling, management, and care of the pregnant woman with epilepsy. Neurology 1992; 42(4 suppl 5):149–160. 19. Spreen O, Tupper D, Risser A, et al. Chronology of neural development. In: Human Developmental Neuropsychology. New York: Oxford University Press, 1984, pp 26–28. 20. Hanson JW, Myrianthopoulos NC, Harvey MA, et al. Risks to the offspring of women treated with hydantoin anticonvulsants, with emphasis on the fetal hydantoin syndrome. J Pediatr 1976; 89(4):662–668. 21. Gaily E, Granstrom ML. Minor anomalies in children of mothers with epilepsy. Neurology 1992; 42(4 suppl 5):128–131. 22. Gaily E, Granstrom ML, Hiilesmaa V, et al. Minor anomalies in the offspring of epileptic mothers. J Pediatr 1988; 112:520–529. 23. Rosa FW. Spina bifida in infants of women treated with carbamazepine during pregnancy. N Engl J Med 1991; 324(10):674–677. 24. Omtzigt JG, Los FJ, Grobbee DE, et al. The risk of spina bifida aperta after first-trimester exposure to valproate in a prenatal cohort. Neurology 1992; 42(4 suppl 5):119–125. 25. Scolnik D, Nulman I, Rovet J, et al. Neurodevelopment of children exposed in utero to phenytoin and carbamazepine monotherapy. JAMA 1993; 271(10):767–770. 26. Ornoy A, Cohen E. Outcome of children born to epileptic mothers treated with carbamazepine during pregnancy. Arch Dis Child 1996; 75(6):517–520. 27. Jones KL, Lacro RV, Johnson KA, et al. Pattern of malformations in the children of women treated with carbamazepine during pregnancy. N Engl J Med 1989; 320(25):1661–1666. 28. Koch S, Jager-Roman E, Losche G, et al. Antiepileptic drug treatment in pregnancy: drug side effects in the neonate and neurological oncome. Acta Paediatr 1996; 85(6):739–746. 29. Ogawa Y, Kaneko S, Otani K, et al. Serum folic acid levels in epileptic mothers and their relationship to congenital malformations. Epilepsy Res 1991; 8(1):75–78. 30. Dansky LV, Rosenblatt DS, Andermann E. Mechanisms of teratogenesis: folic acid and antiepileptic therapy. Neurology 1992; 42(4 suppl 5):32–42. 31. MRC Vitamin Study Research Group. Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. Lancet 1991; 338(8760):131–137.

Treatment for Epilepsy in Pregnancy

83

32. Brodie MJ, Dichter MA. Antiepileptic drugs. N Engl J Med 1996; 334(3):168–175. 33. Malone FD, D’Alton ME. Drugs in pregnancy: anticonvulsants. Semin Perinatol 1997; 21(2): 114–123. 34. Shuster EA. Epilepsy in women (Symposium on Epilepsy—Part VII). Mayo Clin Proc 1996; 71(10):991–999. 35. Yerby M, Koepsell T, Daling J. Pregnancy complications and outcomes in a cohort of women with epilepsy. Epilepsia 1985; 26:631–635. 36. Lopes-Cendes I, Andermann E, Candes F, et al. Risk factors for changes in seizure frequency during pregnancy of epileptic women: A cohort study (abstr). Epilepsia 1992; 33(suppl 3): 57. 37. Hiilesmaa VK. Pregnancy and birth in women with epilepsy. Neurology 1992; 42(4 suppl 5): 8–11. 38. Loebstein R, Lalkin A, Koren G. Pharmacokinetic changes during pregnancy and their clinical relevance. Clin Pharmacokinet 1997; 33(5):328–343. 39. Morrell MJ. Hormones, reproductive health and epilepsy. In: Wyllie E, ed. The treatment of Epilepsy: Principles and Practice, 2nd ed. Baltimore: Williams & Wilkins, 1997. 40. Koren G. Changes in drug disposition in pregnancy and their clinical implications. In: Koren G, ed. Maternal-Fetal Toxicology: A Clinician’s Guide, 2nd ed. New York: Marcel Dekker, 1994, pp 3–13. 41. Pirani BB, Campbell DM, MacGillivray I. Plasma volume in normal first pregnancy. J Obstet Gynaecol Br Commonw 1973; 80(10):884–887. 42. Yerby MS, Freil PN, McCormick K. Antiepileptic drug disposition during pregnancy. Neurology 1992; 42(suppl 5):12–16. 43. Davison JM, Hytten FE. Glomerular filtration during and after pregnancy. J Obstet Gynaecol Br Commonw 1974; 81(8):588–595. 44. Dunihoo DR. Maternal Physiology. In: Dunihoo DR, ed. Fundamentals of Gynecology and Obstetrics. Philadelphia: Lippincott, 1992, pp 280–284. 45. Yerby MS, Friel PN, Miller DQ. Carbamazepine protein binding and disposition in pregnancy. Ther Drug Monit 1985; 7(3):269–273. 46. Tomson T, Almkvist O, Nilsson BY, et al. Carbamazepine-10, 11-epoxide in epilepsy: a pilot study. Arch Neurol 1990; 47(8):888–992. 47. Bourgeois BFD, Wad N. Individual and combined antiepileptic and neurotoxic activity of carbamazepine and carbamazepine-10, 11-epoxide in mice. J Pharmacol Exp Ther 1984; 231: 411–415. 48. Armijo JA, Cavada E. Graphic estimation of phenytoin dose in adults and children. Ther Drug Monit 1991; 13(6):507–510. 49. Brodie MJ. Management of epilepsy during pregnancy and lactation. Lancet 1990; 336(8712): 426–427. 50. Perucca E, Richens A, Ruprah M. Serum protein binding of phenytoin in pregnant women. Proc Br Pharmacol Soc 1981; 11:409P–410P. 51. Bologa M, Tang B, Klein J, et al. Pregnancy-induced changes in drug metabolism in epileptic women. J Pharmacol Exp Ther 1991; 257(2):735–740. 52. Thomson AH, Brodie MJ. Pharmacokinetics optimisation of anticonvulsant therapy. Clin Pharmacokinet 1992; 23(3):216–230. 53. Henriksen O, Johannessen SI. Clinical and pharmacokinetic observations on sodium valproate: a 5-year follow-up study of 100 children with epilepsy. Acta Neurol Scand 1982; 65(5):504– 523. 54. Pugh CB, Garnett WR. Current issues in the treatment of epilepsy. Clin Pharmacol 1991; 10(5):335–358. 55. Wilson JG. Current status of teratology—general principles and mechanisms derived from animal studies. In: Wilson JG, Fraser FC, eds. Handbook of Teratology, vol. 1. New York: Plenum Press, 1977, pp 47–74.

84

Nulman et al.

56. Gabitril package insert, Abbott Laboratories. September 2, 1998. 57. Van Creveld S. Nouveau aspects de la maladie he´imorragique du nouveau-vie. Arch Fr Pediatr 1958; 6:721–735. 58. Moslet U, Hansen ES. A review of vitamin K, epilepsy and pregnancy. Acta Neurol Scand 1992; 85(1):39–43. 59. Sutor AH. Vitamin K deficiency bleeding in infants and children. Semin Thromb Hemostas 1995; 21(3):317–329. 60. Anai T, Hirota Y, Oga M, et al. PIVKA-II (protein induced by vitamin K absence-II) status in newborns exposed to anticonvulsant drugs in utero. Nippon Sanka Fujinka Gakkai Zasshi 1991; 43(3):347–350. 61. Cornelissen M, Steegers-Theunissen R, Kollee L, et al. Supplementation for vitamin K in pregnant women receiving anticonvulsant therapy prevents neonatal vitamin K deficiency. Am J Obstet Gynecol 1993; 168(3 part 1):884–888. 62. Nakane Y, Okuma T, Takahashi R, et al. Multi-institutional study on the teratogenicity and fetal toxicity of antiepileptic drugs: a report of a collaborative study group in Japan. Epilepsia 1980; 21(6):663–680. 63. Fujioka K, Kaneko S, Hirano T, et al. A study of the psychomotor development of the offspring of epileptic mothers. In: Sato T, Shinagawa S, eds. Antiepileptic Drugs and Pregnancy. Amsterdam: Excerpta Medica, 1984, pp 196–206. 64. Samren EB, van Duijn CM, Koch S, et al. Maternal use of antiepileptic drugs and the risk of major congenital malformations: a joint European prospective study of human teratogenesis associated with maternal epilepsy. Epilepsia 1997; 38:981–990. 65. Sophocles AM, Brozovich EM. Birth control failure among patients with unwanted pregnancies: 1982–1984. J Fam Pract 1986; 22(1):45–48. 66. American Collage of Obstetricians and Gynecologists. Seizure Disorders in Pregnancy. ACOG educational bulletin No. 231, December 1996. Committee on Educational Bulletins of the International Journal of Gynaecology and Obstetrics. Int J Obstet Gynaecol 1997; 56(3):279– 286. 67. Mattson RH, Cramer JA, Darney PD, et al. Use of oral contraceptives by women with epilepsy. JAMA 1986; 256(2):238–240. 68. Koren G, Nulman I. Fetal malformations associated with drugs and chemicals: visualization by sonography. In Koren G, ed. Maternal-Fetal Toxicology: A Clinician’s Guide. 2nd ed. New York: Marcel Dekker, 1994, pp 627–639.

6 The Safety of Commonly Used Antidepressants in Pregnancy Gideon Koren, Anne Pastuszak, Sheila Jacobson, and Irena Nulman The Hospital for Sick Children, Toronto, Ontario, Canada

Clinical Case A woman with manic-depressive disorder, dependent on lithium, has just found out she is pregnant. She considers termination of pregnancy, a move that was supported by two other doctors. What should be your advice?

INTRODUCTION A substantial number of women of reproductive age need antidepressant therapy, and many of the conditions represented cannot be well controlled without such medications during pregnancy. The information available on the reproductive safety of the major classes of antidepressants has been sparse. Recently, two large prospective studies, the result of a collaboration by several teratogen information services, have shed light on pregnancy outcome following first-trimester exposure to lithium, tricyclic antidepressants, and fluoxetine. This chapter reviews the results of these studies and their practical implications.

LITHIUM IN PREGNANCY It is estimated that 0.1% of pregnant women use lithium (1). This drug crosses the placenta, and concentrations are much the same in maternal and cord serum. Therapeutic and supratherapeutic doses of lithium have caused craniofacial defects in rodents (2), but malformations have not been demonstrated in primates (3). Since 1970, isolated instances of congenital anomalies (specifically Ebstein’s anomaly) as well as normal outcomes have been reported in association with lithium exposure during pregnancy (4–6). In 1968 the Danish Registry of Lithium Babies was established to obtain further information about infants and children who had been exposed to lithium during the first trimester of pregnancy. Data collected by a voluntary retrospective reporting system included cases from Scandinavia, Canada, and the United States. Of a total of 225 cases 85

86

Koren et al.

reported by 1983, 25 (11%) had major congenital malformations (7). Eighteen (72%) of these patients had cardiac anomalies; of these, one-third were Ebstein’s anomaly, which is a rare congenital heart defect with an incidence of 1 in 20,000 in the general population. On the basis of this information, lithium has become widely regarded as a human teratogen. Voluntary retrospective reports of reproductive outcome have many serious methodological shortcomings. In particular, there is evidence that bias toward reporting of adverse outcomes is common (8). With an unknown denominator (total number of exposures), an increased risk cannot be defined. Furthermore, two case-control studies (1,9) have not demonstrated increased use of lithium in mothers giving birth to children with Ebstein’s anomaly [although one of them reported Ebstein’s anomaly after exposure to lithium (9)], and a retrospective study (10) of 350 women with major affective disorders showed no difference in outcome between infants exposed to lithium and those who had been exposed to other psychotropic drugs, although four cardiac defects were seen in 59 lithium-exposed infants and none in the 38 controls (10). To evaluate the teratogenic potential of lithium in pregnancy, we conducted a controlled, prospective study of women using lithium who contacted teratogen information services in two American and two Canadian university programs (11). Patients and Methods We enrolled 148 women who called one of four teratogen information services to obtain information about the potential risks of therapeutic drugs during pregnancy: these centers were Motherisk (Toronto), the California Teratogen Information Service (CTIS), the Philadelphia Pregnancy Healthline, and Fetal Risk Assessment from Maternal Exposure (FRAME) (London, Ontario). The prospective collection of the study and follow-up data were consistent between centers. Motherisk and FRAME referred patients to a weekly clinic, where at interview a physician obtained information about drugs or other chemicals taken during or before pregnancy, including indication, dose, toxicity of the drug and its toxic effects, and monitoring. A medical and obstetric history was also elicited, as well as occupational exposures and family history. At this visit advice was offered and appropriate referrals were made. After the expected date of delivery, each woman was telephoned and a follow-up history of the pregnancy was obtained. The mother was asked for details about pregnancy outcome, perinatal complications, birth weight, physical findings, and developmental milestones. In addition, the physician caring for the baby was contacted to confirm this information and to provide a written report of the delivery and health status of the child. The CTIS and the Philadelphia Pregnancy Healthline obtained all initial information by telephone interview. In San Diego, postnatal follow-up was done by one of the investigators in the clinic, who collected details of delivery and postnatal course. The Philadelphia Pregnancy Healthline obtained all follow-up data by telephone; detailed records from physicians caring for the babies were also obtained. Patient enrollment into the study began with the initiation of each program: CTIS, 1979; Philadelphia Pregnancy Healthline, 1984; Motherisk, 1985; FRAME, 1989 until February 1991. All pregnant women who called and reported lithium ingestion during part or all of their first trimester were prospectively enrolled. Lithium exposures as early as 3 weeks’ gestation were included, since the half-life of lithium is long (2.4 days) in patients receiving long-term therapy (11). Thus detectable serum concentrations can still be present 2

Antidepressants in Pregnancy

87

weeks after cessation of treatment. All patients were offered fetal echocardiography at 18 weeks’ gestation to rule out cardiac anomalies. If prenatal echocardiography had not been done, infants were referred for postnatal echocardiographs. Patient enrollment was as follows: CTIS, 101; Motherisk, 25; Philadelphia Pregnancy Healthline, 19; and FRAME, 3. Controls were women who were seen at the Motherisk clinic for counseling about drugs that are not known or suspected to be teratogenic. Each study patient was matched with a woman of similar age (to within 2 years). Prenatal and postnatal evaluation was the same as described for the Motherisk program. Echocardiography was not done in this group, since this procedure was not deemed to be medically or ethically justifiable. We used Marden’s definition of a major anomaly—that is, one that has an adverse effect on either the function or social acceptability of the individual (13). We used chi-square analysis to compare the frequency of malformations in study and control groups and the frequency of smoking in the two groups. We calculated risk ratios and 95% confidence intervals (Taylor series) to establish the relation between lithium exposure and malformations, then for cardiac malformations and Ebstein’s anomaly. Birth weight, gestational age, and developmental milestones were compared with Student’s t-test for paired data. Correlations between values were studied by least-squares regression analysis. Results Of the 148 subjects, 10 were lost to follow-up postnatally; however, they were included in part of the analysis because all had had prenatal echocardiography. A total of 68 patients had echocardiographs (46%). Maternal age ranged from 15–40 years (mean 30, SD 5 years). All patients were receiving lithium for major affective disorders. The mean daily lithium dose was 927 (SD 340) mg; the range was 50–2400 mg. Pregnancy outcome did not differ between patients and controls with respect to the total number of live births, frequency of major anomalies, spontaneous or therapeutic abortions, ectopic pregnancy, and prematurity (Table 1). Three major congenital malforTable 1

Pregnancy Outcomea

Outcome

Lithium (n ⫽ 138)

Control (n ⫽ 148)

Normal live births Full-term Premature (⬍36 weeks) Congenital defects Spontaneous abortion Therapeutic abortionb Stillbirth Ectopic pregnancy Unknown

105 99 6 3 13 15 1 1 10

123 116 7 3 12 9 0 1 0

a

(76%) (72%) (44%) (3%) (9%) (10%)

(7%)

(83%) (785) (5%) (2%) (8%) (6%)

Ectopic pregnancies and spontaneous and therapeutic abortions are expressed as percentages of all outcomes (n ⫽ 148). Stillbirths and normal live births are expressed as percentages of known outcomes (n ⫽ 138). Congenital defects are expressed as a percentage of all live births. b One therapeutic abortion because of Ebstein’s anomaly diagnosed in utero.

88

Koren et al.

Table 2 Comparison of Sample Characteristics in Lithium-Exposed and Control Groups: Mean (SD) and Range

Birth weight, grams Gestational age, weeks Maternal age, years

Lithium group

Control group

p-value

3475 (660) (539–5024) 39.1 (3.0) (23–42.4) 30.0 (5.3) (15–40)

3383 (566) (950–4896) 39.2 (2.5) (26–43) 29.8 (5.3) (16–40)

0.026 0.562 0.614

mations occurred in each group: 2.8% of live births in the lithium group and 2.4% in controls. The 10 pregnancies for which final outcome was not known were not included in this part of the analysis. There were four sets of twins in the lithium group; one pair was born at 23 weeks’ gestation, and both infants died shortly after birth as a result of complications of prematurity. There was one set of twins in the control group. Two children in the lithium group had neural tube defects: one had hydrocephalus and meningomyelocele and had also been exposed to carbamazepine during the first trimester; the other had spina bifida and tethered cord. A third infant (a twin) was born at 23 weeks, had meromelia, and died shortly after birth. One fetus in the lithium group had a severe form of Ebstein’s anomaly; this was diagnosed at 16 weeks’ gestation and the pregnancy was terminated. This fetus had also been exposed to fluoxetine, trazodone, and l-thyroxine in the first trimester. In the controls, one child had a ventricular septal defect, one had congenital hip dislocation, and one had cerebral palsy and torticollis. The risk ratio for all congenital defects was 1.2 (95% confidence interval 0.2–5.7) when only live births were compared. When the case of Ebstein’s anomaly was included (since this pregnancy probably would have gone to term if the anomaly had not been detected), the risk ratio became 1.5 (0.4–6.7). The risk ratio for cardiac anomalies was 1.1 (0.1–16.6). The mean age at which postnatal follow-up was done in the lithium-exposed group was 61 (SD 87.5) weeks (range, 1 week–9 years). Lithium-exposed infants weighed a mean of 92 g more than controls ( p ⫽ 0.01) at birth (Table 2). Gestational age and frequency of prematurity did not differ between the groups. Because all controls were enrolled in Toronto, we examined whether the difference in birth weight was related to geographic factors in the various study groups. However, when the Motherisk data were analyzed separately, the difference in birth weight persisted and remained significant ( p ⫽ 0.018). There were more cigarette smokers among the women using lithium than among the controls (31.8 versus 15.5%; p ⫽ 0.002). Data on attainment of major developmental milestones (smiling, lifting head, sitting, crawling, standing, talking, and walking) were available for 21 lithium-exposed patients enrolled by Motherisk and 1 patient enrolled by FRAME. Study and control groups did not differ in age of attainment of any of the milestones (data not shown). Discussion Discontinuation of lithium therapy can have devastating results for the mental and physical health of young women with major affective disorders. On the other hand, many women

Antidepressants in Pregnancy

89

who inadvertently conceive while on lithium therapy choose to terminate pregnancy because of a perception of teratogenic risk based on the unfounded retrospective reports. As far as we are aware, this is the first prospective study of infants exposed to lithium during the first trimester of pregnancy. The recent establishment of teratogen information services has provided a very good opportunity to investigate large numbers of patients exposed to substances of interest within short periods. Our service is specifically designed to counsel pregnant women, and we are contacted directly by those exposed to lithium, almost invariably at the time of exposure. Accurate information about time of exposure and drug dosage is crucial to assess any causal relation between any xenobiotic and outcome. Moreover, information about other risk factors, such as illicit drugs and smoking, is obtained prospectively. Case-control studies rely on maternal recall, which can be poor and is subject to bias (14). Our results accord with those of earlier prospective cohort and case-control studies (1,9,10), suggesting that the risk of congenital defects in babies after lithium exposure is lower than that reported by the Danish Registry of Lithium Babies. The registry investigators claim that lithium is associated with a 10% incidence of cardiac malformations and a 3% incidence of Ebstein’s anomaly; they believe that the drug is a major teratogen. Because the registry is a voluntary, retrospective database, neither the number of cases not reported nor the number of unreported normal outcomes of lithium exposure is known. Our results cannot rule out an association between lithium and major anomalies, even though there was no difference in the number of anomalies between the study and control groups. Since Ebstein’s anomaly is very rare, a much larger sample size would be needed to define the real risk of this abnormality. That the single cardiac malformation in our study group was Ebstein’s anomaly indicates that lithium as a cause of this disorder is rare. Since fewer than half our patients had echocardiographs, some cardiac malformations might have been missed, including milder forms of Ebstein’s anomaly. This might also be true for the control group, in which no echocardiographs were done. Since all patients were examined by a physician, however, it is unlikely that clinically significant lesions were missed. The infant with hydrocephalus and meningomyelocele who was exposed to lithium in utero had also been exposed during the first trimester to carbamazepine, a drug that has been associated with an increased incidence of neural tube defects (14). The fetus that had Ebstein’s anomaly was also exposed in the first trimester to several other drugs, none of which has been associated with cardiac anomalies; lithium remains the most likely causative factor. The teratogenicity of lithium might be dose-related, as has been shown in animals (2). The doses used at the time the Danish registry data were obtained could have been higher than present recommended doses. This could explain the discrepancy between their results and ours. However, there is no evidence that lithium doses have changed during this time (16). In addition, lithium has a very narrow therapeutic window, above which toxic effects are common. Most patients are prescribed the highest dose they can tolerate that will maintain serum concentrations in the therapeutic range (15). We therefore believe that a dose relation does not account for the different findings of the two studies. The probable reason is the different methods of data collection used by the two groups. Although the difference in the number of elective abortions was not statistically significant between the groups, the rate was higher in the lithium group than in the controls

90

Koren et al.

(10 versus 6%). The increased trend to voluntary termination might have been due to the perceived teratogenic risk of lithium, since women who believe that they are at such risk are more likely to terminate their pregnancies (17). Counseling provided by the teratogen information services and referral for appropriate screening tests such as echocardiography might have reduced the perception of teratogenic risk of lithium. As a result, the number of terminations in our study was probably lower than it would have been; thus the figures for elective abortion are higher than ours in centers where no such counseling is provided. American Medicaid data indicate termination rates exceeding 50% in women exposed to lithium during pregnancy (F. Rosa, personal communication). The babies exposed to lithium were heavier than controls, and fetal macrosomia has been associated with lithium therapy (18,19). Yoder (19) proposed that lithium may have an insulin-like effect on carbohydrate metabolism. We did not find that demographic variations contributed to the difference in weight between the groups, and the proportion of diabetic patients did not differ. In addition, the proportion of cigarette smokers in the lithium group was twice that in the controls. One would therefore expect the babies in the lithium group to be smaller than those in the control group. Hence, the effect of the drug on birth weight might be even larger than is evident in this study. Although only a few babies were assessed for developmental milestones, no differences were seen, which is consistent with findings of earlier reports (20). We conclude on the basis of our results and those of others (1,9,10) that lithium is not a major human teratogen. We believe that women with major affective disorders who wish to have children may continue lithium during pregnancy and do not need to terminate pregnancy provided that level 2 ultrasound and fetal echocardiography are done to rule out the presence of major cardiac anomalies. Fluoxetine and Tricyclic Antidepressants During the First Trimester Fluoxetine (Prozac) is a new antidepressant that causes selective inhibition of neuronal serotonin (5-HT) uptake. Presently, the drug is used by millions of patients with major depression in North America. The manufacturer (Eli Lilly Ltd.) suggests that the drug not be given to women of childbearing potential, since the safe use of fluoxetine during human pregnancy has not been conclusively established. In rats and rabbits treated with 9 and 11 times the maximum daily human dose, fetal morphology and pregnancy outcome were not shown to be compromised (21). Experiential data have so far been limited to voluntary reports collected by the manufacturer. In a retrospective analysis of 38 women who reported first-trimester exposure to fluoxetine, there were 16 normal live births, 8 elective abortions, 3 miscarriages, 9 cases that were lost to follow-up, 1 infant born with hepatoblastoma, and 1 twin pregnancy with 1 miscarriage. In a postmarketing report summarizing the outcome of 226 prospective spontaneous reports of pregnant patients exposed to fluoxetine, outcome data were available for only 113 babies. There were 64 normal live births, 8 premature births, 18 elective abortions, 16 miscarriages, 3 twin pregnancies, and 1 infant born with transposition of the great arteries, although fluoxetine exposure occurred in the second trimester, well after embryological formation of the arteries (22). Voluntary reports collected by manufacturers regarding gestational exposures to drugs are fraught with major methodological problems. The original sources of these data are unknown and are not compared to appropriate control groups. Moreover, the quality

Antidepressants in Pregnancy

91

of ascertainment and follow-up is often poor, and—as documented above (22)—there is an almost 50% loss of follow-up data. There is ample evidence that abnormal outcome is more likely to be reported voluntarily; thus, in the absence of a denominator of all pregnant women treated with fluoxetine, it is impossible to ascertain the frequency of defects associated with this drug and to compare them with the baseline risk in the general population. The recommendation that women of reproductive age not conceive while on fluoxetine totally ignores that almost half the pregnancies in North America are unplanned. Indeed, during the last few years, teratology information services throughout North America have received hundreds of calls from pregnant women who conceived in an unplanned manner while on fluoxetine therapy. Following the success of the first collaboration by teratology information services that investigated the first-trimester use of lithium (11), we undertook to collect prospectively data on pregnancy outcome following fluoxetine in the first trimester and compare these to two control groups, one exposed to tricyclic antidepressants and the other to nonteratogens. Patients and Methods Our prospective collaborative study enrolled pregnant women who contacted one of four teratogen information services (TIS) requesting counseling about the teratogenic potential of fluoxetine. The participating centers were Motherisk (Toronto), Pregnancy Healthline (Pennsylvania), Pregnancy and Risk Information Service (New Jersey), and Pregnancy RiskLine (Utah). All women contacting these services during pregnancy were included. Prospective collection of information and follow-up data were consistent between centers, although collection was performed in different manners. Motherisk referred all women concerned about first-trimester fluoxetine exposure to a weekly clinic, during which information was obtained in an interview with a team physician regarding indication, dose, toxicity, and dates of initiation and discontinuation. Four women were referred to Motherisk by a geneticist outside Toronto who collected the cases prospectively in a similar manner. In addition, obstetrical, medical, genetic, and drug exposure history was obtained from both the mother and biological father of the fetus. Eight to 12 months after delivery details about the outcome of pregnancy, birth weight, presence or absence of birth defects, and perinatal and neonatal complications were documented. All follow-up information was corroborated by written documentation from the child’s physician. Pregnancy Hotline, Pregnancy and Risk Information Service, and Pregnancy RiskLine recorded maternal information similar to that gathered by Motherisk by telephone interviews. Postnatal followup data, similar to those collected in Toronto, were obtained by telephone in Pennsylvania, Utah, and New Jersey. The Pennsylvania Service also used follow-up cards received in the mail. Photocopies of each center’s fluoxetine cases were sent to Toronto for matching and statistical analysis after all names and addresses had been deleted. Each woman exposed to flf1flf1fluoxetine during the first trimester was age-matched (⫾2 years) to two controls closest in date to the date of consultation of the fluoxetine case. The first control group consisted of pregnant women who voluntarily sought counseling at Motherisk after first-trimester exposure to tricyclic antidepressants; the second control group consisted of pregnant women who voluntarily sought counseling at Motherisk regarding exposure to a nonteratogen. A nonteratogen is defined as a medication or environmental agent that has been proven in large studies not to increase the baseline teratogenic risk (e.g., acetaminophen,

92

Koren et al.

penicillins, dental x-rays). Both control groups were selected from our computerized database. Although the primary outcome of interest was the rate of birth defects in pregnancies exposed to fluoxetine as compared to the tricyclic and nonteratogen groups, maternal characteristics (age, obstetrical history, alcohol and cigarette use), pregnancy outcome (maternal weight gain, method of delivery, use of forceps, rates of live births, elective abortions, and miscarriages) and offspring characteristics (gestational age and birth weight) were also compared. Chi-square analysis and Fisher’s exact test were used to compare proportions, Student’s t-test was employed to compare paired data (two groups) and analysis of variance (three groups) to compare continuous data. Wilcoxon sign-rank (two groups) and Kruskal-Wallis (three groups) were procedures used to compare data that did not follow normal distribution. Data are expressed as mean plus or minus standard deviation. Results The participating centers followed prospectively a total of 128 pregnant women treated with fluoxetine for depression during the first trimester (45 in Toronto, 44 in Philadelphia, 21 in New Jersey, and 18 in Utah). The mean daily maternal dose of fluoxetine was 25.8 ⫾ 13.1 mg (range, 10–80 mg/day) (n ⫽ 122), and mean weight-adjusted dose (for 75 cases where maternal weight was available) was 0.19 ⫾ 0.1 mg/kg. The drug was taken during the first trimester by all 128 women, during the first and second trimesters by 2, and throughout the pregnancy by 6. Because of the limited number of patients taking tricyclic antidepressants in the Motherisk database suitable for matching (within 2 years of maternal age), our reported data are divided into comparisons between 128 fluoxetine cases and 128 age-matched nonteratogen controls (NTC), and comparisons among 74 fluoxetine cases, 74 age-matched tricyclic antidepressant (TCA) cases, and 74 age-matched NTC. There were no differences in any of the characteristics of the women using fluoxetine across the four participating centers. There was no difference in the rate of major birth defects when the live births exposed to fluoxetine were compared to the NTC live births (2 versus 1.8%, p ⫽ 0.38) or when the smaller fluoxetine groups was compared to both of its controls [3.4 versus 0 (TCA) versus 3% (NTC), p ⫽ 0.8] (Table 3). Table 4 details all major and minor birth defects as well as neonatal complications. There was no statistical difference in pregnancy outcome, maternal weight gain during pregnancy, gestational age at delivery, birth weight, or use of forceps at delivery whether the fluoxetine group was compared to NTC only or to both age-matched control groups (Table 3). Conversely, however, there was a tendency for a higher percentage of miscarriages in the 128 fluoxetine patients compared with NTC, but it did not reach statistical significance [14.8 versus 7.8% reported rate (RR), 1.9; 95% confidence interval (CI), 0.92– 3.92]. A similar trend was observed in the smaller sample size, although the TCA also had a miscarriage rate similar to the fluoxetine group (13.5 versus 12.2 versus 6.8%). There was no statistical difference between rates of vaginal or cesarean section deliveries between the fluoxetine group and age-matched controls. Infants born to depressive women of both groups (fluoxetine and TCA) tended to have more neonatal complications, although when looked at individually, none of those recorded was significantly more common (Table 4). Because one patient from Utah was gravida 33 (para 9, miscarriages 23, likely as a result of diagnosed maternal trisomy 8), obstetrical history data deviated from a normal Gaussian distribution and consequently nonparametric analysis was performed. Excluding

Pregnancy Outcome in Cases (Fluoxetine-Exposed) and Age-Matched Controls (TCA and NTC)

Characteristics Outcome Live birth Elective abortion Spontaneous abortion Major congenital anomalies Weight gain (kg) Gestational age (weeks) ⬍37 weeks ⬎42 weeks Birth weight (g) ⱖ4000 (g) Forceps used Delivery Vaginal Emergency C/S Repeat C/S

Fluoxetine (n ⫽ 128)

NTC (n ⫽ 128)

98/128 11/128 19/128 2/98 17.8 ⫾ 6.3 39.4 ⫾ 1.7 6/85 1/85 3459.7 ⫾ 660.2 15/84 8/38

110/128 8/128 10/128 2/110 16.3 ⫾ 6.0 39.4 ⫾ 1.8 7/85 2/85 3421.1 ⫾ 563 9/81 9/83

61/82 9/82 12/82

62/83 17/83 4/83

p value 0.14a 0.1b 0.3b 0.69 0.96 0.22b 0.38b 0.68 0.08b 0.13 0.04

Fluoxetine (n ⫽ 74)

TCA (n ⫽ 74)

NTC (n ⫽ 74)

p value 0.34a

58/74 6/74 10/74 2/58 17.2 ⫾ 6.2 39.4 ⫾ 1.6

60/74 5/74 9/74 0/60 16.8 ⫾ 5.8 39.1 ⫾ 2.3

67/74 2/74 5/74 2/67 15.9 ⫾ 5.3 39.6 ⫾ 1.9

0.31b 0.3b 0.53 0.40

3421.9 ⫾ 664.1

3515.9 ⫾ 672.3

3408.6 ⫾ 602.2

0.62

4/23

6/42

5/41

0.85

32/42 4/42 6/42

32/43 5/43 6/43

31/41 8/41 2/41

0.43

Antidepressants in Pregnancy

Table 3

Chi-square for 2 ⫻ 3 table. Fisher (two groups) and chi-square (three groups) (miscarriages versus live birth, excluding elective abortions). Abbreviations: NTC ⫽ nonteratogen control; TCA ⫽ tricyclic antidepressant; C/S ⫽ caesarean section. a

b

93

94

Koren et al.

Table 4 Major Malformations and Neonatal Complications Fluoxetine Major malformations

Neonatal complicationsa

a

Jejunal obstruction Ventricular septal defect Jaundice needing therapy (2) Shoulder dystocia and apnea (1) Patent ductus arteriosus and cyanosis (1) Sepsis and seizures (1) Hemangioma (2) Lacrimal stenosis (1) Aspiration pneumonia (1) Club feet (1) Congenital dislocation of hip (1)

Tricyclic antidepressants None

Metatarsus adductus (1) Congenital dislocation of hip (1) Slight hypotonia (1) β-Hemolytic streptococcus (1) Apnea (1) Hydrocele (1) Respiratory distress syndrome (1) Meconium aspiration and sepsis (1) Metatarsus varus (1)

Nonteratogens Pulmonary atresia Ventricular septal defect Jaundice (1) Clipped tongue (1)

The number of infants with the complication is indicated in parentheses.

this patient from the statistical analysis did not alter the results in any way. There was no statistical difference in gravidity, parity, and previous elective and previous spontaneous abortions between the three groups, although comparison of the fluoxetine and NTC groups revealed a higher mean gravidity and parity in the fluoxetine group (Table 5). There was a similar distribution of abstainers of ethanol and tobacco in all groups (Table 5). Discussion Many women of reproductive age suffer from depression, which necessitates chronic therapy with antidepressants. According to O’Hara et al., as many as 10% of pregnant women meet the criteria for major or minor depression (23). Exposing any fetus to drugs is perceived by most women as being associated with an increase in teratogenic risk (24), and women with psychiatric disorders are likely to be more prone to emotional instability due to unfounded perceptions of teratogenic risk following such exposures. While manufacturers and health providers often suggest that depressed women not be treated during gestation, such practice may endanger their health and hence the safety of their pregnancies. This reality, coupled with the fact that half of North American pregnancies are unplanned, implies that there is urgent need for credible data on the safety/risk ratio of antidepressant use in pregnancy. This need is in sharp contradistinction to the complacency exhibited by many pharmaceutical manufacturers in helping to collect such data and many of the regulatory agencies in enforcing such collection. Presently, systematic collection of postmarketing data on pregnancy exposure to drugs is not mandatory, and many manufacturers feel that their duties and responsibilities are met by disclaiming their product’s use in pregnancy, thus often orphaning pregnant women from essential drugs. Our study, like the earlier lithium project (11), demonstrates

Maternal Characteristics of Cases (Fluoxetine-Exposed) and Age-Matched Controls (TCA and NTC)

Characteristics Maternal age (years) Gestational age at first consultation (weeks) Obstetrical history Gravidity Parity Elective abortions Spontaneous abortions Ethanol Abstainers ⬍2.5 drinks per week ⱖ2.5 drinks per week Tobacco Abstainers ⬍0.5 package per day ⬍1 package per day ⱖ1, ⬍1.5 pack per day

Fluoxetine (n ⫽ 128)

NTC (n ⫽ 128)

p value

31.6 ⫾ 5.7

31.2 ⫾ 4.9

0.51

⫾ ⫾ ⫾ ⫾

0.01b 0.01b 0.3b 0.13b

2.9 0.9 0.5 0.5

⫾ ⫾ ⫾ ⫾

3.2 1.1 0.9 2.5

2.1 0.6 0.3 0.2

1.3 0.8 0.7 0.7

Fluoxetine (n ⫽ 74)

TCA (n ⫽ 74)

NTC (n ⫽ 74)

p value

31.7 ⫾ 5.2 8.5 ⫾ 4.9

31.6 ⫾ 5 8.6 ⫾ 6.3

31.0 ⫾ 4.3 8.9 ⫾ 5.8

0.66 0.93a

2.5 0.9 0.5 0.2

⫾ ⫾ ⫾ ⫾

1.6 0.9 0.8 0.6

2.3 0.7 0.4 0.4

⫾ ⫾ ⫾ ⫾

1.7 1 0.8 0.8

2.0 0.7 0.3 0.2

⫾ ⫾ ⫾ ⫾

1.2 0.8 0.6 0.4

⬎0.05c ⬎0.05c ⬎0.05c ⬎0.05c

62/109 35/109 12/109

68/125 49/125 8/125

0.31 0.8d

41/62 14/62 7/62

35/72 32/72 5/72

40/73 27/73 6/73

0.12 0.02d

74/105 17/105 11/105 3/105

93/125 21/125 6/125 5/125

0.42 0.61e

42/61 11/61 6/61 2/61

46/73 19/73 7/73 1/73

52/73 15/73 2/73 4/73

0.38 0.55e

Antidepressants in Pregnancy

Table 5

a

Analysis of variance. Wilcoxon sign-rank analysis. c Kruskal-Wallis analysis. d Chi-square analysis: abstainers versus admitted drinkers. e Chi-square analysis: abstainers versus admitted smokers. Abbreviations: NTC, nonteratogen control; TCA, tricyclic antidepressant. b

95

96

Koren et al.

the ability of TIS to prospectively collect and ascertain large cohorts of women exposed during the first trimester to specific medications. Because pregnant women contact these services at the time of exposure, complete documentation of exposure and other potential confounders is possible, as well as prospective follow-up of pregnancy. We have chosen to compare pregnant women exposed to fluoxetine to two control groups: one comprising women treated with tricyclic antidepressants, the selection of which was intended to control for potential effects of depression, and a second group of women who were exposed to nonteratogens, a group that should represent healthy pregnant women. All three groups were drawn from pregnant women counseled by TIS in an attempt to obviate potential bias introduced by different referral patterns. Prospective collection of pregnant women exposed to the drug in question and detailed ascertainment of the outcome of pregnancy have allowed us to construct a meaningful denominator and subsequently to calculate and compare rates of outcome measurements between the fluoxetine and control groups. Although this study is based on maternal reports, the Motherisk program routinely corroborates these reports with physicians’ written reports; the former have been found to be very accurate. Because the elimination halflife of fluoxetine ranges between 2 and 6 days, it is likely that many women were still exposed to the drug into their second trimester (e.g., five half-lives or 30 days after cessation of therapy). The rate of major malformations in live births exposed to fluoxetine during the first trimester was within the expected normal range (1–3%) and comparable to the two control groups. Although the strength of this study to rule out minimal increased risk above baseline is limited (n ⫽ 128), it is very unlikely that fluoxetine is a major human teratogen. The similar rate of major malformations in the control group indicates that the power of a much larger cohort would not likely be sufficient to identify a different trend. The sample size of this prospective cohort would have the power to detect a fourfold increased risk of major malformations, assuming a baseline risk of 2% in the NTC group, with a power of 80% and an α of 0.05. When compared to control women not exposed to teratogens, the fluoxetine group had a tendency for a higher rate of reported miscarriages; this did not reach statistical significance (14.8 versus 7.8%; RR 1.9, 95% CI 0.92–3.92). Our cohort has a limited power to show that an RR of 1.9 is significant; for such an RR to be statistically significant, more than 700 women would be needed in each group. The three-group comparison reveals that women exposed to TCA also had a similar tendency for a higher rate of reported miscarriages, which suggests that this tendency may be associated with the depression and/or the emotional instability or other putative biological changes associated with these psychiatric conditions. Another possible explanation is that when questioned during follow-up interviews, some women reported a miscarriage when in fact they had chosen to terminate their pregnancy. Our methodology did not have means to corroborate their reports. Our recent prospective study with lithium in pregnancy has shown similar miscarriage rates between the lithium and control groups (9.4 versus 8.1%, respectively; Table 6), which is similar to our present control group, whereas women exposed to lithium tended to electively terminate more often. This may suggest that the report obtained by the present study is genuine and that either the depressive condition or some of its therapies (fluoxetine and TCA) are inducing higher rates of miscarriage. To date, no other studies address this potential association between depression or its therapies and reported rates of miscarriage. Comparison of gestational age at first consultation revealed that the lower spontaneous abortion rate in the NTC group was not due to these patients contacting

Antidepressants in Pregnancy

97

Table 6 Rates of Spontaneous Abortion and Live Births in Women with Depression Exposed to Fluoxetine, Tricyclic Antidepressants, or Lithiuma

Fluoxetine Miscarriages Live births a

10/74 19/128 58/74 98/128

Tricyclic antidepressants (TCA)

Nonteratogen controls (NTC)

9/74

5/74

60/74

67/74 110/128

Lithium

Controls

13/138 10/128 105/138

12/148 123/148

NTC (n ⫽ 128) versus lithium controls, p ⫽ 0.19 (Fisher’s exact); fluoxetine (n ⫽ 74) versus TCA (n ⫽ 74) versus lithium, p ⫽ 0.76 (chi-square); fluoxetine (n ⫽ 128) versus lithium, p ⫽ 0.08 (Fisher’s exact); TCA (n ⫽ 74) versus lithium, p ⫽ 0.17 (Fisher’s exact).

Motherisk later in gestation compared to the fluoxetine and the TCA groups; the stages in pregnancy were identical [8.5 ⫾ 4.9 weeks (fluoxetine) versus 8.6 ⫾ 6.3 weeks (TCA) versus 8.9 ⫾ 5.8 weeks (NTC), p ⫽ 0.93] (Table 5). Very sparse published information exist on the potential teratogenicity of TCAs (25); the present prospective study failed to show this group as teratogenic. Women exposed to fluoxetine has an obstetric history as well as patterns of ethanol and tobacco use that were, in general, not different from those of the two control groups. Moreover, their characteristics are comparable to the general population of women contacting Motherisk. The mean dose of fluoxetine consumed by the participating women was within the recommended range; therefore it is unlikely that the lack of evidence of teratogenicity is due to a suboptimal dose. Neurodevelopment of Children Exposed in Utero to Antidepressant Drugs Despite the wide use of tricyclic antidepressant drugs and fluoxetine by women of reproductive age, the paucity of information on fetal effects has not allowed physicians to reassure women that either exposure during the first trimester or continuous therapy throughout gestation is safe. This lack of data has created anxiety among women planning pregnancies and pregnant women as well as among their families and physicians. Although neither tricyclic antidepressant drugs nor fluoxetine causes major malformations, the possibility of long-term damage to the developing central nervous system may deter women from taking these drugs even when clinically indicated. This study was designed to assess prospectively cognitive and language development and behavior in children exposed in utero to tricyclic antidepressant drugs or fluoxetine.

METHODS The Motherisk Program The Motherisk Program is an information and consultation service for women, their families, and health professionals regarding drug, chemical, radiation, and infectious exposures during pregnancy and lactation. Women with major depression who contact the program are invited to visit the clinic to be counseled by a physician.

98

Koren et al.

Selection of Patients We recruited three groups of mother-child pairs for this study. Two of the groups included the offspring of all women counseled by the program during the first trimester of their pregnancy regarding therapy with either tricyclic antidepressant drugs (since the inception of the Program in September of 1985) or fluoxetine (since its introduction in Canada in 1988). We excluded from the study any woman in whom antidepressant drug therapy had been discontinued before conception, if the woman was exposed to more than one antidepressant drug or to known teratogens during the pregnancy, or if the mother refused to participate in our follow-up program as described below. We also studied a control group of mothers not exposed to any drug, chemical, radiation, or infection known to affect the fetus adversely. The latter group, also collected prospectively, comprised women who had taken innocuous drugs such as acetaminophen or oral penicillin or had been exposed to dental x-rays. The control mothers were chosen from the list of women who had the closest visiting dates to the others in the study groups (⫾2 months). The study was approved by the Hospital’s Research Ethics Board and informed consent was provided by all women.

Assessments Antenatal Assessment During the initial assessment at or up to several weeks after pregnancy was diagnosed, we collected data on alcohol ingestion, medicinal and recreational drugs, smoking, lifestyle, medical and nutritional status, and sexually transmitted diseases. A detailed medical, genetic, and obstetric history was obtained from the mother and details concerning the time of exposure were recorded, as were the doses of tricyclic antidepressant drugs or fluoxetine and any concomitant drug therapy. Postnatal Assessment The first postnatal assessment occurred 6 to 9 months after delivery. During this interview the mother was questioned about the course of her pregnancy subsequent to the first meeting, including verification of the duration and dose of tricyclic antidepressant or fluoxetine treatment during gestation, maternal illnesses, and perinatal and postnatal complications. Details about the type of delivery, the perinatal period, and the times of developmental milestones were collected. This assessment also included a written report from the physician caring for the child. Neurobehavioral Testing All infants and children were assessed by a psychometrist who did not know the nature of the intrauterine exposure. To test for neurocognitive development, infants between 16 and 30 months of age were given the Baley Scales of Infant Development (26) and children above that age were tested with the McCarthy Scales of Children’s Abilities (27). The infant’s temperament and behavior were evaluated using the age-appropriate Carey Temperament Scale (28,29); in toddlers above 24 months, the age-appropriate Achenbach Behavior Checklist was used (30). Language skills were assessed in all infants and children with the Reynell Developmental Language Scales (31). Maternal IQ was assessed by the Wechsler Adult Intelligence Scale-Revised (32) and socioeconomic status with the Hollingshead Four Factor Inventory (33).

Antidepressants in Pregnancy

99

The mother’s level of depression and function from the birth of the infant to the time of the present assessment were assessed using the Global Assessment Scale, which rates the woman’s lowest level of functioning by selecting the lowest range that describes her functioning on a continuum of mental illness (34); the Center for Epidemiologic Studies Depressed Mood Scale (CES-D), a 20-item scale that measures depression symptoms for both epidemiological research and clinical purposes (35); and the Index of Parental Attitudes (IPA), a 25-item scale designed to measure the extent, severity, or magnitude of parent–child relationship problems as seen and reported by a parent (36). Maternal assessments were conducted at the time of examination of the offspring. As part of these assessments, we recorded whether the mother continued drug therapy in the postpartum period and for how long. The outcome measures in each group (tricyclic antidepressant drugs, fluoxetine, and control) were compared by one-way ANOVA and Tukeys-HSD. All tests were two-tailed. Subsequently, multiple regression analysis was conducted in order to determine the effects of potential confounders on the outcome measures. Differences in proportions among the groups were compared by the chi-square test. Results A total of 129 pregnant women taking a tricyclic antidepressant drug were counseled by the Motherisk Clinic since its inception in 1985; 24 were lost to follow-up, 8 refused follow-up, 3 were exposed to known teratogens, 12 reported spontaneous abortions, and 2 had therapeutic abortions. In total, we tested 80 infants exposed in utero to tricyclic antidepressant drugs; of the 80 mothers, 62 were treated for depression and 18 for other indications including migraine (9 women), pain (6 women), and bladder control (3 women). A total of 40 women took a tricyclic antidepressant drug during the first trimester, 36 throughout pregnancy, 2 during the first and second trimesters, and 2 during the first and third trimesters. A total of 29 women took amitriptyline, 20 imipramine, 10 clomipramine, 9 desipramine, 8 nortriptyline, and one each maprotiline, doxepin, amoxapine, and trimipramine. Since its introduction in Canada in 1988, 88 pregnant women taking fluoxetine were counseled. Of these, 6 were lost for follow-up, 8 refused participation, 12 had spontaneous abortions, and 7 had therapeutic abortions. This group, therefore, consisted of 55 infants, 37 of whose mothers had taken fluoxetine during the first trimester and 18 of whose mothers had taken the drug throughout pregnancy. The control group consisted of 84 mother-child pairs exposed to nonteratogenic drugs. The characteristics of the mothers in the three groups are shown in Table 7. The women in the fluoxetine group were of significantly higher gravidity and parity, had more previous therapeutic abortions, and had a lower mean socioeconomic status. The women of both antidepressant drug groups tended to consume more alcohol and smoke more cigarettes during the index pregnancy than women in the control group. The women in the groups taking tricyclic antidepressant drugs and fluoxetine had similar scores on the three tests used to quantify their levels of depression and function since after the birth of the index child (Table 7). At birth and at the time of testing, the weight, height, and head circumference percentiles of the infants in the three groups were similar (Table 8), and there were no differences in the rates of perinatal complications. The incidence of major malformations among the three groups was also similar; three exposed to tricyclic antidepressant drugs (ventricular septal defect, hypospadias, and pyloric stenosis), two exposed to fluoxetine (ventricular

Table 7 Characteristics of Women Who Were Taking Tricyclic Antidepressant Drugs or Fluoxetine During Pregnancy and Pregnant Control Subjectsa Tricyclic antidepressant drugs (n ⫽ 80)

Fluoxetine (n ⫽ 55)

Control (n ⫽ 84)

31 ⫾ 4 2⫾1

31 ⫾ 4 3⫾2

30 ⫾ 5 2⫾1

1⫾1

1⫾1

Previous spontaneous abortion Previous therapeutic abortion Weight gain during (kg) Socioeconomic status score

0.3 ⫾ 0.6

0.2 ⫾ 0.5

0.39 0.001 (fluoxetine vs. the other two groups) 0.4 ⫾ 0.6 0.001 (control vs. the other two groups) 0.2 ⫾ 0.4 0.61

0.3 ⫾ 0.6

0.6 ⫾ 1.1

0.1 ⫾ 0.4

16 ⫾ 6.5 46 ⫾ 12

16 ⫾ 8 40 ⫾ 13

14 ⫾ 6 44 ⫾ 13

IQ Alcohol use during pregnancy (no. of women)c None Light Heavy Cigarette smoking during pregnancy (no. of women)d None Light Heavy Severity of depressione Global Assessment Scale CES Depressed Mood Scale Score on best days Score on worst days Index of Parental Attitudes

100 ⫾ 14

97 ⫾ 13

97 ⫾ 14

36 44 0

19 34 2

56 27 0

0.001 (control vs. the other two groups)

54 25 1

29 25 1

65 18 1

0.001 (control vs. the other two groups)

62 ⫾ 15

60 ⫾ 15

—

0.10

10 ⫾ 9 33 ⫾ 16 11 ⫾ 10

13 ⫾ 10 35 ⫾ 15 11 ⫾ 9

— — —

0.08 0.48 0.81

Characteristic Age at conception (years) Gravidity Parity

p valueb

0.001 (fluoxetine vs. the other two groups) 0.39 0.04 (fluoxetine vs. the other two groups) 0.34

Plus-minus values are means ⫾ SD. The p values correspond to the overall heterogeneity of the three groups, or in the case of severity of maternal depression, the two drug groups, according to Tukey’s multiple-range test. Significant differences between specific groups are noted in parentheses. c Light alcohol ingestion was defined as the consumption of up to two drinks per week and heavy alcohol ingestion as the consumption of more than two drinks per week. Information was not available for one woman in the control group. d Light cigarette smoking was defined as smoking of up to five cigarettes per day and heavy cigarette smoking as smoking of more than five cigarettes per day. e The score on the Global Assessment Scale can range from 1 (indicating the need for constant supervision) to 100 (indicating normal function). The score on the Center for Epidemiologic Studies (CES) Depressed Mood Scale can range from 0 (normal) to 60 (severe depression). The score on the Index of Parental Attitudes can range from 0 (no problems) to 100 (major attitudinal problems); scores above 30 are considered to indicate clinically important problems. a

b

Antidepressants in Pregnancy

101

Table 8 Physical Characteristics of Infants at Birth and at the Time of Testing According to Whether They Were Exposed In Utero to Antidepressant Drugsa

Characteristic Gestational age at birth (weeks) Birth weight (g) Weight at testing (percentile) Height at testing (percentile) Frontooccipital circumference at testing (percentile) a b

Tricyclic antidepressant drugs (n ⫽ 80) 39 3490 58 52 46

⫾ ⫾ ⫾ ⫾ ⫾

2 642 30 29 26

Fluoxetine (n ⫽ 55) 39 3567 54 47 45

⫾ ⫾ ⫾ ⫾ ⫾

2 683 30 27 28

Control (n ⫽ 84)

p valueb

⫾ ⫾ ⫾ ⫾ ⫾

0.10 0.18 0.32 0.62 0.80

40 3373 51 49 48

1 540 31 31 27

Plus-minus values are means ⫾ SD. Testing occurred between 16 and 86 months of age (mean, 33 ⫾ 14). The p values correspond to the overall heterogeneity of the three groups.

septal defect and patent ductus arteriosus), and two among the controls (cyanotic heart disease and ventricular septal defect). The mean global IQ values in the three groups of infants were similar, both for younger children (tested with the Bayley’s scales) and older ones (tested with the McCarthy test) (Table 9). After adjustment for independent variables that may affect language development, the scores of the Reynell verbal or expressive tests were similar in the three groups. Regarding temperament, there was no difference in temperament styles in either the group taking tricyclic antidepressant drugs or the fluoxetine group when compared to the controls. Similarly, there were no differences in scores of mood, arousal functions, activity level, distractibility, or behavior problems. Multiple regression analysis of the effect of duration of antidepressant drug therapy (first trimester versus the whole pregnancy) revealed no differences for either infants exposed to tricyclic antidepressant drugs or fluoxetine as compared with the control infants on any of the neurobehavioral tests. In the tricyclic antidepressant drug group, 18 women were treated for conditions other than depression. Their offsprings’ IQ, Reynell scores, and scores of temperament, mood, arousal, activity, or distractibility were not different from those of the rest of the group (data not shown).

DISCUSSION In planning our study, we wished to address potential confounders that may affect infant achievements on standard neurocognitive development and behavioral tests regardless of the mother’s therapy, including maternal IQ and socioeconomic class. Because the depressed mother usually raises the infant, the infant’s emotional, cognitive, and behavioral development may be adversely affected as a result of postnatal mother-child interactions. For example, disturbances in affect regulation, attachment, temperament difficulties, depression, and cognitive and other developmental achievement have all been described in infants of depressed mothers (37). To address these potential confounders, we quantified the mother’s depression and her resulting function using widely used research tools (35– 37). Despite wide variability in the degree of depression, the mean scores of the women

102

Table 9

Results of Neurobehavioral Tests in Infants According to Whether They Were Exposed In Utero to Antidepressant Drugsa Adjusted differences (95% CI)c

Testb Bayley Mental Developmental Index McCarthy General Cognitive Index Reynell Verbal Comprehension Scale Reynell Expressive Language Scale

Tricyclic antidepressant drugs (n ⫽ 80)

Fluoxetine (n ⫽ 55)

⫾ ⫾ ⫾ ⫾

117 ⫾ 17 114 ⫾ 16 1.2 ⫾ 1.2 ⫺0.2 ⫾ 1.0

118 117 1.3 0.3

17 10 0.8 0.9

Control (n ⫽ 84)

Tricyclic antidepressant drugs vs. control

Fluoxetine vs. control

⫾14 ⫾ 13 ⫾ 0.9 ⫾ 1.0

2.4 (⫺4.5 to 9.4) 2.7 (⫺2.3 to 7.6) 0.3 (⫺0.1 to 0.5) 0 (⫺0.3 to 0.3)

2.1 4.7 0.3 ⫺0.1

115 114 1.1 0.1

(⫺5.0 (⫺4.0 (⫺0.1 (⫺0.4

to to to to

9.2) 13.4) 0.6) 0.3)

Plus-minus values are means ⫾ SD. The children were tested between 16 and 86 months of age (mean, 33 ⫾ 14). The Bayley and McCarthy scores are typical for this age. The normal range for both tests is 100 ⫾ 1 SD. Lower scores mean lower cognitive function. The mean Reynell score in normal children of this age is 0 ⫾ 1 (range of possible scores, ⫺3 to ⫹3). c Multiple regression analysis was used after adjustment for children’s age; maternal IQ, socioeconomic status, score on the Center for Epidemiologic Studies Depressed Mood Scale, and score on the Global Assessment Scale; and duration of exposure to drug (first trimester versus entire pregnancy). CI denotes confidence interval. a

b

Koren et al.

Antidepressants in Pregnancy

103

receiving tricyclic antidepressant drugs and those receiving fluoxetine were similar on the three tests. In a like manner, other maternal factors known to affect child achievements in standard cognitive tests were quantified by us, including socioeconomic status and maternal IQ. We failed to find any differences in cognitive, language, and behavioral development among the children exposed to tricyclic antidepressant drugs and fluoxetine as compared with the unexposed control group. Because half of the pregnancies in North America are unplanned, many women take a tricyclic antidepressant drug or fluoxetine during the first few weeks of their pregnancy. However, the reassurance available to women that firsttrimester exposure to these drugs will not affect brain development in their unborn babies is not likely to reassure those who need continuous pharmacotherapy throughout gestation. When maternal depression is not optimally controlled, there is ample evidence for adverse outcome in infants and young children in a variety of domains, including cognitive, language, and behavioral development as well as higher rates of perinatal risk (38). In our study, one-third of the women taking fluoxetine and almost half of those taking a tricyclic antidepressant drug continued the medication throughout pregnancy, allowing us to compare first-trimester to whole-pregnancy exposure to these drugs. We found that exposure to either tricyclic antidepressant drugs or fluoxetine throughout gestation did not affect the IQ or language and behavioral development of the offspring as measured during preschool years. In summary, in utero exposure to either tricyclic antidepressant drugs or fluoxetine does not adversely affect child neurodevelopment. Answer Lithium has a minimal teratogenic risk for the rare Einstein’s anomaly. Fetal echocardiography can help rule out severe cases of this malformation before 20 weeks of gestation.

REFERENCES 1. Zalstein E, Koren G, Einarson T, et al. A case-control study on the association between first trimester exposure to lithium and Ebstein’s anomaly. Am J Cardiol 1990; 65:817–818. 2. Smithberg M, Dixet PK. Teratogenic effects of lithium in mice. Teratology 1982; 26:239–246. 3. Gralla EJ, McIlhenny HM. Studies in pregnant rats, rabbits and monkeys with lithium carbonate. Toxicol Appl Pharmacol 1972; 21:428–433. 4. Nora JJ, Nora AH, Toews WH. Lithium, Ebstein’s anomaly, and other congenital heart defects. Lancet 1974; 2:594–595. 5. Long WA, Park WW. Maternal lithium and neonatal Ebstein’s anomaly: evaluation with crosssectional echocardiography. Am J Perinatol 1984; 1:182–184. 6. Fries H. Lithium in pregnancy. Lancet 1970; 1:1233. 7. Frankenberg FR, Lipinski JF. Congenital malformations. N Engl J Med 1983; 309:311–312. 8. Koren G. Retinoid embryopathy. N Engl J Med 1986; 315:262. 9. Kallen B. Comments on teratogen update: lithium. Teratology 1988; 38:597–598. 10. Kallen B, Tandberg A. Lithium and pregnancy. Acta Psychiatr Scand 1983; 68:134–139. 11. Jacobson SJ, Jones K, Johnson K, et al. Prospective multicentre study of pregnancy outcome after lithium exposure during first trimester. Lancet 1992; 339:530–533. 13. Maren PM, Smith DW, McDonald MJ. Congenital anomalies in the newborn infant, including minor variations. J Pediatr 1964; 64:357–371. 14. Koren G. Teratogenic drugs and chemicals in humans. In: Koren G, ed. Maternal-Fetal Toxicology. New York: Marcel Dekker, 1990, p 17.

104

Koren et al.

15. Rosa F. Spina bifida in infants of women treated with carbamazepine during pregnancy. N Engl J Med 1991; 324:674–677. 16. Goodman L, Gilman A, eds. The pharmacological Basis of Therapeutics, 7th ed. New York: Macmillan, 1985, p 429. 17. Koren G, Bologa M, Pastuszak A. The way women perceive teratogenic risk: the decision to terminate pregnancy. In: Koren G, ed. Maternal-Fetal Toxicology. New York: Marcel Dekker, 1990, pp 373–381. 18. Belik J, Yoder M, Pereira GR. Fetal macrosomia: an unrecognized adverse effect of maternal lithium therapy. Pediatr Res 1983; 17:304A. 19. Yoder MC, Belik J, Lannon RA, et al. Infants of mothers treated with lithium during pregnancy have an increased incidence of prematurity, macrosomia and perinatal mortality. Pediatr Res 1984; 18:404A. 20. Schou M. What happened later to the lithium babies? Acta Psychiatr Scand 1976; 54:193–197. 21. Byrd RA, Brophy GT, Markham JK. Developmental toxicology studies of fluoxetine hydrochloride administered orally to rats and rabbits. Teratology 1989; 39:444. 22. Goldstein DJ. Outcome of fluoxetine-exposed pregnancies: proceedings of the Fourth International Conference of Teratogen Information Services, Chicago, April 18–20, 1991. 23. O’Hara MW, Neunober DJ, Zekoski GH. Prospective study of postpartum depression: prevalence and predictive factors. J Abnorm Psychol 1984; 93:158–171. 24. Koren G, Pastuszak A. Prevention of unnecessary pregnancy terminations by counselling women on drug, chemical and radiation exposure during the first trimester. Teratology 1990; 41:657–661. 25. Cohen LS, Heller VL, Rosenbaum JF. Treatment guidelines for psychotropic drug use in pregnancy. Psychosomatics 1989; 30:25–33. 26. Bayley N. Bayley Scales of Infant Development. 2nd ed. San Antonio, TX: Psychological Corporation, 1993. 27. McCarthy D. McCarthy Scales of Children’s Abilities. New York: Psychological Corporation, 1972. 28. Fullard W, McDevitt S, Carey W. Toddler Temperament Scale for 1- to 3-Year-Old Children. Philadelphia: Temple University Press, 1978. 29. McDevitt SC, Carey WB. The measurement of temperament in 3–7 year old children. J Child Psychol Psychiatry 1978; 19:245–253. 30. Achenbach TM. Manual for the Child Behavior Checklist/4–18 and 1991 profile. Burlington, VT: University of Vermont Department of Psychiatry, 1991. 31. Reynell JK. Reynell Developmental Language Scales Manual. 2nd ed. Windsor, England: NFER-NSON, 1985. 32. Wechsler D. Wechsler Adult Intelligence Scale, rev ed. New York: Psychological Corporation, 1981. 33. Hollingshead AB. Four factor index of social status. New Haven, CT: Yale University Department of Sociology, 1975. 34. Endicott J, Spitzer RL, Fleiss JL, Cohen J. The Global Assessment Scale: a procedure for measuring overall severity of psychiatric disturbance. Arch Gen Psychiatry 1976; 33:766–771. 35. Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977; 1:385–401. 36. Hudson WW. The Clinical Measurement Package: A Field Manual. Homewood, IL: Dorsey Press, 1982. 37. Goodman SH. Understanding the effects of depressed mothers on their children. Prog Exp Pers Psychopathol Res 1992; 15:47–109. 38. Morrison HL, ed. Children of depressed parents: risk, identification, and intervention. New York: Grune & Stratton, 1983.

7 Benzodiazepine Use in Pregnancy and Major Malformations or Oral Cleft: Meta-Analysis of Cohort and Case-Control Studies Lisa R. Dolovich, Gideon Koren, and J. M. Re´gis Vaillancourt The Hospital for Sick Children, Toronto, Ontario, Canada

Antonio Addis Istituto di Ricerche Farmacologiche Mario Negri, Milan, Italy

J. D. Barry Power Ottawa, Ontario, Canada

Thomas R. Einarson The University of Toronto, Toronto, Ontario, Canada

INTRODUCTION Benzodiazepines are commonly used for anxiety, insomnia, and epilepsy. Their use is substantial, even by pregnant women. Bergman et al. found that 2% of pregnant women in the United States who were receiving Medicaid benefits filled one or more prescriptions for benzodiazepines during pregnancy (1). As about half of pregnancies in the United States are unplanned (2), many women may inadvertently expose the fetus to benzodiazepines during the first trimester. Therefore, women require valid information regarding the risks of benzodiazepine use during pregnancy to avoid exposure to teratogens but also to ensure that they are not denied medication during pregnancy because of unfounded fear of unknown consequences. Antepartum exposures to benzodiazepines have been associated with teratogenic effects (for instance, facial cleft, skeletal anomalies) in some animal studies (3,4) but not others (5,6). Early case-control studies in humans found that maternal benzodiaze-

Reprinted from the British Medical Journal 1998; 317:839–843.

105

106

Dolovich et al.

pine exposure increased the risk of fetal cleft lip and cleft palate (7,8). Subsequent reports implicated benzodiazepines as the cause of major malformations (9–11) and a benzodiazepine syndrome similar to fetal alcohol syndrome (9,12,13). Numerous studies, however, have refuted these findings (1,14–16). These contradictory results have led to considerable controversy surrounding the use of benzodiazepines in pregnancy. We carried out a meta-analysis to examine whether exposure to benzodiazepines during at least the first trimester is associated with increased risk of major malformations or oral cleft. METHODS Data Sources We systematically searched Medline (1966 to December 1997 via Ovid), Embase (1980 to December 1997), Reprotox (a database of reviews on reproductive toxicity topics), references in textbooks on drugs in pregnancy, references of included studies, and review articles. Benzodiazepine(s) (exploded as a subject heading or the various preparations put in as text words) was combined with the following words as subject headings or text words: fetal diseases, infant, fetal organ maturity, cleft lip, cleft palate, major malformations, and prenatal exposure. The Toronto-based Motherisk Program—a consultation service for drug, chemical, and radiation exposure during pregnancy—helped to locate unpublished papers and provided one unpublished study and one abstract. The original authors provided unpublished data. Study Selection Searches were reviewed or completed independently and in duplicate. Cohort or casecontrol studies in any language considered pertinent were retrieved and included if they examined the relation between human maternal exposure to benzodiazepines in at least the first trimester and major malformations or oral cleft alone and included an unexposed concurrent control group. Major malformations were those described by Heinonen et al., which, among others, include cleft palate and cleft lip (17). Hereafter oral cleft is used for cleft lip or cleft palate, or both. Studies examining only certain subtypes of malformations or studies in patients with epilepsy were included but considered separately from the main analysis. Only studies where exposure occurred during the first trimester were considered as the fetus is most susceptible to teratogens during the period between the first and eighth weeks of organogenesis, the lip forms between weeks 4 and 8, and the oral palate forms between weeks 5 and 12. Studies were excluded if they were case series or reports, editorials, reviews, animal studies or used only stillbirths or abortions or the data could not be extracted. All published studies deemed suitable were retrieved. Unpublished studies were treated methodologically in the same way as published studies. The methods sections with study identifiers removed were reviewed independently and in duplicate to determine inclusion. Consensus or a third party whose decision was final resolved disagreements. Data Extraction Once the study was included data were extracted and quality assessed independently and in duplicate. Discrepancies were resolved through consensus. Study quality was assessed

Use of Benzodiazepines in Pregnancy

107

by using predetermined criteria. These aspects of study quality are provided in the results section as descriptive information. Data Analysis Studies of different design—namely, cohort and case-control studies—were analysed separately because of differing threats to their internal validity (18,19). Data were analysed by calculating the odds ratio and 95% confidence interval with a random effects model (20). We also calculated χ 2 tests for heterogeneity (21). Further sensitivity analyses were performed for case-control studies to assess the impact of recall bias through the use of normal babies compared with malformed babies as controls. In a further examination of homogeneity of effects we plotted the data with the rates of malformations in the control groups on the X axis and in the exposed subjects on the Y axis as suggested by L’Abbe´ et al. (22). We first visually inspected the plot for evidence of obvious outliers. We then regressed the malformation rate in the exposed group on that of the controls. The slope of that regression line was compared with the null hypothesis (that is, a slope of 1) by using standard techniques as a test for effects. We also examined residuals to determine if any observations were outliers (that is, ⬎1.96 SE) from a statistical point of view. Publication bias was examined through visual inspection of a funnel plot whereby odds ratios were plotted against study sample size.

RESULTS Over 1400 studies were considered. Most were not retrieved because two independent reviewers considered that they did not relate to the question under review. Of the studies considered, 74 studies were retrieved and 51 of these were excluded. Studies were excluded because they had no concurrent control group (eighteen), they did not examine major malformations (nine), they were carried out on animals (one), benzodiazepines were not studied (one), exposure was not during the first trimester (one), studies were review articles or commentaries (five), data presented were duplicated in an included trial (six), results for benzodiazepines were not reported separately from other agents (five), only a range of results were reported (one), only partial data were provided (two), benzodiazepine exposure was not linked to malformations (one), and the study was not available in North America (one). A complete list of excluded studies is available from the authors. Thirteen studies that examined major malformations (11,13,14,23–32), 11 studies that examined oral cleft alone (1,7,10,13,27,30,31,33–36), and three studies that examined other specific malformations (37–39) were included (some providing information for more than one evaluation). One study unpublished at the time of consideration has since been published (26). Of the 23 included studies, 20 (87%) predefined exposure (1,10,11,13,14,23,25– 28,30–39) and 22 (96%) predefined the outcome (1,7,10,11,13,14,23,25–32,34–40). Exposure was ascertained mainly through interview with the mother (61% of studies) (7,10,14,25–30,33,34,36–38) and outcome was confirmed mainly by using physician examination or records (44% of studies) (11,13,14,25,28,30,32,34,39,40) or malformation registries (30% of studies) (7,10,29,31,35–37). Equal diagnostic examination between exposed and unexposed groups occurred in all but three studies (14,25,35). Hartz et al. gathered and confirmed information about malformed babies from different sources but

108

Dolovich et al.

did not do so for control babies (14). Czeizel et al. sent surveys to up to three controls if initial controls did not respond (35). Laegreid et al. used blood samples to confirm benzodiazepine exposure and had blood sample results for 78% of cases but only 66% of controls (25). Various benzodiazepines were used or prescribed, although 48% of the studies (11 of 23) examined the use of chlordiazepoxide or diazepam only (10,11,14,23,28,30,32– 34,37,40). Only two studies provided any information regarding the duration of maternal exposure (25,26). The indications for use were infrequently provided (11,25,26,41). Sixtyone percent (14 of 23) of studies reported concurrent use of at least some prescription medications (1,10,11,13,25,26,28–32,37–39). Associations with Major Malformations Data pooled from seven cohort studies did not show an association between fetal exposure to benzodiazepines during pregnancy and major malformations (odds ratio 0.90; 95% confidence interval 0.61–1.35; homogeneity χ 2 ⫽ 1.74; p ⫽ 0.62; Table 1, Fig. 1) (11,14,23– 27). Two cohort studies carried out in patients with epilepsy were located. Results of both were not significant (30,31). Combination of four case-control studies showed that major malformations were associated with the use of benzodiazepines during pregnancy (3.01; 1.32–6.84; χ 2 ⫽ 9.87;

Table 1 Association of Major Malformations in Fetuses with Prenatal Benzodiazepine Exposure Exposed First author (year) Cohort studies Nonepileptic patients: Milkovich (1974) (11) Crombie (1975) (23) Hartz (1975) (14) Kullander (1976) (24) Laegreid (1992) (25) Pastuszak (1996) (26) Ornoy (1997) (27) Combined effect Epileptic patients: Nakane (1980) (30) Robert (1986) (31) Case-control studies Greenberg (1977) (29) Bracken (1981) (28) Noya (1981) (32) Laegreid (1990) (13) Combined effect a b

χ 2 ⫽ 1.74; p ⫽ 0.62. χ 2 ⫽ 9.87; p ⫽ 0.008.

Not exposed

No malformed

Total

No malformed

Total

5 3 11 2 1 1 9

86 200 257 89 17 106 335

10 382 2,179 198 1 3 10

229 19,143 46,233 5,664 29 115 363

16 0

117 4

42 8

490 144

36 39 1 8

60 72 24 10

800 1,331 0 10

1,612 4,266 24 68

Odds ratio (95% CI)

1.35 0.75 0.90 0.63 1.75 0.36 0.97 0.90

(0.45–4.07) (0.24–2.35) (0.49–1.66) (0.16–2.60) (0.10–29.92) (0.04–3.47) (0.39–2.43) (0.61–1.35) a

1.69 (0.91–3.13) 1.78 (0.09–35.94) 1.52 2.61 3.13 23.20 3.01

(0.9–2.58) (1.63–4.16) (0.12–80.68) (4.29–125.55) (1.32–6.84) b

Use of Benzodiazepines in Pregnancy

109

Figure 1 Association of major malformations with prenatal exposure to benzodiazepines.

p ⫽ 0.008) (13,28,29,32). All included case-control studies that evaluated major malformations used normal babies as controls so subgroup analyses based on types of controls could not be done. Regression analyses for both cohort and case-control studies showed no obvious heterogeneity. Associations with Oral Cleft Data pooled from three cohort studies showed no relation between fetal exposure to benzodiazepines during pregnancy and oral cleft (1.19; 0.34–4.15; χ 2 ⫽ 0.01; p ⫽ 0.997; Table 2, Fig. 2) (1,27,33). The analysis of six case-control studies produced a significant odds ratio for oral cleft of 1.79 (1.13–2.82; χ 2 ⫽ 11.39; p ⫽ 0.01) (7,10,13,34–36). Subgroup analysis of the case-control studies with normal babies as controls showed no significant association with oral cleft (1.63; 0.89–2.96; χ 2 ⫽ 3.81; p ⫽ 0.15) (7,13,35). Similarly, no significant association was found in analyses of case-control studies with malformed babies as controls (2.03; 0.88–4.71; χ 2 ⫽ 6.90; p ⫽ 0.10) (10,34,36). Regression analyses for both cohort and case-control studies showed no obvious heterogeneity. In general, for the analyses of major malformation and oral cleft the risks for casecontrol studies were grouped at a different end of the distribution than the risks for cohort studies, showing that the relative risks within each study design are of the same magnitude but the absolute differences in risk are of a different order of magnitude between studies (case-control about 10 times greater than cohort). This finding suggests a possible systematic difference between study designs. Funnel plot analyses produced funnel shaped plots, indicating that there was no obvious publication bias. Two case-control studies examined the association of benzodiazepine use with fetal cardiac malformations. One did not show an association between exposure and outcome; the other did (37,38). One study examined benzodiazepine use with malformations of the central nervous system and did not find any association between exposure and outcome (39).

110

Dolovich et al.

Table 2 Results of Studies Examining Association of Specific Malformations with Prenatal Exposure to Benzodiazepines Exposed Author (year) Cohort studies (oral cleft) Nonepileptic patients Shiono (1984) (33) Bergman (1992) (1) Ornoy (1997) (27) Combined effect Epileptic patients Nakane (1980) (30) Robert (1986) (31) Case-control studies Oral cleft Safra (1975) (10) Saxen (1975) (7) Rosenberg (1983) (34) Rodriguez (1986) (36) Czeizel (1987–88) (35) Laegreid (1990) (13) Combined effect Cardiac malformations Tikkanen (1992) (37) Correa-Villasenor (1994) (38) Malformations of central nervous system Winship (1984) (39) a b

Not exposed

No No malformed Total malformed

Total

32,395 102,985 363

Odds ratio (95% CI)

1 0 0

854 1,354 335

31 62 0

1.22 1.21 1.08 1.19

(0.17–8.98) (0.17–8.71) (0.07–17.39) (0.34–4.15) a

3 0

117 4

12 1

7 27 13 8 48 2

16 40 67 61 91 10

42 511 590 442 1,153 4

262 1,044 3,011 7,990 2,311 68

4.07 2.17 0.99 2.58 1.12 4.00 1.79

2 57

10 92

404 3,318

1,152 6,855

0.46 (0.10–2.19) 1.74 (1.14–2.65)

14

750

14

750

1.00 (0.47–2.11)

490 1.05 (0.29–3.78) 144 10.63 (0.38–298.57)

(1.44–11.54) (1.11–4.24) (0.54–1.82) (1.22–5.45) (0.74–1.71) (0.63–25.43) (1.13–2.82) b

χ 2 ⫽ 0.01; p ⫽ 0.997. χ 2 ⫽ 11.39; p ⫽ 0.01.

DISCUSSION Data taken from cohort studies showed no significant association between benzodiazepines taken during the first trimester and either major malformations or malformations of the oral cleft alone. However, data from case-control studies showed a small but significant increased risk for these events. This finding may reflect the substantially higher sensitivity of case-control studies to examine the risk of specific malformations or it may be chance. The tests of heterogeneity also showed that the cohort studies were not heterogeneous for both major malformation and oral cleft, whereas the case-control studies for oral cleft were heterogeneous, which decreases the reliability of these marginally significant results. A case series of eight children exposed to benzodiazepines in utero suggested the existence of a benzodiazepine syndrome (9). This syndrome was described as dysmorphic features, growth aberrations, and abnormalities of the central nervous system (9,12,13). Our results, however, do not confirm the presence of this syndrome. Even before this

Use of Benzodiazepines in Pregnancy

111

Figure 2 Association of oral cleft with prenatal exposure to benzodiazepines.

report alternative causes for these findings, such as Zellweger syndrome or other genetic abnormalities, have been suggested (42).

Possible Confounding and Bias Concomitant exposure to other medications can result in an overestimation of the risk of benzodiazepines. Fourteen studies, eight of which were case-control, allowed exposure to other potentially teratogenic medications (1,10,11,13,25,26,28–32,37–39). This large number confounds the results. In most studies no information on duration or indication for use of benzodiazepines was provided. Therefore it was difficult to determine if any of the populations included have an increased or decreased risk of major malformations. Studies that evaluated the risk of fetal malformations in women with epilepsy were separated from the main analysis as fetuses born to such women already have an increased risk of major malformations (43). Information is lacking regarding the risk of developing specific malformations. The use of a normal baby as a control in a case-control study can produce recall bias, as mothers of malformed babies may be more likely to recall exposures than mothers of normal babies. The subgroup analysis that compared benzodiazepine use in mothers of healthy babies as controls compared with mothers of malformed babies as controls produced similar effect sizes, suggesting that recall bias did not have a large effect on study outcome. The present meta-analysis has several limitations. The number of reports was relatively small and may have limited the power of our analysis. Also, although the overall sample was large, most cases for analyses of both oral cleft and major malformations were derived from only three studies (1,14,23). With regard to assessment of malformations, the studies used wide-ranging definitions for identification of malformations to be considered. When we examined the association of benzodiazepines with cleft lip and cleft palate we

112

Dolovich et al.

had to combine these two malformations as oral cleft because many studies combined these malformations as one entity and it was not possible to stratify the data (1,10,13,36). This study differs from previous reviews. Altshuler et al. reported an association between benzodiazepines and oral cleft, but the review included studies that did not have any control groups or studies that did not have concurrent control groups (44). That method of analysis may have seriously increased the strength of association found and the heterogeneity found when studies were combined and thereby produced different results. McElhatton provided a narrative review that succinctly summarized the opposing information, but because the studies presented were not combined systematically or quantitatively, the conclusions remain controversial and inconclusive (45).

CONCLUSIONS Because women commonly use benzodiazepines and half of all pregnancies are unplanned, counselling of women on the safety of such exposure is clinically important. Pooled data from cohort studies showed no apparent association between benzodiazepine use and the risk for major malformations or oral cleft alone. There was, however, a small but significantly increased risk for oral cleft according to data from the available case-control studies. More case-control studies examining these events are needed especially because the available studies are not homogeneous. Even when the ‘‘worst-case scenario’’ is assumed, benzodiazepines do not seem to be major human teratogens, but because some cases of cleft lip can be visualized by fetal ultrasound, level 2 ultrasonography should be used to rule out this malformation.

ACKNOWLEDGMENTS This project was completed as part of a requirement for a doctor of pharmacy course on critical appraisal, PHM 605, at the University of Toronto. Contributors: LRD coordinated the study, including discussion of core ideas, design of study, information retrieval, study selection, data extraction, statistical analysis, data analysis and interpretation, and writing the paper; AA participated in discussion of core ideas, design of study selection, data extraction, data analysis and interpretation, and editing the paper; JMRV and JDBP participated in discussion of core ideas, designing the study, information retrieval, study selection, data extraction, and writing the paper; GK and TRE helped initiate the project, participated in research design, analysis, and interpretation and in editing the paper. LD will act as guarantor for the study. Funding: No additional funding. Conflict of interest: None.

REFERENCES 1. Bergman U, Rosa FW, Baum C, et al. Effects of exposure to benzodiazepine during fetal life. Lancet 1992; 340:694–696. 2. Skrabanek P. Smoking and statistical overkill. Lancet 1992; 340:1208–1209.

Use of Benzodiazepines in Pregnancy

113

3. Miller RP, Becker BA. Teratogenicity of oral diazepam and diphenylhydantoin in mice. Toxicol Appl Pharmacol 1975; 32:53–61. 4. Walker BE, Patterson A. Induction of cleft palate in mice by tranquillizers and barbiturates. Teratology 1974; 10:159–163. 5. Beall JR. Study of the teratogenic potential of oral diazepam and SCH 12041. Can Med Assoc J 1972; 106:1061. 6. Chesley S, Lumpkin M, Schatzki A, et al. Prenatal exposure to benzodiazepine: I. Prenatal exposure to lorazepam in mice alters open-field activity and GABA receptor function. Neuropharmacology 1991; 30:53–58. 7. Saxen I, Saxen L. Association between maternal intake of diazepam and oral clefts. Lancet 1975; ii:498. 8. Saxen I, Lahti A. Cleft lip and palate in Finland: incidence, secular, seasonal, and geographical variations. Teratology 1974; 9:217–224. 9. Laegreid L, Olegard R, Walstrom J, Conradi N. Teratogenic effects of benzodiazepine use during pregnancy. J Pediatr 1989; 114:126–131. 10. Safra MJ, Oakley GP. Association between cleft lip with or without cleft palate and prenatal exposure to diazepam. Lancet 1975; ii:478–480. 11. Milkovich I, van den Berg BJ. Effects of prenatal meprobamate and chlordiazepoxide hydrochloride on human embryonic and fetal development. N Engl J Med 1974; 291:1268–1271. 12. Laegreid L, Olegard R, Wahlstrom J, Conradi N. Abnormalities in children exposed to benzodiazepines in utero. Lancet 1987; i:106–109. 13. Laegreid L, Olegard R, Conradi N, et al. Congenital malformations and maternal consumption of benzodiazepines: a case-control study. Dev Med Child Neurol 1990; 32:432–441. 14. Hartz SC, Heinonen OP, Shapiro S. et al. Antenatal exposure to meprobamate and chlordiazepoxide in relation to malformations, mental development, and childhood mortality. N Engl J Med 1975; 292:726–728. 15. St Clair SM, Schirmer RG. First trimester exposure to alprazolam. Obstet Gynecol 1992; 80: 843–846. 16. Jick H, Holmes LB, Hunter JR, et al. First trimester drug use and congenital disorders. JAMA 1981; 246:343–346. 17. Heinonen OP, Sloane D, Shapiro S. Birth Defects and Drugs in Pregnancy: Maternal Drug Exposure and Congenital Malformations. Littleton, MA: Publishing Sciences Group, 1977. 18. Horwitz RI, Feinstein AR. Methodologic standards and contradictory results in case-control research. Am J Med 1979; 66:556–564. 19. Levine M, Walter S, Lee H, et al. Users’ guides to the medical literature. IV. How to use an article about harm. JAMA 1994; 271:1615–1619. 20. DerSimonian R, Laird N. Meta-analysis in clinical trials. Cont Clin Trials 1986; 7:177–188. 21. Fleiss JL. The statistical basis of meta-analysis. Stat Methods Med Res 1993; 2:121–145. 22. L’Abbe´ KA, Detsky AS, O’Rourke K. Meta-analysis in clinical research. Ann Intern Med 1987; 107:224–233. 23. Crombie DL, Pinsent RJ, Fleming DM, et al. Fetal effects of tranquilizers in pregnancy. N Engl J Med 1975; 293:198–199. 24. Kullander S, Kallen B. A prospective study of drugs and pregnancy. I. Psychopharmaca. Acta Obstet Gynecol Scand 1976; 55:25–33. 25. Laegreid L, Hagberg G, Lundberg A. Neurodevelopment in late infancy after prenatal exposure to benzodiazepines—a prospective study. Neuropediatrics 1992; 23:60–67. 26. Pastuszak A, Milich V, Chan S, et al. Prospective assessment of pregnancy outcome following first trimester exposure to benzodiazepines. Can J Clin Pharmacol 1996; 3:167–171. 27. Ornoy A, Moerman L, Lukashova I, Arnon J. The outcome of children exposed in-utero to benzodiazepines. Teratology 1997; 55:102A. 28. Bracken MB, Holford TR. Exposure to prescribed drugs in pregnancy and association with congenital malformations. Obstet Gynecol 1981; 58:336–344.

114

Dolovich et al.

29. Greenberg G, Inman WH, Weatherall JA, et al. Maternal drug histories and congenital abnormalities. BMJ 1977; 2:853–856. 30. Nakane Y, Okuma T, Takahashi R, et al. Multi-institutional study on the teratogenicity and fetal toxicity of antiepileptic drugs: a report of a collaborative study group in Japan. Epileptsia 1980; 21:663–680. 31. Robert E, Lofkvist E, Mauguiere F, Robert JM. Evaluation of drug therapy and teratogenic risk in a Rhone-Alpes district population of pregnant epileptic women. Eur Neurol 1986; 25: 436–443. 32. Noya CA. Epidemiological study on congenital malformations. Rev Cubana Hig Epidemiol 1981; 19:200–210. 33. Shiono PH, Mills JL. Oral clefts and diazepam use during pregnancy. N Engl J Med 1984; 311:919–920. 34. Rosenberg L, Mitchell AA, Parsells JL, et al. Lack of relation of oral clefts to diazepam use during pregnancy. N Engl J Med 1983; 309:1282–1285. 35. Czeizel A. Lack of evidence of teratogenicity of benzodiazepine drugs in Hungary. Reprod Toxicol 1987–1988; 1:183–188. 36. Rodriguez PE, Salvador PJ, Garcia AF, Martinez FM. Relationship between benzodiazepine ingestion during pregnancy and oral clefts in the newborn: a case-control study. Med Clin 1986; 87:741–743. 37. Tikkanen J, Heinonen OP. Congenital heart disease in the offspring and maternal habits and home exposures during pregnancy. Teratology 1992; 46:447–454. 38. Correa-Villasenor A, Ferencz C, Neill CA, et al. Ebstein’s malformation of the tricuspid valve: genetic and environmental factors. Teratology 1994; 50:137–147. 39. Winship KA, Cahal DA, Weber JP, Griffin JP. Maternal drug histories and central nervous system anomalies. Arch Dis Child 1984; 59:1052–1060. 40. Gregroire G, Derderian F, LeLorier J. Selecting the language of the publications included in a meta-analysis: is there a tower of Babel bias? Pharmacoepidemiol Drug Safety 1994; 3:S18. 41. Viggedal G, Hagberg BS, Laegreid L, Aronsson M. Mental development in late infancy after prenatal exposure to benzodiazepines—a prospective study. J Child Psychol Psychiatry 1993; 34:295–305. 42. Winter RM. In utero exposure to benzodiazepines. Lancet 1987; i:627. 43. Samrem EB, van Duijn CM, Hiilesmaa VK, et al. Maternal use of antiepileptic drugs and the risk of major congenital malformations: a joint European prospective study of human teratogenesis associated with maternal epilepsy. Epilepsias 1997; 38:981–990. 44. Altshuler LI, Cohen L, Szuba M, et al. Pharmacologic management of psychiatric illness during pregnancy: dilemmas and guidelines. Am J Psychiatry 1996; 153:592–606. 45. McElhatton PR. The effects of benzodiazepine use during pregnancy and lactation. Reprod Toxicol 1994; 8:461–475.

8 Drugs and Chemicals Most Commonly Used by Pregnant Women Doreen Matsui and Michael J. Rieder The Children’s Hospital of Western Ontario, London, Ontario, Canada

Monica Bologa The Hospital for Sick Children and the Upjohn Company of Canada, Toronto, Ontario, Canada

Frank Fassos, Michael McGuignan, and Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

Clinical Case A woman who has had repeated miscarriages tells you that she was found to have lupus antibodies and that a specialist in the university hospital put her on low-dose aspirin for the rest of her current pregnancy. She has heard bad things about salicylates in pregnancy. How would you counsel her?

INTRODUCTION The thalidomide tragedy of the late 1950s has caused many medical practitioners to view every drug and chemical as harmful to the pregnant woman. The reality is that only a few compounds have been proven to be teratogenic in humans (see Chap. 3), whereas the vast majority probably do not pose a reproductive hazard. In a recent analysis, we demonstrated that pregnant women exposed to drugs and chemicals known not to be teratogenic assigned themselves an unrealistically high teratogenic risk of 25%. This assigned risk is comparable to the risk associated with thalidomide (1). There is a clear need for authoritative information and consultation services to help pregnant women and their physicians understand fetal risks. The advice provided should be based on unbiased, up-to-date information on the reproductive effects of drugs, chemicals, and radiation. In September 1985, we started a consultation program for women concerned about antenatal exposure to drugs, chemicals, and radiation as well as exposure during lactation. A detailed description of the Motherisk Program appears elsewhere (2,3). A summary of the data from the first 6 months of telephone consultations (total ⫽ 450) appears in Table 1, which lists, in descending order of occurrence, the most common drugs and chemicals 115

116

Matsui et al.

Table 1 Most Common Antenatal Exposures Asked About in the Motherisk Clinic in Toronto 1.

2.

3.

4.

5.

6.

Antibiotics n ⫽ 55 (12.8%) Erythromycin: 16 Ampicillin: 12 Penicillin: 9 Trimethoprim-sulfamethoxazole: 9 Tetracycline: 6 Cephalosporins: 3 Analgesic-anti-inflammatory medication n ⫽ 37 (8.6%) Acetaminophen: 22 Acetylsalicylic acid: 6 Percocet, Tylenol #3: 3 Percodan 222, 282: 3 Codeine: 3 Paints, organic solvents n ⫽ 34 (7.9%) Water-based: 11 Oil-based: 10 Unspecified: 10 Lead: 3 Cold medications n ⫽ 32 (7.5%) Sympathomimetic action (decongestants): 6 Antihistamines: 11 Sympathomimetic ⫹ antihistaminic action: 8 Antitussives (dextromethorphan) ⫹ throat antiseptics: 7 Environmental pesticides n ⫽ 28 (6.5%) Unspecified: 16 Organophosphates: 7 Carbamates: 5 Cosmetic products for hair care n ⫽ 24 (5.6%) Dye: 10 Permanent-wave solution: 4 Both: 7 Bleach: 3

7. Video display terminals: 20 (4.7%) 8. Antiasthmatic medications n ⫽ 17 (4%) Adrenergic stimulants (Ventolin, Berotec): 9 Corticosteroid (Beclovent, Beconase): 5 Theophylline: 3 9. X-rays: 17 (4%) 10. Pediculocides n ⫽ 13 (3%) Chlorinated insecticides (Lindane): 5 Pyrethrins—piperonyl butoxide (R&C brand shampoo): 4 Unspecified: 4 11. Anxiolytic medications n ⫽ 11 (2.6%) Diazepam and related drugs: 7 Chlordiazepoxide: 4 12. Corticosteroids: 10 (2.3%) 13. Oral contraceptives: 8 (1.9%) 14. Rubella vaccine: 7 (1.6%) 15. Sugar substitutes: 6 (1.4%) 16. Local anesthesia (usually for dental work): 5 (1.1%) 17. H 2-receptor antagonists (ranitidine, cimetidine): 5 (1.1%) 18. Antihelmintic medications: 5 (1.1%) 19. Antifungal medications: 4 (0.9%) 20. Antitrichomonal medications (metronidazole): 4 (0.9%) 21. Laxatives: 4 (0.9%) 22. Antiemetics (Gravol brand): 4 (0.9%) 23. Mercury compounds: 4 (0.9%) 24. Ammonia: 4 (0.9%) 25. Natural gas (gas furnace and fuel gas): 4 (0.9%) 26. Antimalarial medications: 3 (0.7%) 27. General anesthesia: 3 (0.7%)

of concern to women who called the Motherisk Program. By being prepared to answer questions about the listed exposures, the health professional is likely to cover more than 80% of the common exposures during pregnancy. This chapter, which reviews many of the exposures mentioned above, has been updated to include other exposures about which inquiries have been found to be common in recent years. In the majority of cases a representative article for each exposure

Drugs and Chemicals Used by Pregnant Women

117

is referenced. For more information, the reader may refer to one of several textbooks (4–7).

ANTI-INFECTIVE MEDICATIONS Antibiotics Although an initial animal study suggested an association between erythromycin and congenital malformations (8), this relationship was not confirmed by subsequent studies in animals or humans (9). Erythromycin estolate is contraindicated in pregnancy because it appears to increase the risk of cholestatic hepatitis (10). Penicillin and ampicillin are commonly used antibiotics in pregnancy. Neither animal nor human studies have provided evidence that either antibiotic is harmful to the fetus (11–13). Similarly, cephalosporins have not been demonstrated to have teratogenic potency in humans or animals (13–15). Both trimethoprim and sulfonamides have been shown to cross the placenta (16); however, neither drug has not been incriminated as a teratogen (17,18). Acute hemolytic anemia may occur in fetuses with glucose-6-phosphate dehydrogenase (G6PD) deficiency exposed to sulfonamides in utero. When used during late pregnancy, sulfonamides may cause neonatal hyperbilirubinemia owing to displacement of bilirubin from albumin binding sites (19). The main risk from tetracycline is a yellow-brown discoloration of teeth due to deposition of the antibiotic in calcifying teeth (20,21). The risk exists only later than 4– 5 months of gestation, when the deciduous teeth begin to calcify. As for other malformations, large studies are contradictory or negative. Some authors could not demonstrate an increase in malformations associated with tetracycline (22,23). Others found a statistical association with minor malformations (inguinal hernia, hypospadias) (7). In a recent casecontrol study, the rate of treatment with doxycycline during the second and third months of gestation was not significantly higher in any group of congenital abnormalities (24). It is generally felt that tetracyclines are embryotoxic (in their effect on teeth) rather than teratogenic (25,26). Norfloxacin and ciprofloxacin are now common therapies for urinary tract infections; however, their use during pregnancy is not recommended because of arthropathy, which has been noted in beagle dogs. Use of these new quinolones during the first trimester was not associated with an increased risk of malformations or musculoskeletal problems in a small study of 35 pregnant pregnant women (27).

Pediculocides There are no reports of an increased incidence of malformations associated with the topical use of antilouse medication. Lindane (γ-benzene hexachloride) is a chlorinated insecticide. Animal studies showed no teratogenicity associated with the use of this drug (4). The manufacturer recommends caution in pregnancy, although there are no reports of congenital defects in humans. As lindane is readily absorbed and potentially neurotoxic (28), percutaneous absorption should be minimized by decreasing the concentration of the solution (⬍1%) and the duration of exposure (few minutes) (29).

118

Matsui et al.

Pyrethrins with piperonyl butoxide are poorly absorbed and their potential toxicity is lower than that of lindane. Therefore they are generally considered safer in pregnancy (30). Sulfur compounds are often used as alternative scabicides in children, pregnant women, and patients with massively excoriated skin. Approximately 1% of the topically applied sulfur compound is absorbed, but it does not appear to cause systemic toxicity. There are no reports of teratogenicity from sulfur compounds. In animal studies, despite some evidence of adverse reproductive effects, sulfur compounds had no developmental toxicity (31). Antihelminthic Medications Among the various drugs used for the treatment of helminthic infestations, the common ones used during pregnancy were pyrantel pamoate, pyrvinium pamoate, and piperazine. These drugs act within the gastrointestinal (GI) lumen, and their intestinal absorption is ideally small. Piperazine is readily absorbed from the GI tract, whereas the absorption of pyrantel pamoate and pyrvinium pamoate is low (⬍15% for pyrvinium pamoate) (32). Animal studies on piperazine (in rats and pigs) and on pyrantel pamoate (in rats, pigs, and goats) were all negative (33,34). There are no data available on human exposure during pregnancy for pyrantel pamoate or pyrvinium pamoate. In a prospective study of three mother-child pairs with first-trimester exposure to piperazine, no evidence was found that would suggest a relationship to malformations (7). In a case report, piperazine tartrate was effective and safe in eliminating a helminthic infection (35). Antifungal Medications Few of the antifungals have been proven to be teratogenic in animal studies. Systemic administration of flucytosine or griseofulvin has caused multiple malformations in different animal species (36–38). In human studies, systemic griseofulvin has been associated with congenital malformations. In one report, three abortions and two congenital malformations were described (39). More recently, an association between high-dose fluconazole therapy during pregnancy and multiple congenital anomalies has been suggested (40,41). In contrast, a prospective study of 226 women exposed to a low-dosage regimen of this antifungal agent during the first 12 weeks of pregnancy did not demonstrate an increased risk for congenital anomalies (42). A study of prescription frequency of therapy of vaginitis during the first trimester did not find significant excesses in total or specific birth defects (43). Vaginal application should be avoided after the amniotic membranes have ruptured (29). Harm to the fetus has not been demonstrated with the use of topical azoles (44). Clotrimazole has been used topically in pregnancy, and no associations with congenital malformations have been reported (45). Miconazole has been associated with an increase in fetal mortality in different animal studies when administered during active organogenesis (46). In humans, there is no reported adverse effect of miconazole on the fetus (47). Nystatin is poorly absorbed, and its use has not been associated with teratogenesis (45,48). Antitrichomonal Medication Rats given up to five times the human dose of metronidazole showed no apparent adverse effect on either fertility or fetal development. Lifetime studies in hamsters were negative

Drugs and Chemicals Used by Pregnant Women

119

(49). Two infants with midline facial defects were reported whose mothers were treated with metronidazole during the first trimester for amebiasis (50). However, metronidazole has not been incriminated as a teratogen in more than 800 pregnancies, including almost 300 with first-trimester exposure (51). Further evidence is provided by two meta-analyses, both of which concluded that metronidazole does not appear to be associated with an increased teratogenic risk (52,53).

Antimalarial Medication Chloroquine and hydroxychloroquine are the drugs of choice for prophylaxis and treatment of sensitive malaria species during pregnancy (54). There was no evidence of teratogenicity in a study of 169 infants exposed to 300 mg of chloroquine base per week in utero (55) or in a study at the Motherisk Clinic, where 14 healthy babies were born to women exposed to one of these drugs in the first trimester of pregnancy (56). Currently these drugs are used for the treatment of systemic lupus erythematosus in doses much higher than those used for malaria prophylaxis. The safety of these higher doses has not been established, although normal pregnancy outcomes have been described (57,58). There is less experience with the use of mefloquine during pregnancy. There were no congenital malformations among 11 babies born to women who had taken mefloquine prophylaxis during the first trimester of pregnancy (59). The other antimalarials—quinine, primaquine, and pyrimethamine with dapsone (Maloprim) or with sulfadoxine (Fansidar)—have been incriminated as possible abortifacients and should be avoided during pregnancy (54).

ANALGESICS AND ANTI-INFLAMMATORY MEDICATION Acetaminophen is believed to be nonteratogenic at therapeutic doses (60,61). However, the effect of very high doses has not been assessed. There is a case report of a woman who consumed 1.3 g of acetaminophen daily throughout pregnancy for headache and nausea. She also developed severe anemia, which required several transfusions. During the fifth month, polyhydramnios developed; eight amniocenteses recovered 16 L of fluid. The infant died of severe renal insufficiency (62). However, an association between the drug intake and infant’s anomalies cannot be established on the basis of a single case report. Aspirin is one of the most frequently ingested drugs in pregnancy (63). Intake of aspirin during the first trimester has not been associated with adverse effects (60). Aspirin intake during late pregnancy, especially within 1 week of delivery, may affect neonatal coagulation and may cause premature closure of the ductus arteriosus. An increased incidence of intraventricular hemorrhage (IVH) in premature or low-birth-weight infants has been associated with maternal aspirin intake: 12 out of 17 aspirin-exposed newborns (71%) developed IVH compared with 41 out of 91 non-aspirin-exposed infants (41%) (64). Aspirin consumption late in pregnancy may also produce adverse effects in the mother—such as antepartum and/or postpartum hemorrhage, prolonged gestation, and prolonged labor— resulting from the inhibition of prostaglandin synthetase (65). In a study of 45 pregnant patients with rheumatic disease treated with nonsteroidal anti-inflammatory drugs (NSAIDs) other than aspirin, no teratogenic effect was found

120

Matsui et al.

(66). However, use of all NSAIDs should be avoided during the third trimester due to their antiprostaglandin synthesis action. Codeine is a widely used narcotic analgesic. It has not been associated with an increased risk of malformations (67); however, use of large doses of codeine near term may result in a neonatal narcotic withdrawal syndrome (68–70).

COLD MEDICATIONS Decongestants Sympathomimetic amines are teratogenic in some animal species (71); however, phenylephrine and pseudoephedrine as well as other sympathomimetic agents have not been shown to increase the risk of teratogenicity in humans (72). An association was found between the intake of sympathomimetic drugs in the first trimester of pregnancy and minor malformations, like inguinal hernia and clubfoot (7), but this was not confirmed by other studies. Although the use of certain decongestants during pregnancy has been suggested as a risk factor for gastroschisis (73,74), further evidence is required. Antihistamines Most of the antihistamine compounds available in over-the-counter preparations have not been implicated as bearers of untoward effects. A large prospective study of antihistamine exposure in pregnancy showed that only with brompheniramine was there a statistically significant association with teratogenicity: 10 of 65 cases of first-trimester exposures (in the Collaborative Perinatal Project: CPP) had malformed offspring (7). However, a recent meta-analysis failed to show teratogenicity of brompheniramine (75). The use of antihistamines in the first trimester does not appear to increase the risk for major birth defects (76). Diphenhydramine, despite an initial report of possible association with an increased incidence of oral clefts, is now considered to have no significant teratogenic potential. More recently nonsedating antihistamines have become popular; however, less data are available regarding their use during pregnancy. Neither cetirizine nor astemizole were associated with an increased teratogenic risk in a limited number of patients (77,78). Antitussives Dextromethorphan has not been implicated as a potential teratogen.

ANTIASTHMATIC MEDICATIONS None of the medications used for the treatment of acute asthmatic attacks has been incriminated as a teratogen. The current approach is to treat asthmatic attacks optimally during pregnancy, since the complications of untreated asthma far outweigh unproven reproductive effects (79,80) associated with the medications. Salbutamol has not been shown to be teratogenic in animals. The drug may cause fetal tachycardia and maternal hyperglycemia, resulting in increased serum insulin and potential postnatal hypoglycemia (81,82). In a prospective study of 259 pregnant women

Drugs and Chemicals Used by Pregnant Women

121

with asthma, use of inhaled β 2-agonists was not associated with an increased frequency of adverse maternal or fetal outcomes, including congenital malformations (83). First-trimester exposure to theophylline in 121 women with asthma was associated with malformations in three (2.5%) infants, a rate similar to that expected in the general population (84). Transient toxicity manifested as jitteriness and tachycardia has been reported in newborns whose mothers took theophylline during pregnancy (85). Systemic corticosteroids have been shown to induce cleft palate and lip in numerous animal species (86,87). However, reports of more than 1100 pregnancies have failed to demonstrate teratogenicity in humans (7,88). Systemic exposure from inhaled corticosteroids is believed to be small and no harm to the fetus from inhaled steroids has been documented (89). In a study of the effects of inhaled beclomethasone in 45 pregnancies, the prevalence of congenital malformations (1 of 43 live births) was felt to be within the normal range (90).

CORTICOSTEROIDS Corticosteroids are indicated for a variety of conditions that occur during pregnancy, including asthma, arthritis, nephrotic syndrome, and inflammatory bowel disease (see discussion on corticosteroids under ‘‘Antiasthmatic Medications,’’ above).

ANXIOLYTIC MEDICATIONS Initial studies reported a three- to fourfold increased risk of cleft palate in the offspring of women exposed to diazepam (91,92). This would increase the specific risk to 3–4 per 1000 in the general population. (When compared to the general population baseline teratogenic risk of 1–5 per 100, the figures are not alarming.) Conversely, other studies did not confirm this finding (93–95). In surveys, no such association was detected (96,97). A large single dose (⬎30 mg) or sustained prenatal diazepam use can lead to the ‘‘floppy infant syndrome’’ (98), in which the infant demonstrates hypotonia, respiratory embarrassment, difficulty in suckling, and hypothermia (98,99).

ANTIEMETICS Dimenhydrinate is the chlorotheophylline salt of diphenhydramine. Two prospective studies (7,100) showed no increased risk of teratogenicity from the use of dimenhydrinate. These data are supported by negative results in animals (101). Bendectin, a combination of doxylamine and pyridoxine, was developed as an antinauseant for use during pregnancy. Shortly after it came on the market, isolated case reports of limb-reduction defects led to a large number of litigations against the manufacturer, although cohort and case-control studies did not show a higher than baseline risk for malformations (102,103). The manufacturer removed the product from the American market in 1983 because of the exceedingly high costs of insurance. A company in Quebec now makes an identical product, which has been approved in Canada to be used in pregnancy.

122

Matsui et al.

LAXATIVES Laxatives are among the most frequently used drugs during pregnancy, and there is no evidence to indicate adverse effects on human development from the majority of them. Since their action is local, their absorption and, therefore, access to the fetus is minimal (32). Most animal studies were negative; however, a few cathartics or laxatives are teratogenic or embryotoxic in animals, inducing minor skeletal changes (104). Processed hydrophilic bulking agents appear to be safe during pregnancy (105). Docusate (sodium, potassium, or calcium) is a common ingredient in many laxative preparations. A prospective study of 116 patients exposed to docusate sodium during pregnancy revealed no evidence of an association with malformations (7). Chronic use of large doses of the laxatives throughout pregnancy may cause hypomagnesemia in the mother and offspring (106). Ten studies involving 937 pregnant women treated with a variety of senna preparations showed few side effects (107). Castor oil may induce uterine contractions and therefore is contraindicated in pregnancy (32).

H 2-RECEPTOR ANTAGONISTS There are no reports of teratogenicity associated with either ranitidine or cimetidine. Animal studies were negative in the rabbit for ranitidine and negative in the mouse, rat, and rabbit for cimetidine (108,109). Cimetidine has been used at term with antacids to prevent gastric acid aspiration and subsequent pneumonitis (Mendelson’s syndrome). No adverse effects were noted in the neonate in these studies (110,111). Transient liver impairment has been described in one newborn after cimetidine exposure at term (112). In a prospective study of 178 pregnant women, no increase in major malformations was found following first-trimester exposure to H 2 blockers (113).

ORAL CONTRACEPTIVES The literature on the teratogenic effect of oral contraceptives is extensive. Although limbreduction defects, neural tube defects, cardiovascular lesions, and renal, anal, tracheal, and esophageal malformations from steroidal estrogens have been suggested, a large number of studies contradict these reports (114–118). Most studies, both positive and negative, have been heavily criticized for their methodology. The current consensus of opinion is: ‘‘Oral contraceptives present no major teratogenic hazard’’ (118). Wilson and Brent have summarized recent data by saying that ‘‘the use of exogenous hormones during human pregnancy has not been proved to cause developmental abnormalities in non-genital organ and tissue . . .’’ (119). Lack of an association between oral contraceptives and birth defects was supported by a meta-analysis of available prospective studies (118). Progesterone Current literature on the teratogenic effects of progesterone is voluminous. Several synthetic progestins have been documented to cause female pseudohermaphroditism (121). Of the progestins reported, ethisterone and norethindrone are the most active and account for most of the 200 cases of masculinization reported to date. Natural progestin (i.e.,

Drugs and Chemicals Used by Pregnant Women

123

progesterone) appears to be responsible for very few cases of sex changes. The overall risk appears to be 0.3–2.2%, with one early series describing 18% with norethindrone (122). The small amount of progestogens present in oral contraceptives is unlikely to cause virilization of a female fetus (114). A recent meta-analysis showed no significant relationship between first-trimester exposure to sex hormones generally (or to oral contraceptives specifically) and external genital malformations (123). It is currently believed that these hormones do not contribute measurably to the frequency of hypospadias (124).

VACCINES In follow-up of 514 susceptible women inadvertently immunized with rubella vaccine within 3 months of conception or during pregnancy who elected to go to term, there was no evidence of congenital abnormalities compatible with congenital rubella among their infants (125). Two infants with asymptomatic glandular hypospadias (which has been suggested to be part of the congenital rubella syndrome constellation of symptoms) have been reported; however, both had negative rubella-specific IgM titers in cord blood at birth (126). According to a 1989 Centers for Disease Control (CDC) report, the observed risk for congenital rubella syndrome following rubella vaccination continued to be zero (at that time) (127). Other investigators have concluded that the available data support the statement that the fetal risk associated with vaccination during pregnancy is so small as to be negligible (128). Although pregnancy is a contraindication for vaccination, vaccination is not normally a reason for termination of pregnancy (125). Vaccination against hepatitis B has become more widespread. No congenital abnormalities were observed among the infants born to 10 women exposed to hepatitis B vaccine during the first trimester of pregnancy (129).

ANESTHESIA Local Teratogenicity (skeletal anomalies, cataracts) has been associated with the use of lidocaine hydrochloride in animals (130). In humans, lidocaine crosses the placenta and may cause some neonatal depression and neurobehavioral changes after maternal intravenous administration of doses slightly higher than the antiarrhythmic doses (29). Lidocaine does not cause malformations after topical administration or injection for local anesthesia (131).

General There is no evidence that a single course of general anesthesia in early pregnancy is teratogenic. A recent study (131) supports the findings of a prospective study (132), demonstrating the safety of nitrous oxide. Similarly, thiopental, enflurane, and halothane were not shown to cause untoward embryonal or fetal effects (131,133,134). These results are different from occupational exposure to inhalational anesthetics, where cumulative exposure may be hazardous to the developing fetus (38) or may cause an increased risk for miscarriages (131).

124

Matsui et al.

SUGAR SUBSTITUTES There is no strong evidence to indicate that the use of moderate amounts of artificial or nonsugar sweetening agents has any adverse effect on human fetal development. One study outlined an increased prevalence of behavior problems (hyperactivity, irritability, and nervousness as well as mental retardation) and physical anomalies (deformities of bones) among offspring of mothers who were users of artificial sweeteners during pregnancy as compared to nonusers (135). However, the type of sweetening agent was not identified in the above-mentioned reports. Saccharin was shown to have no teratogenic activity in a number of studies in mice, rats, and rabbits (136,137). Some commercially available by-products (e.g., the benzoic forms) were teratogenic in a few animal studies, inducing ocular and other developmental anomalies (137). Aspartame is broken down in the small intestine into aspartic acid, methanol, and phenylalanine. Aspartic acid does not readily cross the placenta (138). Methanol is oxidized to formaldehyde and subsequently to formic acid; however, the rate of formate synthesis does not exceed the rate of its urinary excretion (139). Furthermore, the amount of methanol ingested per can of aspartame-sweetened beverage would be less than that from an equal volume of fruit juice (138). Phenylalanine is concentrated on the fetal side of the placenta but does not pose any risk to the fetus at the doses studied (139).

COSMETIC PRODUCTS FOR HAIR CARE Dyes The most frequent consultations to the Motherisk Clinic involved the ‘‘permanent’’ formulations. These are complex compounds of a variety of chemical classes, which may include phenylenediamine, toluenediamine, resorcinol, aminophenol, naphthol, 1-methyl-[ 20 H]aminobenzene, nonoxynol, oleic acid, isopropyl alcohol, ammonium hydroxide, trisodium acetate, ascorbic acid, sodium sulfite, and sodium hydroxide (140). Questions about the safety of hair products were raised as a result of some studies showing mutagenic activity in the Ames test (141). Recently, a study of 13 Chinese hair dyes showed negative Ames test results (142). There are no human studies available on these substances. Studies in laboratory animals (rats, rabbits) are available for some, and no teratogenicity has been shown for most. Some compounds (phenylenediamine and toluene compounds, possibly aminophenol and resorcinol) are considered to be potentially mutagenic and teratogenic (140,142). As to potential toxicity, aromatic nitro and amino compounds can become cyanogenic at toxic levels, thus theoretically raising the possibility that the fetus will be affected as well. However, documentation of these potentially toxic effects has involved cases of industrial intoxication, with digestive absorption of large amounts of chemicals (140). The transcutaneous absorption of these substances is not well defined, but the amount of hair coloring agents entering the systemic circulation is probably small (79). In a study of 12 hair dye formulations tested for systemic toxicity by topical application to rabbits, no clinical or histomorphological evidence of systemic effects was found. These formulations (three semipermanent formulations and nine oxidation dyes) were also tested for teratogenic effects following dermal application to pregnant rats. The course of

Drugs and Chemicals Used by Pregnant Women

125

pregnancy was not significantly affected, and no biologically significant soft tissue or skeletal changes were noted (141). In another study, five oxidation dyes were administered by gavage to pregnant rats. No signs of toxicity were observed during the treatment period. Even after administration of doses exceeding 110-fold the human exposure, there were no significant changes in fetal development and structure (143), but a significant decrease in the mean maternal weight gain was recorded. No evidence of a teratologic effect was demonstrated in pregnant rabbits exposed by gavage to a composite of dyes and base components found in commercial semipermanent hair color products (144). Permanent Wave Solutions Permanent waves in hair are produced by the use of two solutions: the waving fluid, which is an alkaline thioglycolate solution (e.g., ammonium thioglycolate, thioglycerol), and the fixation/neutralization solution, which is an acid hydrogen peroxide solution. The waving solution may have irritant properties; the ammonium thioglycolate may cause respiratory symptoms owing to immediate-type hypersensitivity. No teratogenicity has been suggested by use of these products in animals. The hydrogen peroxide, which has a local irritant effect, is rapidly degraded to water and oxygen and does not cause systemic effects (79). Bleach Hair bleaching formulas contain hydrogen peroxide and ‘‘per-’’ salt, such as ammonium persulfate, which may cause contact allergy, rhinitis, or serious respiratory symptoms owing to immediate-type hypersensitivity (79). In general, most exposures involve low dose levels. If there should be evidence of apparent toxicity, the pregnant woman should be seen by a physician. There are no reports of reproductive effects in either humans or animals.

PAINTS AND ORGANIC SOLVENTS Organic solvents are ubiquitous in the industrialized environment, appearing as individual agents or in complex associations. These compounds are volatile, and their vapors may gain access to the maternal and fetal circulations. While the most common site of toxicity is the central nervous system (CNS), the wide structural diversity of organic solvents results in potentially different teratogenic effects (145). Many industrial solvents are teratogenic in laboratory animals. While malformations depend on the specific solvent and animal species, malformations that have been described include poor fetal development, hydrocephalus, exencephaly, liver abnormalities, blood changes, and skeletal and cardiovascular defects. In vitro studies show that some organic solvents are weak mutagens (146). Chronic exposure to benzene induces chromosomal changes in human hematopoietic cells and may lead to an increased risk of leukemia or aplastic anemia. This compound is teratogenic or embryolethal in some animal species (147). Induction of birth defects in humans by industrial solvents is controversial. A widely quoted case-control study claimed increased exposure to organic solvents among mothers of children with CNS malformations when compared to mothers of normal controls (148).

126

Matsui et al.

Several epidemiological studies have demonstrated that women who work in laboratories or other industrial environments have an increased risk of miscarriage (145,149,150) or, in their offspring, gastrointestinal malformations such as omphalocele, gastroschisis, and esophageal stenosis or atresia (147). A small but statistically significant association with major malformations was found in a recent meta-analysis of studies of maternal occupational exposure to organic solvents (151). Thus, it is important when working with these chemicals to follow the recommended precautions to limit exposure. Most pregnant women who use oil-based paints apply them to one or more rooms of their home, generally in preparation for the new child. It is unlikely that such a brief, low-dose exposure will cause any of the conditions that may be associated with occupational exposure throughout pregnancy. Water-based paints, or latex paints, pose no increased reproductive risk because they have low volatility. Lead is a common constituent in glass-staining materials. It is teratogenic in laboratory animals, causing skeletal and CNS defects, oral clefts, and fetal death (145,152). In humans, lead is known to induce abortions; however, no increase in teratogenic risk has been described. Because of its abortifacient properties, there are occupational regulations for women’s exposure to lead (153). In a study of 71 pregnancies in women working with lead, 11% ended in miscarriages and neonatal mortality was almost 40% (154). When the course of 253 pregnancies of women residing in America’s ‘‘lead belt’’ was compared with an equal number of pregnancies occurring elsewhere, there were more cases of premature rupture of membranes and more preterm babies in the first group. Malformations were not reported (155). One infant with a high blood lead level and erythrocyte protoporphyrin at birth developed neurological disability at 13 months of age (156). A recent study pointed to a possible association between umbilical lead above 10 µg/dL and slightly lower scores in the Bayley test (157). Pregnant women who may have occupational exposure to lead during pregnancy should be referred to a reproductive toxicologist or other health professionals with similar expertise.

AMMONIA Ammonia vapors can cause respiratory tract irritation but no systemic toxicity (158). The different halogenated ammonia compounds and ammonia salts have been associated with negative studies as well as some positive reports of developmental anomalies, which were most likely caused by the halogen, not by the ammonia (159).

ENVIRONMENTAL PESTICIDES Increasing interest in the reproductive toxicity of pesticides and other agricultural chemicals arose after the extensive use during the Vietnam War of the defoliant Agent Orange, which was believed to be associated with an increased incidence of spontaneous abortions, stillbirths, and malformations (153). However, current data do not reveal strong evidence of fetotoxicity unless the dose was excessive. Insecticides are basically divided into organophosphates, aromatic carbamate esters, and chlorinated compounds. Many of them have teratogenic potential in laboratory animals, inducing various structural abnormalities (skeletal and brain defects) (160–162). Organophosphates are commonly used insecticides with irreversible cholinesterase-

Drugs and Chemicals Used by Pregnant Women

127

inhibiting properties (162). In animal studies, dose levels that were not maternally lethal did not produce fetotoxicity, fetal lethality, or malformations (163). Malathion is a common household organophosphate insecticide that is also used topically for the treatment of lice. There are no reports of human teratogenicity from the use of malathion, and animal studies were negative (164). Carbamates exert a reversible cholinesterase-inhibiting activity (162). In animal studies, neither propoxur nor carbofuran caused fetal lethality or developmental abnormalities at dose levels that were not maternally lethal (163). Methyl benzimidazole carbamate (MBC) has a demonstrated capacity to interfere with mitosis in different species (fungi, viruses, bacteria, cells of bovine brain). Although the studies concluded that MBC was a potential teratogen, a study in rats and rabbits revealed some degree of embryolethality but no visceral or skeletal malformations (165). Although the cholinesterase inhibitors are commonly used, there is a lack of human reports on their reproductive effects. Very few of the women consulting the Motherisk Clinic with concerns about environmental pesticides had been exposed to toxic amounts of these compounds; rather, they had been in an area where pesticides were sprayed and consequently they could smell them. It is conceivable that in the absence of symptomatic maternal toxicity, such short exposure will not cause embryonal or fetal damage. Whenever a woman exposed to these chemicals experiences any degree of clinical toxicity, she and her fetus should be closely monitored by a physician. MERCURY COMPOUNDS In our survey, the concerns about mercury almost always involved the accidental ingestion of metallic (elemental) mercury in the form of dental amalgams or the contents of a thermometer. Swallowed elemental mercury is not absorbed to a significant degree from the GI tract and does not pose any risk of systemic toxicity. The amount of mercury vapor released from dental amalgam is very low and should not cause teratogenic effects (166). Once absorbed, practically all mercury compounds are teratogenic in animals and humans (167). The most potent teratogen is methyl mercury, the causative agent of the well-defined Minamata disease (168). NATURAL GAS Natural gas is approximately 85% methane. In high concentrations, methane is an asphyxiant, but it does not cause systemic toxicity (169). Of more concern is the combustion product carbon monoxide (CO). Carbon monoxide is a known teratogen, causing CNS abnormalities in the fetus (170). A prospective multicenter study of acute CO exposure in pregnancy was conducted by Motherisk and included some women who had been exposed to CO from a malfunctioning furnace. Mild poisoning was not associated with an increased fetal risk; however, adverse fetal outcome was noted with severe maternal CO toxicity (171). These results are consistent with a previous review of the literature (172). Pregnant women exposed to leaking or malfunctioning gas furnaces should have their carboxyhemoglobin level determined and should be referred to a specialized service. Fuel gas—a mixture of methane, ethane, propane, and butane—is considered to be potentially teratogenic in animals. Pregnant mice exposed to 58% concentrations of fuel gas on day 8 of gestation produced offspring with hydrocephalus and exencephaly (173).

128

Matsui et al.

VIDEO DISPLAY TERMINALS Suspicion that the use of video display terminals (VDTs) had adverse effects on the embryo and fetus started after several clusters of miscarriages and malformations were reported (174). However, investigation reconfirmed that VDTs do not emit x-rays, microwaves, or other radiation at levels that would be harmful (175,176). Current data indicate that VDTs do not increase the risk to either the pregnant woman or the fetus (175–177).

X-RAYS Exposure to radiation doses less than 5 rad is not associated with an increase in congenital malformations (178). In the dose range of 5–15 rad, there may be an increased teratogenic risk. For dosages of more than 15 rad, there appears to be a two- to threefold increase in the incidence of major malformations (179,180). The usual exposure from diagnostic xrays should generally be far below the teratogenic range. In any case, the radiologist involved should be contacted to estimate the apparent dose of exposure.

GENERAL INSTRUCTIONS When advising a pregnant woman over the telephone about the potential reproductive effects of drugs, chemicals, or radiation, the health professional should rule out other risk factors that may affect her pregnancy outcome. These factors include age, obstetrical history, other exposures including alcohol and smoking, genetic background, paternal exposures, and the underlying medical condition. Socioeconomic status also plays an important role in the normal nutrition and progress of a pregnancy. The interviewer should advise the woman that in every pregnancy there is a 1–5% risk of major malformations and that even if her exposure does not appear to increase the teratogenic risk, the baseline risk still exists. Clinical Case Answer Salicylates, alone or combined with low-dose corticosteroids, have been shown to prevent repeated miscarriages in selected groups of women. These drugs have not been shown to affect the fetus adversely. Theoretically, they may cause premature closure of the fetal ductus arteriosus or bleeding complications, but these complications were not documented with small doses.

NOTE ADDED IN PROOF Since this chapter was written, more recent studies have provided additional information on three of the drug and chemical classes discussed earlier in this chapter. A case-control study showed a relationship between exposure to systemic corticosteroids during the first trimester of pregnancy and an increased risk of cleft lip (with or without cleft palate) in the newborn infants (OR ⫽ 6.55) (181). The results of a metanalysis of studies of benzodiazepine use during pregnancy were also recently published. Data from case-control

Drugs and Chemicals Used by Pregnant Women

129

studies showed a small increased risk for major malformations or oral cleft alone. The investigators concluded that although benzodiazepines do not seem to be major human teratogens, level 2 ultrasonography should be used to rule out visible forms of cleft lip (182). In a recent prospective study of the outcome following maternal occupational exposure to organic solvents, significantly more major malformations occurred among fetuses of women exposed to organic solvents than controls (RR ⫽ 13.0). The risk was increased among women who reported symptoms associated with their exposure (183).

REFERENCES 1. Koren G, Bologa M, Long D, et al. The perception of teratogenic risk by pregnant women exposed to drugs and chemicals during the first trimester. Am J Obstet Gynecol 1989; 160: 1190–1194. 2. Koren G, Feldman Y, Shear N. Motherisk—a new approach to antenatal counselling of drug/ chemical exposure. Vet Hum Toxicol 1986; 28:563–565. 3. Koren G, MacLeod SM. Monitoring and avoiding drug and chemical teratogenicity. Can Med Assoc J 1986; 125:1079–1081. 4. Schardein JL. Chemically Induced Birth Defects. New York: Marcel Dekker, 1985. 5. Briggs GG, Freeman RK, Yaffe SJ. Drugs in Pregnancy and Lactation, 5th ed. Baltimore: Williams & Wilkins, 1998. 6. Sheppard TH. Catalog of Teratogenic Agents, 7th ed. Baltimore: Johns Hopkins University Press, 1992. 7. Heinonen OP, Slone D, Shapiro S. Birth Defects and Drugs in Pregnancy. Littleton, MA: PSG Publishing Company, 1977. 8. Miyoshi T. Experimental studies on the effects of toxicants on pregnancy of rats. J Osaka City Med Center 1959; 8:309–318. 9. Philipson A, Sabath LD, Charles D. Erythromycin and clindamycin absorption and elimination in pregnant women. Clin Pharmacol Ther 1976; 19:68–77. 10. McCormack WM, George H, Donner A, et al. Hepatotoxicity of erythromycin estolate during pregnancy. Antimicrob Agents Chemother 1977; 12:630–635. 11. Wasz-Hockert O, Nummi S, Voupala S, Jarvinen P. Transplacental passage of azidocillin, ampicillin and penicillin G during early and late pregnancy. Acta Paediatr Scand 1970; 206(suppl):109–110. 12. Ceccarelli P, Rossi M, Cianti F, Domenici C. Use of amoxicillin in the obstetrical and gynecological field. Minerva Gynecol 1977; 29:137–142. 13. Cheung M. Counselling the pregnant patient. Contemp Pract 1988; 15(2):7–14. 14. Hasegawa Y, Yoshida T, Kozen T. A teratological study on cefamandole in rats and rabbits. Chemotherapy 1979; 27(suppl 5):658–681. 15. Chu Chen K, Sabeti S. L’evaluation clinique de la cephalexin orale. Int J Clin Pharmacol 1970; 2(suppl):124–128. 16. Reid DWJ, Caille G, Kaufmann NR. Maternal and transplacental kinetics of trimethoprim and sulfamethoxazole, separately and in combination. Can Med Assoc J 1975; 112:67S–72S. 17. Ochoa AG. Trimethoprim and sulfamethoxazole in pregnancy. JAMA 1971; 217:1244–1245. 18. Brumfitt W, Prusell R. Double-blind trial to compare ampicillin, cephalexin, cotrimoxazole, and trimethoprim in treatment of urinary infection. BMJ 1972; 2:673–676. 19. Jamerot G, Into-Malmberg MB, Esbjorner E. Placental transfer of sulphasalazine and sulphapyridine and some of its metabolites. Scand J Gastroenterol 1981; 16:693–697. 20. Cohlan SQ. Tetracycline staining of teeth. Teratology 1977; 15:127–130. 21. Rubin P. Prescribing in pregnancy. Practitioner 1990; 234:556–560. 22. Culshaw JA. TTC and congenital limb abnormalities. Br Med J 1962; 2:924–927.

130

Matsui et al.

23. Elder HA, Santamarina BAG, Smith S, Kass EH. The natural history of asymptomatic bacteriuria during pregnancy: the effect of tetracycline on the clinical course and the outcome of pregnancy. Am J Obstet Gynecol 1971; 111:441–462. 24. Czeizel AE, Rockenbauer M. Teratogenic study of doxycycline. Obstet Gynecol 1997; 89: 524–528. 25. Corcoran R, Castles JM. Tetracycline for acne vulgaris and possible teratogenesis. Br Med J 1977; 2:807–808. 26. Graham JM, Blanco JD. Chlamydial infections. Primary Care 1990; 17:85–93. 27. Berkovitch M, Pastuszak A, Gazarian M, et al. Safety of new quinolones in pregnancy. Obstet Gynecol 1994; 84:535–538. 28. Sammiguel GS, Ferrer AP, Alberich MT, Genaoui BM. Consideraciones sobre el tratamiento de la infancia y en el embarazo. Actas Dermosifilogr 1980; 71:105–108. 29. Berkowitz RL, Constan DR, Mochizuki TK. Handbook for Prescribing Medications During Pregnancy. Boston: Little Brown, 1981. 30. Robinson DH, Shepherd DA. Control of head lice in school children. Curr Ther Res 1980; 27:1–6. 31. Palanker AL, Keating JW, Weinberg MS, et al. Reproductive teratogenic and egg production studies in animals fed SO 2-treated activated sewage sludge. Toxicol Appl Pharmacol 1973; 25:454–459. 32. Goodman Gilman A, Goodman LS, Rale TH, Murad F. The Pharmacological Basis of Therapeutics, 7th ed. New York: Macmillan, 1985. 33. Owaki Y, Sakai T, Momiyama H. Teratological studies on pyrantel pamoate in rats and rabbits. Oyo Takuri 1970; 5:33–50. 34. Wilk AL. Relation between teratogenic activity and cartilage-binding affinity of norchlorcyclizine analogues. Teratology 1969; 2:172. 35. Young RL, Zund G, Mason BA, Faro S. Pelvic inflammatory disease complicated by massive helminthic hyperinfection. Obstet Gynecol 1989; 74:484–486. 36. Chaube S, Murphy ML. The teratogenic effects of 5-fluorocytosine in the rat. Cancer Res 1969; 29:554–557. 37. Jindra J, Anjezdska A, Janousek V. Embryotoxic effects of high doses of griseofulvin on the skeleton of the albino mouse. In: Evaluation of Embryotoxicity, Mutagenicity and Carcinogenicity Risks in New Drugs. Proceedings of the Third Symposium on Toxicological Testing for Safety of New Drugs. Prague, 1979, pp 161–165. 38. Lecyk M. Toxicity of copper sulfate in mice embryonic development. Zool Pol 1980; 28: 101–106. 39. Gotz H, Reichenberger M. Ergebrisse einer Fragebogenaktion bei 1670 Dermatologen der BD uber Nebenwirkungen bei der Griseofulvin-therapie. Hautarzt 1972; 23:485–492. 40. Lee BE, Feinberg M, Abraham JJ, Murthy ARK. Congenital malformations in an infant born to a woman treated with fluconazole. Pediatr Infect Dis J 1992; 11:1062–1064. 41. Wray CM. Congenital anomalies and fluconazole. Can Med Assoc J 1997; 156:91. 42. Mastroaicovo P, Mazzone T, Botto LD, et al. Prospective assessment of pregnancy outcomes after first-trimester exposure to fluconazole. Am J Obstet Gynecol 1996; 175:1645–1650. 43. Rosa FW, Baum C, Shaw M. Pregnancy outcomes after first-trimester vaginitis drug therapy. Obstet Gynecol 1987; 69:751–755. 44. Ernest JM. Topical antifungal agents. Obstet Gynecol Clin North Am 1992; 19:587–607. 45. Doering PL, Santiago TM. Drugs for treatment of vulvovaginal candidiasis: comparative efficacy of agents and regimens. Drug Intell Clin Pharm 1990; 24:1078–1983. 46. Ito C, Shibutani Y, Taya K, Ohnishi H. Toxicological studies of miconazole: 3. Teratological studies of miconazole in rabbits. Jyakuhin Kenkyu 1976; 7:377–381. 47. Wallenberg HCS, Wladimiroff JW. Recurrence of vulvovaginal candidiasis during pregnancy: comparison of miconazole vs nystatin treatment. Obstet Gynecol 1976; 48:491–494. 48. Donders GGG, Moerman P, Caudron J, Van Assche FA. Intrauterine Candida infection: a

Drugs and Chemicals Used by Pregnant Women

49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75.

131

report of four infected fetuses from two mothers. Eur J Obstet Gynecol Reprod Biol 1990; 38:233–238. Hammill HA. Trichomonas vaginalis. Obstet Gynecol Clin North Am 1989; 16:531–540. Cantu JM, Carcia-Cruz D. Midline facial defect as a teratogenic effect of metronidazole. Birth Defects 1982; 18:85–88. Morgan J. Metronidazole treatment in pregnancy. Int J Gynaecol Obstet 1978; 15:501–502. Burtin P, Taddio A, Ariburnu O, et al. Safety of metronidazole in pregnancy: a meta-analysis. Am J Obstet Gynecol 1995; 172:525–529. Caro-Paton T, Carvajal A, Martin de Diego I, et al. Is metronidazole teratogenic? A metaanalysis. Br J Clin Pharmacol 1997; 44:179–182. Recommendations for the prevention of malaria in travelers. MMWR 1988; 37:277–284. Hart CW, Naunton RF. The ototoxicity of chloroquine phosphate. Arch Otolaryngol 1964; 80:407–412. Levy M, Buskila D, Gladman DD, et al. Pregnancy outcome following first trimester exposure to chloroquine. Am J Perinatol 1991; 8(3):174–178. Parke A. Antimalarial drugs and pregnancy. Am J Med 1988; 85(suppl 4A):30–33. Parke A, West B. Hydroxychloroquine in pregnant patients with systemic lupus erythematosus. J Rheumatol 1996; 23:1715–1718. Balocco R, Bonati M. Mefloquine prophylaxis against malaria for female travellers of childbearing age. Lancet 1992; 340:309–310. Rudolph AM. Effects of aspirin and acetaminophen in pregnancy and in the newborn. Arch Intern Med 1981; 141:358–363. Levy G, Garretson LK, Soda DM. Evidence of placental transfer of acetaminophen. Pediatrics 1975; 55:895–899. Char VC, Chandra R, Fletcher AB, Avery GB. Polyhydramnios and neonatal renal failure— a possible association with maternal acetaminophen ingestion. J Pediatr 1975; 86:638–639. Corby DG. Aspirin and pregnancy: maternal and fetal effects. Pediatrics 1978; 62(suppl): 930–937. Rumack CM, Guggenheim MA, Rumack BM, et al. Neonatal intracranial hemorrhage and maternal use of aspirin. Obstet Gynecol 1981; 58(suppl):525–565. Wolff F, Berg R, Bolte A. Clinical study of the labor-inhibiting effects and side effects of ASA. Geburtshilfe Frauenheilkd 1981; 41:96–100. Ostensen M, Ostensen H. Safety of nonsteroidal antiinflammatory drugs in pregnant patients with rheumatic disease. J Rheumatol 1996; 23:1045–1049. Little BB, Snell LM, Klein VR, et al. Maternal and fetal effects of heroin addiction during pregnancy. J Reprod Med 1990; 35:159–162. Mangurten HH, Benawra R. Neonatal codeine withdrawal in infants of nonaddicted mothers. Pediatrics 1980; 65:159–160. Van Leeuwen G, Guthrie R, Strange F. Narcotic withdrawal reaction in a newborn infant due to codeine. Pediatrics 1965; 36:635–636. Smith CG, Smith MT. Substance abuse and reproduction. Semin Reprod Endocrinol 1990; 8:55–64. Loevy H, Roth BF. Induced cleft palate development in mice: comparison between the effect of epinephrine and cortisone. Anat Rec 1968; 160:386–390. Smith CV, Rayburn WF, Anderson JC, et al. Effect of a single dose of oral pseudoephedrine on uterine and fetal Doppler blood flow. Obstet Gynecol 1990; 76:803–806. Werler MM, Mitchell AA, Shapiro S. First trimester maternal medication use in relation to gastroschisis. Teratology 1992; 45:361–367. Torfs CP, Katz EA, Bateson TF, et al. Maternal medications and environmental exposures as risk factors for gastroschisis. Teratology 1996; 54:84–92. Seto A, Einarson TR, Koren G. Evaluation of brompheniramine safety in pregnancy. Reprod Toxicol 1993; 7:393–395.

132

Matsui et al.

76. Seto A. Pregnancy outcome following first trimester exposure to antihistamines: meta-analysis. Am J Perinatol 1997; 14:119–124. 77. Einarson A, Bailey B, Jung G, et al. Prospective controlled study of hydroxyzine and cetirizine in pregnancy. Ann Allergy Asthma Immunol 1997; 78:183–186. 78. Pastuszak A, Schick B, D’Alimonte D, et al. The safety of astemizole in pregnancy. J Allergy Clin Immunol 1996; 98:748–750. 79. Nater JP, DeGroot AC. Unwanted Effects of Cosmetics and Drugs Used in Dermatology. Amsterdam: Elsevier, 1985. 80. Dunlap NE, Bailey WC. Corticosteroids in asthma. South Med J 1990; 83:428–432. 81. Hastwell G. Salbutamol aerosol in premature labour. Lancet 1975; 2:1212–1213. 82. Thomas DJB, Dove AF, Alberti KGMM. Metabolic effects of salbutamol infusion during premature labour. Br J Obstet Gynaecol 1977; 84:497–499. 83. Schatz M, Zeiger RS, Harden KM, et al. The safety of inhaled β-agonist bronchodilators during pregnancy. J Allergy Clin Immunol 1988; 82:686–695. 84. Stenius-Aarniala B, Riikonen S, Teramo K. Slow-release theophylline in pregnant asthmatics. Chest 1995; 107:642–647. 85. Labovitz E, Spector S. Placental theophylline transfer in pregnant asthmatics. JAMA 1982; 247:786–788. 86. Walker BE. Effect of glucocorticoids on palate development in the rat. Anat Rec 1969; 163: 281–285. 87. Walker BE. Induction of cleft palate in rats with anti-inflammatory drugs. Teratology 1971; 4:39–42. 88. Greenberger PA, Patterson R. Management of asthma during pregnancy. N Engl J Med 1985; 312:897–903. 89. Nelson-Piercy C, Moore-Gillon J. Treatment of Asthma. In: Rubin P, ed. Prescribing in Pregnancy, 2nd ed. London: BMJ Publishing Group, 1995, pp 46–58. 90. Greenberger PA, Patterson R. Beclomethasone diproprionate for severe asthma during pregnancy. Ann Intern Med 1983; 98:478–480. 91. Saxdn 1. Associations between oral clefts and drugs taken during pregnancy. Int J Epidemiol 1975; 4:37–44. 92. Safra JM, Oakley GP. Association between cleft lip with or without cleft palate and neonatal exposure to diazepam. Lancet 1975; 2:478–480. 93. Czeizel A. Diazepam, phenytoin and etiology of cleft lip and/or cleft palate. Lancet 1976; 1:810–811. 94. Rosenberg L, Mitchell AA, Parsells JL, et al. Lack of relation of oral clefts to diazepam use during pregnancy. N Engl J Med 1983; 309:1282–1285. 95. Pastuszak A, Milich V, Chan S, Chu J, et al. Prospective assessment of pregnancy outcome following first trimester exposure to benzodiazepines. Can J Clin Pharmacol 1996; 3:167–171. 96. Czeizel A. Lack of evidence of teratogenicity of benzodiazepine drugs in Hungary. Reprod Toxicol 1988; 1:183–188. 97. Bergman U, Boethius G, Swartling PG, et al. Teratogenic effects of benzodiazepine use during pregnancy. J Pediatr 1990; 116:490–491. 98. Loudon JB. Psychotropic drugs. BMJ 1987; 294:1. 99. Grimm NE. Diazepam and other benzodiazepines in pregnancy. In: Yanai J, ed. Neurobehavioral Teratology. New York: Elsevier, 1984, pp 153–163. 100. Mellin GW, Katzenstein M. Meclozine and fetal abnormalities. Lancet 1963; 1:222–223. 101. McColl JD. Dimenhydrinate in pregnancy. Can Med Assoc J 1963; 88:861. 102. Zierler S, Rothman KH. Congenital heart disease in relation to maternal use of Bendectin and other drugs in early pregnancy. N Engl J Med 1985; 313:347–352. 103. Einarson TR, Leeder JS, Koren G. A method for meta-analysis of epidemiological studies. Drug Intell Clin Pharm 1988; 22:813–824. 104. Asuzu IU, Shetty SN, Anika SM. Effects of chronic oral administration in mice of the

Drugs and Chemicals Used by Pregnant Women

105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133.

133

gut-stimulating crystals Croton penduliflorus seed oil. J Ethnopharmacol 1990; 30:135– 143. West L, Warren J, Cutts T. Diagnosis and management of irritable bowel syndrome, constipation, and diarrhea in pregnancy. Gastroenterol Clin North Am 1992; 21:793–802. Schindler AM. Isolated neonatal hypomagnesaemia associated with maternal overuse of stool softeners. Lancet 1984; 2:822. Anonymous. Risk assessment for senna during pregnancy. Pharmacology 1992; 44(suppl 1): 20–22. Brogden RN, Speight TM, Avery GS. Cimetidine: a review of its pharmacological properties and therapeutic efficacy in peptic ulcer disease. Drugs 1978; 15:93–131. Tamura J, Sato N, Ezaki H. Teratological study on ranitidine hydrochloride in rabbits. J Toxicol Sci 1983; 8(suppl 1):141–150. Crawford JS. Cimetidine in elective caesarian section. Anaesthesia 1981; 36:641–642. Koren G, Zemlickis DM. Outcome of pregnancy after first trimester exposure to H 2 receptor antagonists. Am J Perinatol 1991; 8:37–38. Glade G, Saccar CL, Pereira GR. Cimetidine in pregnancy: apparent transient liver impairment in the newborn. Am J Dis Child 1980; 134:87–88. Magee LA, Inocencion G, Kamboj L, et al. Safety of first trimester exposure to histamine H2 blockers—a prospective cohort study. Dig Dis Sci 1996; 41:1145–1149. Simpson JL. Review: relationship between congenital anomalies and contraception. Adv Contracept 1985; 1:3–30. Simpson JL, Phillips OP. Spermicides, hormonal contraception and congenital malformations. Adv Contracept 1990; 6:141–167. Katz Z, Lancet M, Skornick J, et al. Teratogenicity of progestogens given during the first trimester of pregnancy. Obstet Gynecol 1985; 65:775–780. Yovich JL, Turner SR, Draper R. Medroxyprogesterone acetate therapy in early pregnancy has no apparent fetal effects. Teratology 1988; 38:135–144. Bracken MB. Oral contraception and congenital malformations in offspring: a review and meta-analysis of the prospective studies. Obstet Gynecol 1990; 76:552–557. Wilson JG, Brent RL. Are female sex hormones teratogenic? Am J Obstet Gynecol 1981; 141:567–580. Roe TF, Alfi OS. Ambiguous genitalia in XX male children: report of two infants. Pediatrics 1977; 60:55–59. Jacobson BD. Hazards of norethindrome therapy during pregnancy. Am J Obstet Gynecol 1962; 84:962–968. Raman-Wilms L, Tseng AL, Wighardt S, et al. Fetal genital effects of first-trimester sex hormone exposure: a meta-analysis. Obstet Gynecol 1995; 85:141–149. Kallen B, Mastroiacovo P, Lancaster PAL, et al. Oral contraceptives in the etiology of isolated hypospadias. Contraception 1991; 44:173–182. Best JM. Rubella vaccines: past, present and future. Epidemiol Infect 1991; 107:17–30. Enders G. Rubella antibody titers in vaccinated and nonvaccinated women and results of vaccination during pregnancy. Rev Infect Dis 1985; 7(suppl 1):103–107. Centers for Disease Control. Rubella vaccination during pregnancy—United States, 1971– 1988. MMWR 1989; 38:289–293. Preblud SR. Some current issues relating to rubella vaccine. JAMA 1985; 254:253–256. Levy M, Koren G. Hepatitis B vaccine in pregnancy: maternal and fetal safety. Am J Perinatol 1991; 8:227–232. Zhivkov E, Atanasov L. Experiments in obtaining and preventing congenital cataracts in rats. Ophthalmologia 1965; 2:105–112. Friedman JM. Teratogen update: Anesthetic agents. Teratology 1988; 37:69–77. Crawford JS, Lewis M. Nitrous oxide in early human pregnancy. Anaesthesia 1986; 41:900–905. Mellin GW. Comparative teratology. Anesthesiology 1968; 29:1–4.

134

Matsui et al.

134. Pope WD, Halsey MJ, Lansdown AB, Bateman PE. Lack of teratogenic dangers with halothane. Acta Anaesthesiol 1975; 23(suppl):169–173. 135. Stone D, Matalka E, Pulaski B. Do artificial sweeteners ingested in pregnancy affect the offspring? Nature 1971; 231:53–59. 136. Taylor JM, Friedman L. Combined chronic feeding and three-generation reproduction study of sodium saccharin in the rat. Toxicol Appl Pharmacol 1974; 29:154–158. 137. Lederer J. Saccharin, its by-products and their teratogenic effects. Louvain Med 1977; 96: 495–501. 138. Pitkin RN. Aspartame ingestion during pregnancy. In: Stegink LD, Filer LJ Jr, eds. Aspartame Physiology and Biochemistry. New York: Marcel Dekker, 1984, pp 555–563. 139. Sturtevant FM. Use of aspartame in pregnancy. Int J Fertil 1985; 30:85–87. 140. Gosselin RE, Smith RP, Hodge HC, Braddock JE. Clinical Toxicology of Commercial Products, 5th ed. Baltimore: Williams & Wilkins, 1984. 141. Burnett C, Goldenthal EI, Harris SB, et al. Teratology and percutaneous toxicity studies on hair dyes. J Toxicol Environ Health 1976; 1:1027–1040. 142. Wong L, Li SL, Qin YH, et al. Studies on mutagenicity of hair dyes made in China. Biomed Environ Sci 1991; 4:310–316. 143. DiNardo JC, Picciano JC, Schnetzinger RW, et al. Teratological assessment of five oxidative hair dyes in the rat. Toxicol Appl Pharmacol 1985; 78:163–166. 144. Wernick T, Lanman BM, Fraux JL. Chronic toxicity, teratologic, and reproduction studies with hair dyes. Toxicol Appl Pharmacol 1975; 32:450–460. 145. Taskinen HK. Effects of parental occupational exposures on spontaneous abortion and congenital malformations. Scand J Work Environ Health 1990; 16:297–314. 146. Hemminki K, Sorsa M, Vainio H. Occupational Hazards and Reproduction. Washington, DC: Hemisphere, 1985. 147. Editorial. Is there a fetal solvent syndrome? A medical letter. Reprod Toxicol 1983; 2(5). 148. Holmberg PC. CNS defects in children born to mothers exposed to organic solvents during pregnancy. Lancet 1979; 2:177–179. 149. Huel G, Meriger D, Bowler R. Evidence for adverse reproductive outcomes among women microelectronic assembly workers. Br J Ind Med 1990; 47:400–404. 150. Lindbohm MJ, Taskinen H, Sallman M, Herminki K. Spontaneous abortions among women exposed to organic solvents. Am J Ind Med 1990; 17:449–463. 151. McMartin KI, Chu M, Kopecky E, et al. Pregnancy outcome following maternal organic solvent exposure: a meta-analysis of epidemiological studies. Clin Invest Med 1997; 20(suppl):8. 152. Roeleveld N, Zielhuis GA, Gabreels F. Occupational exposure and defects of the central nervous system in offspring: review. Br J Ind Med 1990; 47:580–588. 153. Barlow SM. Reproductive effects of occupation. In: Fabro S, Scialli AR, eds. Drug and Chemical Action in Pregnancy. New York: Marcel Dekker, 1986, pp 353–373. 154. Rom WN. Effects of lead on the female and reproduction: a review. Mt Sinai J Med 1976; 43(5):542–551. 155. Fahim MS, Fahim Z, Hall DG. Effects of subtoxic lead levels on pregnant women in the state of Missouri. International Conference on Heavy Metals in the Environment, Toronto, 1975. 156. Singh N, Donovan CM, Hanshaw JB. Neonatal lead intoxication in a prenatally exposed infant. J Pediatr 1978; 93:1019–1021. 157. Bellinger D, Leviton A, Watermaux C, et al. Longitudinal analysis of prenatal and postnatal lead exposure and early cognitive development. N Engl J Med 1987; 316:1037–1043. 158. Close LG, Catlin FI, Cohn AM. Acute and chronic effects of ammonia burns of the respiratory tract. Arch Otolaryngol 1980; 106:151–158. 159. Livingstone CS. Neonatal goitre. BMJ 1966; 2:50–54.

Drugs and Chemicals Used by Pregnant Women

135

160. Kavlock RJ, Chernoff N, Gray LE, et al. Teratogenic effects of benomyl in the Wistar rat and CD-1 mouse, with emphasis on the route of administration. Toxicol Appl Pharmacol 1982; 62:44–54. 161. Ottolenghi AD, Haseman JK, Suggs F. Teratogenic effects of aldrin, dieldrin, endrin in hamsters and mice. Teratology 1974; 9:11–16. 162. Banerjee J. Inhibition of human fetal brain acetylcholinesterase: marker effect of neurotoxicity. J Toxicol Environ Health 1991; 33:283–290. 163. Courtney KD, Andrees JE, Springer J, Dalley L. Teratogenic evaluation of the pesticides baygon, carbofuran, dimethoate, and EPN. J Environ Sci Health 1985; B20:373–406. 164. Kanja LW, Skaare JV, Ojwang SBO, Maitai CK. A comparison of organochlorine pesticide residues in maternal adipose tissue, maternal blood, cord blood, and human milk from mother/infant pairs. Arch Environ Contam Toxicol 1992; 22:21–24. 165. Janardhan A, Sattur PB, Sisodia P. Teratogenicity of methyl benzimidazole carbamate in rats and rabbits. Bull Environ Contam Toxicol 1984; 33:257–263. 166. Larsson KS. Teratological aspects of dental amalgam. Adv Dent Res 1992; 6:114–119. 167. Koos BJ, Longo LD. Mercury toxicity in the pregnant woman, fetus and newborn infant. Am J Obstet Gynecol 1976; 126:390–409. 168. Roeleveld N, Zielhaus GA, Gabreels F. Occupational exposure and defects of the central nervous system in offspring: review. BMJ 1990; 47:580–588. 169. Windholz M, ed. The Merck Index, 9th ed. Rahway, NJ: Merck, 1976. 170. Longo LD. The biological effects of carbon monoxide in the pregnant woman, fetus and newborn infant. Am J Obstet Gynecol 1977; 129:69–103. 171. Koren G, Sharav T, Pastuszak A, et al. A multicentre, prospective study of fetal outcome following accidental carbon monoxide poisoning in pregnancy. Reprod Toxicol 1991; 5:397– 403. 172. Norman CA, Halton DM. Is carbon monoxide a workplace teratogen? A review and evaluation of the literature. Ann Occup Hyg 1990; 34:335–347. 173. Kato T. Embryonic abnormalities of the CNS caused by the fuel-gas inhalation of the mother animal. Folia Psychiatr Neurol Jpn 1958; 11:301–324. 174. Aldridge JFL. Visual display units and health. Practitioner 1985; 229:539–545. 175. Schnorr TM, Grajewski BA, Hornung RW, et al. Video display terminals and the risk of spontaneous abortion. N Engl J Med 1991; 324:727–733. 176. Cluff S. Health hazards of video display terminals. Modern Med Can 1986; 41:501–509. 177. Parazzini F, Luchini L, La Vecchia C, Crosignani PG. Video display terminal use during pregnancy and reproductive outcome—a meta-analysis. J Epidemiol Commun. Health 1993; 47:265–268. 178. Bentur Y, Horlatsch N, Koren G. Exposure to ionizing radiation during pregnancy: perception of teratogenic risk and outcome. Teratology 1991; 43:109–112. 179. Brent RL. Evaluating the alleged teratogenicity of environmental agents. Clin Perinatol 1986; 13:615–648. 180. Brent RL. The effects of embryonic and fetal-exposure to x-rays, microwaves and ultrasound. In: Brent RL, Beckman DA, eds. Clinics in Perinatology: Vol 13. Teratology Philadelphia, Saunders, 1988, pp 301–330. 181. Rodriguez-Pinilla E, Martinez-Frias ML. Corticosteroids during pregnancy and oral clefts: A case-control study. Teratology 1998; 58:2–5. 182. Dolovich LR, Addis A, Regis Vaillancourt JM, et al. Benzodiazepine use in pregnancy and major malformations or oral cleft: Meta-analysis of cohort and case-control studies. BMJ 1998; 317:839–843. 183. Khattak S, K-Moghtader G, McMartin K, et al. Pregnancy outcome following gestational exposure to organic solvents. JAMA 1999; 281:1106–1109.

9 Periconceptional Folate and Neural Tube Defects Time for Rethinking Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

Clinical Case A patient of yours who is now 6 weeks into gestation has heard that maternal supplementation of folic acid may decrease the risk for neural tube defects. She wants you to recommend vitamin pills that will have enough folate. How much daily folate does she need?

INTRODUCTION The Motherisk Program in Toronto evaluates and counsels pregnant women, their families, and health professionals on the teratogenic risks of drugs, chemicals, environmental agents, and infections. At present we deal with 60 inquiries a day, mainly from Ontario but also from other parts of Canada and the United States. In addition to performing follow-up of our own patients and conducting prospective studies, we continuously review and analyze new published studies. When important information becomes available, we feel it is our mandate to publicize such data, which may have a direct impact on the health of many unborn babies. We believe the new data on preconceptual folate supplementation qualify as such a major breakthrough.

THE PROBLEM Neural tube defects (NTD) affect about 0.6 to 1.5 per 1000 born babies in Canada. Consisting mainly of anencephaly and meningomyelocele, NTDs are serious lethal or severely debilitating congenital malformations. It is estimated that the risk of a recurrence of NTD in a woman who has had a previous NTD is about 3–4% (1). During the last decade effective antenatal methods have been developed to diagnose NTD in utero, with secondtrimester ultrasound combined with amniotic or maternal blood α-fetoprotein offering an almost 100% sensitivity in some centers (2). Presently, ultrasound is recommended at 16 137

138

Koren

and 19 weeks. In most centers, amniocentesis for raised α-fetoprotein is the diagnostic approach, whereas maternal α-fetoprotein and ultrasound are used for screening. However, these tests are not performed routinely on women with unknown risk for NTD (e.g., because of previous NTD or because they are receiving valproic acid, carbamazepine, or retinoids). Evidence has been accumulated during the last two decades suggesting that low preconceptual consumption of folate and vitamins is associated with an increased risk of NTD. These retrospective studies could not distinguish between the potential effects of vitamins versus folate, however. The main criticism of these observational studies was that women consuming low amounts of folate and vitamins may have clustering of many other nutritional and socioeconomic risk factors, leading them to increased teratogenic risk (3). However, this controversy was laid to rest when the British Medical Research Council study, a multinational, double-blind, placebo-controlled effort, clearly proved the protection effect of pharmacological folate dose (4 mg/day) over placebo on the recurrence rate of NTD in women who had a previous NTD. Vitamins alone (without folate) had no protective effect (1). It was subsequently argued that this protective effect may not be relevant to the prevention of the occurrence of a first NTD in the general population. However, a recently completed double-blind, placebo-controlled study from Hungary has clearly shown the protective effect of folate at 0.8 mg/day plus vitamins over placebo (4) in women with no previous history of NTD. It is probable that in a heterogeneous society such as that in Canada, different ethnic groups have different magnitudes of risk for NTD; however, this new evidence suggests that, at least in part, NTDs are caused by folate deficiency. An association between folate deficiency and other malformations (e.g., cleft lip or palate) has been suggested but has not been verified. In Canada, the recommended dose of dietary folate for the general population was 0.2 mg/day. This recommendation stems from the concept that 0.2 mg/day of folate is high enough to prevent saturation of hepatic stores of folate. With the new data presented above, it is quite clear that this decrease in recommended dose of folate may increase the risk of NTD in Canada. Recently, the national Health and Welfare changed its recommendations to 0.4 mg/day. There is recent evidence that large segments of North American women below or near poverty have median intakes of folic acid of 0.15 mg/day. Even among those above poverty, many have very low folate intake (e.g., 25% of women have average folate intake of only 0.142 mg/day) (5). Moreover, insulin-dependent diabetic mothers and women with potentially low folate levels, due to inflammatory bowel disease and other disease states, may be at a higher risk than the normal population. At present it is not clear whether 0.4 mg/day of folate is inferior to the 0.8 mg/day tested in Hungary; however, it is very probable that intake of 0.2 mg/day is below the preventive dose.

RETHINKING THE SOLUTION Because NTDs are induced in the first 28 days of pregnancy, adequate maternal folate intake must start preconceptionally. Since, however, almost half of all pregnancies in North America are unplanned, it is evident that recommendations alone are not likely to reach many women and are very likely to miss high-risk women of low socioeconomic status who tend to consume substantially less folate.

Periconceptional Folate and NTDs

139

A variety of methods have been suggested to ensure that women of reproductive age receive adequate folate supplements (6). Educating women to consume more fruits and vegetables might achieve that goal, but supplementation of basic food sources such as bread, salt, cereals, or milk might turn out to be more efficacious because it does depend on changing the behavior patterns of women. The potential deleterious effects of incorporating folic acid in basic food supplementation must be carefully reviewed, however. Some masking of B 12 deficiency may occur in patients with combined deficiencies of vitamin B 12 and folic acid (7). NTDs induced by folate deficiency are now proven to be preventable at the primary level (versus secondary prevention by pregnancy termination). This is an exciting opportunity, similar in magnitude to the prevention of cretinism by means of iodine supplements. The health community, in concert with the various governments, should address these issues as soon as possible. Any delay in response will result in unnecessary occurrence of NTDs, with immense suffering and cost to the children, their families, and the public at large. Clinical Case Answer At 6 weeks of gestation, supplementation of folate will not reverse an NTD if it was already formed. The point here is that folate must be given periconceptionally at a dose of 400 µg/day.

REFERENCES 1. MRC Vitamin Study Research Group. Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. Lancet 1991; 338:131–137. 2. Nodel AS, Green NK, Holmes LB, et al. Absence of need for amniocentesis in patients with elevated levels of maternal alpha fetoprotein and normal ultrasonographic examinations. N Engl J Med 1990; 323:557–561. 3. Mills JL, Rhoads GG, Simpson JL, et al. The National Institute of Child Health and Human Development Neural Tube Defect Study Group. N Engl J Med 1989; 321:430–435. 4. Czeizel A. Controlled studies of multivitamin supplementation on pregnancy out-come. In: Keen CL, Bendich A, Willhite CC, eds. Maternal Nutrition and Pregnancy Outcome. New York: New York Academy of Sciences, May 17–20, 1992, San Diego, CA. 5. Block G, Abrams B. Vitamin and mineral status of women of childbearing potential. In: Keen CL, Bendich A, Willhite CC, eds. Maternal Nutrition and Pregnancy Outcome. New York: New York Academy of Sciences, May 17–20, 1992, San Diego, CA. 6. Oakley G. Periconceptional folic acid supplementation for the prevention of spina bifida and anencephaly. In: Keen CL, Bendich A, Willhite CC, eds. Maternal Nutrition and Pregnancy Outcome. New York: New York Academy of Sciences, May 17–20, 1992, San Diego, CA. 7. Babior BM, Bunn HF, Megaloblastic anemias. In: Braunwald E, Isselbacher KJ, et al., eds. Harrison’s Principles of Internal Medicine. 1987, pp 1498–1504.

10 The Effectiveness of Preconceptional Counseling on Women’s Compliance with Folic Acid Supplementation Anne Pastuszak, Dimple Bhatia, Bunmi Okotore, and Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

INTRODUCTION Folic Acid and Neural Tube Defects The accumulated data from the last 14 years—in the form of randomized controlled trials (1,2), nonrandomized intervention studies (3,4), and observational studies (5–8)—have provided evidence that periconceptional folic acid ingestion can reduce the number of pregnancies resulting in neural tube defects (NTDs) in high-risk patients and low-risk primigravidas (9). From the available evidence, there is potential to avert at least 50% (10) and as many as 72% (3) of incident cases of NTDs if women at high risk for NTD recurrence consume 4.0 mg/day of folic acid during the critical stage of neural tube development in a subsequent pregnancy. Similarly, there is potential to decrease the incidence rate of NTDs in low-risk pregnancies by 40–60% if women consume 0.4 mg/day of folic acid during the critical stage of neural tube closure (10,11). Reflecting these levels of evidence, current Canadian guidelines from the Society of Obstetricians and Gynecologists of Canada (12), the Canadian Council of Medical Geneticists (13), the Canadian Task Force on the Periodic Health Examination (14), and the recent Canadian National Workshop on Folic Acid Supplementation recommend that low-risk women of reproductive age who are planning pregnancy should supplement their diets with 0.4 mg/day folic acid and high-risk women of reproductive age who are planning pregnancy should supplement their diets with 4.0 mg/day. A major issue in translating this breakthrough knowledge into a primary prevention strategy is that on average, Canadian and American women consume only half of the needed 0.4 µg/day (15) and that at least half of pregnancies are unplanned (16). Therefore, the issue of informing women on their preventable risk for NTDs prior to conception becomes most crucial. The Motherisk Program is a teratology information service (TIS) at the Hospital for Sick Children in Toronto, which provides evidence-based fetal risk assessments to pregnant women and health professionals. Self-selected callers who contact the program and request information about the safety or risk of maternal exposure to prescription or over141

142

Pastuszak et al.

the-counter medications, chemicals, radiation, or infectious diseases during any stage of gestation, preconception, or during lactation are interviewed initially by a counselor over the telephone who has the prerogative to refer the patient to a once-weekly clinic for additional counseling. Risk assessments may be different for women who are planning a pregnancy, are at various stages postconception, or are breast-feeding; because of this, data regarding the start and stop dates of maternal exposure, fetal or neonatal age at exposure, and maternal medical and obstetrical history are documented on a standardized intake form. Callers whose exposures occurred pre- or postconceptionally undergo a risk assessment that is compared to the baseline risk of 1–3% for major congenital anomalies in the general population. In addition to receiving an explanation about the baseline risk, women who are planning a pregnancy at the time they call are advised to supplement their daily diet with folic acid prior to conception and to continue this dose throughout pregnancy. Women are told that, ideally, supplementation should start once birth control is discontinued. However, given that the majority of pregnancies is unplanned, Motherisk uses the phrase prior to conception. Counselors explain that in order for folic acid to be maximally effective, it is imperative that it be consumed during the time that the neural tube closes, which, in the human embryo, occurs 24–30 days after fertilization. The goal at Motherisk is to ensure that, through risk-factor modification, women who are planning a pregnancy supplement their diet with folic acid in amounts appropriate for their established risk status; low-risk women are to supplement with 0.4 mg/day and high-risk women, with 4 mg/day. The objective of the present study was to assess the effectiveness of our folic acid supplementation counseling program by comparing compliance in a group of counseled women to that of a group of uncounseled women.

STUDY DESIGN This was a prospective study in which we artificially manipulated a study factor—patient knowledge about folic acid supplementation—and subsequently measured patient compliance with this advice. This design is analogous to an intervention in a cross-section of a population without randomization.

PATIENTS AND METHODS Intake data from all callers to Motherisk are routinely entered into a computerized database. This was used to retrieve the charts of patients planning a pregnancy who called during the 15-week period of May 2 to August 19, 1994. To be included in the study, the patient’s intake form had to show that she was planning a pregnancy, that she had no prior knowledge about the need to supplement with folic acid, and that counseling on folic acid supplementation had been given to her by the program. During the months of November 1994 and April 1995, we attempted to contact these women to record their compliance with our advice. A supplementary questionnaire form included the following questions: Do you recall receiving advice from Motherisk about folic acid supplementation? Do you recall the

Preconceptional Counseling and Folic Acid

143

suggested dose? Other than the Motherisk Program, what other sources of folic acid information have you received? Are you currently taking the dose we suggested? If no, why not? If yes, what dose are you taking? How often? In what form? When did you start this dose? When did you stop this dose? Have you conceived since your initial call? If no, are you still planning a pregnancy? If yes, what is your current gestational age? Has an ultrasound been done? Please describe what you do in your occupation. In addition to verbal counseling, women are receiving through the mail an information pamphlet prepared by the Motherisk team. As a control group for the planning women with no prior knowledge about folic acid, we collected a group of women who were at ⱕ14 weeks’ gestation at the time of their first call to Motherisk. The following supplementary questions were asked of them: Are you currently taking a multivitamin? If no, why not? If yes, what brand, what dose, when did you start, and when did you stop? Why are you taking this vitamin? Did someone recommend that you take it? If yes, who did, and when did this occur? Have you had your first prenatal visit yet? If yes, when? By choosing this group, we corrected for the possibility that, prior to calling, women may have received information on folic acid supplementation from sources other than the Motherisk Program. We used the Blishen, Carrol, and Moore socioeconomic index for occupations (17) in Canada, as this is most applicable in situations where access to socioeconomic data is limited to occupational titles. This scale (from 0, lowest, to 100, highest) is based on 1981 Canadian census data and is a composite of the prevailing income and education levels in each occupation; it is not a measure of occupational prestige. Z comparisons between the counseled and uncounseled groups were performed by Student’s t-test for unpaired data or the chi-square test where appropriate.

RESULTS In the 15-week study period, there were 3438 Ontario callers who requested information from Motherisk, 319 (9.3%) of whom were planning a pregnancy and not routinely taking folic acid at the time of the initial call. Of those, we followed up and reported on the compliance of a cohort of 145 women. The demographics of these women are presented in Table 1, showing the majority to be between 30 and 34 years of age, nulliparous, and in the middle of the socioeconomic scale. At the time of follow-up, 66 of the 145 women had conceived (46%). Regardless of whether any had conceived, 105 women (72%) were taking folic acid at the time of the follow-up call and 40 (28%) were not. Of the 105 women who supplemented, 103— who were rated as at low risk by the counselor at the initial call—were taking 0.4–1.0 mg of folic acid per day and 2 women were taking 5 mg/day. Those who did not take folic acid cited various reasons for their ultimate decision not to supplement (Table 2). All 145 women were asked to identify from which sources other than Motherisk they obtained information about folic acid. A total of 36 (25%) cited their health professional, 36 (25%) cited a magazine or newspaper, 3 (2%) cited a friend, 1 (1%) cited a relative, 4 (3%) cited other sources, 32 (22%) cited a combination of the previous, and 33 (23%) said they had not heard about the beneficial effects of folic acid before (Table 3). Women who took folic acid (72%) were different from women who did not comply (28%) in that they were more likely to abstain from cigarettes, alcohol, and cocaine use and were more likely to have received multiple reminders about folic acid (Table 1).

Table 1 Characteristics Between Women Planning Pregnancy, Who Subsequently Did and Did Not Supplement with Folic Acid

Maternal age at initial call Gravidity at initial call Gravida 0 Gravida 1 Gravida 2 Gravida 3 ⱖGravida 4 Parity at initial call Parity 0 Parity 1 Parity 2 ⱖParity 4 Socioeconomic score# 0–19 20–39 40–59 60–79 80–99 100 Time interval between initial call and follow-up Folic acid product used Jamieson (1 mg) Materna (1 mg) Orifer-F (0.8 mg) Centrum Forte (0.4 mg) Other (0.4–0.99 mg) Other (ⱖ1 mg) Pregnant at follow-up? Yes No Cigarette use at follow-up None ⬎20 per week Alcohol use at follow-up None ⬍5 drinks per week ⱖ5 drinks per week Cocaine use at follow-up None At least once since initial call Use of folate-depleting drug(s) at followup? No Yes a

Took folic acid

Did not take folic acid

p value

32.0 ⫾ 4.2 years range 23–43

32.5 ⫾ 4.9 years range 23–48

0.52

40 31 17 8 8

(38%) (30%) (16%) (8%) (8%)

55 (52%) 28 (27%) 17 (16%) 5 (5%) mean 51.9 37 (35%) 13 (12%) 33 (31%) 21 (20%) 0 1 (1%) 30.5 ⫾ 9.6 weeks range 12–45 36 37 2 4 15 11

12 17 7 4

(30%) (42%) (18%) (10%) 0

21 (52%) 14 (35%) 3 (8%) 2 (5%) mean 50.2 15 (38%) 6 (15%) 12 (30%) 7 (18%) 0 0 28.7 ⫾ 8.7 weeks range 14–45

0.47 0.7 0.91 0.81

0.6 0.95 0.89 0.97 0.49 — 0.6 0.31

(34%) (35%) (2%) (4%) (14%) (10%)

60 (57%) 45 (43%)

6 (15%) 34 (85%)

98 (93%) 7 (7%)

31 (78%) 9 (22%)

0.02

97 (92%) 8 (8%) 0

27 (68%) 10 (25%) 3 (8%)

0.006

105 (100%) 0

38 (95%) 2 (5%)

25 1*

15 0

This SES index does not provide scores for persons with certain occupations. In our uncounseled population, there were 3 students, 4 unemployed, 4 who were self-employed, and 41 who described themselves as homemakers. b At follow-up, the patient reported that she started carbamazepine therapy for grandmal seizures and supplemented with 5 mg/day folic acid.

Preconceptional Counseling and Folic Acid

145

Table 2 Reasons Cited by Patients at Follow-up for the Decision Not to Take Folic Acid No longer planning, no further reasons given Husband laid off, recent divorce, planning IVF a in future Concurrent illness and drug therapy were too preoccupying Folic acid made her sick Her physician didn’t believe in the benefits of folic acid Didn’t know that it had to be taken prior to pregnancy Had heard nothing about folic acid No reasons given Didn’t think ‘‘the fuss’’ was ‘‘for real’’ Didn’t really think this issue applied to her Didn’t have enough information to convince her Found out she was 6 weeks pregnant 1 week after initial call a

14 1 each 4 1 1 1 6 5 1 1 2 1

In vitro fertilization.

At follow-up, 66 of the 145 women (46%) had conceived; 55 (83%) were still gravid at the time of the call, 6 had miscarried, 1 had a tubal pregnancy, and 4 had delivered healthy infants without congenital defects. Of these 66, a total of 60 (91%) had complied with the Motherisk advice and taken folic acid; 47 (71%) took folic acid prior to, during, and after the critical period for neural tube development; whereas 13 (20%) started folic acid supplementation once pregnancy was diagnosed. Six women (9%) had conceived without supplementation during the critical period. At follow-up, 79 women (54%) had not conceived; of these, 45 were taking folic acid and 42 women were still planning a pregnancy, while 34 were not taking folic acid and 14 of them were still planning a pregnancy. In our control group of 147 women who were ⱕ14 weeks gestation at the initial contact and who reflected folic acid supplementation in the general population of uncounseled pregnant patients, 90 (61%) were already taking folic acid at the time of their call to Motherisk. However, only 25 (17%) of them initiated use of folic acid prior to conception, significantly less than among the counseled women [47 of the 66 (71%) who conceived, p ⫽ 0.0001]. The average age in this population was 29.6 ⫾ 4.9 years; however, those who took a multivitamin were older than those who did not (30.3 years versus 28.5 years, p ⫽ 0.03). Those not supplementing cited various reasons for this decision (Table 4).

Table 3

Sources of Folic Acid Information as Described by Patients

Source Health professional Magazine or newspaper Friend Relative Motherisk only Other Multiple

Number who Took folic acid 26 31 3 1 14 3 27

(25%) (30%) (3%) (1%) (13%) (3%) (26%)

Number who Did not take folic acid

p value

10 (25%) 5 (12%) 0 0 19 (48%) 1 (2%) 5 (12%)

0.85 0.06 0.67 0.61 0.0001 0.06 0.14

146

Pastuszak et al.

Table 4 Women Who Were ⱕ14 Weeks post LMPa at the Time of the Initial Call to the Motherisk Program Age Gravidityb Parity Gestational age post-LMP, weeks Patient’s risk for neural tube defects Patients with first prenatal visit prior to initial call to Motherisk Patients taking multivitamins If no, why not?

Was date started prior to conception? Did someone recommend you take a supplement? If yes, who? a b

29.6 years ⫾ 4.9 G 0 (1), G 1 (45), G 2 (49), G 3 (30), G 4 (13), G 5 (4), G 6 (1), G 7 (3), G 8 (1) P0 (67), P1 (58), P2 (14), P3 (5), P4 (1), P5 (0), P6 (2), P7 (0) 0 to ⬍2 (0), 2 to ⬍4 (2), 4 to ⬍6 (15), 6 to ⬍8 (37), 8 to ⬍10 (34), 10 to ⬍12 (23), 12 to ⬍14 (28), 14 to ⬍16 (8) Low (146), high (1) 54 (37%)

90 (61%) Believes she is healthy enough without (9), multivitamins make her sick (11), her physician says its not necessary (3), pregnancy just confirmed/will start soon (4), has severe hyperemesis (1), got out of the habit (1), stopped because of diarrhea (1), no one has told her to/given her any (8), no reason given by patient (10), cannot swallow prenatal vitamins (1), disorganized/hasn’t bought any yet (3), forgot to take them (1), never used vitamins in previous pregnancies (3), just got them (1) Yes 25 (27%), no 65 (71%), unsure 1 (1%) Yes 65 (71%), no 26 (29%) Physician (57), nurse (2), Employee of health food store (2), dietician (1), relative (1), pharmacist (2)

LMP ⫽ last monthly period. Numbers in parentheses are the number of patients or where indicated, percentages.

DISCUSSION In a recent report of compliance with folic acid supplementation in low-risk women of reproductive age, Clark and Fisk (18) reported that of 411 women attending an antenatal clinic for their first pregnancy visit (9–26 weeks, median 15), only 5% supplemented with folic acid and started this prior to conception; 8 (2%) did so through diet modification and 12 (3%) through folate tablets. In our baseline group of 147 women, 25 (17%) supplemented with folic acid and started this prior to conception. Our figure is significantly higher than that of Clark and Fisk, possibly due to the additional 2–3 years that have elapsed since their publication; yet it reflects a very low chance that uncounseled women will be protected against the risk of neural tube defects. In a recent random-digit-dialing telephone survey that interviewed 1005 women aged 18–45 in the United States (19), only 25% of nonpregnant women of childbearing age reported taking a folic acid–containing supplement daily. Our study documents that by counseling women prior to pregnancy, we have significantly improved women’s compliance with folic acid supplementation. In a recent

Preconceptional Counseling and Folic Acid

147

study (submitted), we have shown that only a small proportion of family physicians in Toronto routinely and correctly advise women on folate supplementation. Taken together with these present results, it is evident that if physicians or other health professionals fulfill this counseling role, they can achieve very effective results in preventing fetuses from being exposed to low levels of folic acid. It is conceivable that the women in our study are inherently more motivated to comply, given that they represent a self-selected population of women who voluntarily seek advice from a TIS on pregnancy risks and how to avoid them. Yet, without counseling, only 17% of the women we surveyed effectively protected their fetuses with sufficient folic acid. Regardless of this selection bias, our study highlights that a counselling program can effectively influence patient behavior. Counseling and education do not, however, always result in compliance with supplementation. The extent to which a woman’s behavior coincides with medical advice is influenced by her perception of disease severity, the clinical setting, the dose schedule, and the associated side effects of therapy. Non-compliance with supplementation may result from a woman’s inability to perceive the impact of neural tube defects as a disease in relation to other more publicized, prevalent conditions such as AIDS, breast cancer, or Down’s syndrome. Noncompliance may also result directly consequence from the patient’s failure to read or hear about the benefits of supplementation, her feeling that she is not susceptible, or her view of NTDs as insignificant. She may also feel that the costs of supplementation are too high or that her personality is not conducive to risk-prevention behaviour. During the 1950s a group of social psychologists developed a theoretical framework for explaining the likelihood of an individual to undertake a recommended preventative health action (20). Termed the Health Belief Model, it is based on the decision-making concepts of attractiveness of the goal to the individual and on a personal estimate of the likelihood of attaining this goal. Applying this model to exposure to the success of folic acid supplementation counselling, it follows that whether or not a woman will comply with supplementation is dependent upon a complex, interactive interplay of (1) her perceptions of personal susceptibility to having a child with a neural tube defect, (2) her perceived potential benefits derived from supplementing, (3) the degree of severity of the consequences of a neural tube defect, and (d) preexisting physical, psychological, financial and other barriers or costs related to supplementing throughout pregnancy. Certain cues to action must occur in order to trigger behaviour in patients so that they will adhere to recommendations. This interaction between individual perception, modifying characteristics, and internal and external cues to action which ultimately affect whether or not a woman will supplement must be identified before any primary prevention program can have a fair chance of success. The Motherisk Program has recently demonstrated the ability to modify patients’ perceptions of risk (21) and thus prevent unnecessary pregnancy terminations by women exposed to nonteratogenic drugs. This counselling technique can has been transferred to the present group of patients who are planning a pregnancy. Despite showing that more than 70% of patients who conceived took folic acid during the critical time for neural tube closure, we did document that 30% of women did not take it correctly, prior to conception. In addition, of the 79 who did not conceive, 34 were not taking folic acid daily and 14 of these were still planning to conceive. To overcome this noncompliance, we have prepared a one-page patient fact sheet on folic acid. This is being mailed after the first telephone consultation to all patients who are planning

148

Pastuszak et al.

Figure 1 The Health Belief Model adapted to conditions of folic acid supplementation. Prediction of compliance is not straightforward, since numerous factors interact to affect supplementation. Only if all factors are perceived as maximally beneficial by the patient will the supplementation campaign work with full effectiveness.

to conceive at the time of their initial call in order to reiterate the importance of the message given over the telephone.

CONCLUSION This is the first study to demonstrate that a TIS can successfully intervene to modify women’s compliance with folic acid supplementation. By comparing a group counseled to supplement with folic acid to a group not previously counseled, we have shown that one-on-one counseling resulted in correct supplementation in 71% of those who conceived. In the current absence of either a provincial or a national policy for folic acid supplementation, we expect that there will be many women who have never heard or read about the beneficial effects of periconceptional folic acid. Effective counseling by health professionals can play a pivotal role in reducing the risk of neural tube defects.

ACKNOWLEDGMENT Supported by a grant from The Hospital for Sick Children Foundation.

Preconceptional Counseling and Folic Acid

149

REFERENCES 1. Laurence KM, James N, Miller M, et al. Double-blind randomized controlled trial of folate treatment before conception to percent recurrence of neural-tube defects. BMJ 1981; 282: 1509–1511. 2. MRC Vitamin Study Research Group. Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. Lancet 1991; 338:131–137. 3. Smithells RW, Nevin NC, Seller MJ et al. Further experience of vitamin supplementation for the prevention of neural tube defect recurrences. Lancet 1983; 1:1027–1031. 4. Vergel RG, Sanchez LR, Heredero BL, et al. Primary prevention of neural tube defects with folic acid supplementation: Cuban experience. Prenat Diagn 1990; 10:149–152. 5. Mulinare J, Cordero JF, Erickson JD, Berry RJ. Periconceptional use of multivitamins and the occurrence of neural tube defects. JAMA 1988; 260:3141–3145. 6. Bower C, Stanley FJ. Dietary folate as a risk factor for neural-tube defects: evidence from a case-control study in Western Australia. Med J Aust 1989; 150:613–619. 7. Mills JL, Rhoads GG, Simpson JL, et al. The absence of a relation between the periconceptional use of vitamins and neural tube defects. N Engl J Med 1989; 321:430–435. 8. Milunsky A, Jick A, Jick SS, et al. Multivitamin/folic acid supplementation in early pregnancy reduces the prevalence of neural tube defects. JAMA 1989; 262:2847–2852. 9. Czeizel A, Dudas I. Prevention of the first occurrence of neural-tube defects by periconceptional vitamin supplementation. N Engl J Med 1992; 327:1832–1835. 10. CDC. Recommendations for use of folic acid to reduce the number of spina bifida cases and other neural tube defects. JAMA 1993; 269:1233–1238. 11. Werler MM, Shapiro S, Mitchell AA. Periconceptional folic acid exposure and risk of occurrence of neural tube defects. JAMA 1993; 269:1257–1261. 12. SOGC Genetics Committee: Recommendations on the use of folic acid for the prevention of neural tube defects. J Soc Obstet Gynecol Can 1993; Mar (suppl):41–46. 13. Van Allen M, Fraser FC, Dallaire L, et al. Recommendations on the use of folic acid supplementation to prevent the recurrence of neural tube defects. Can Med Assoc J 1993; 149:1239– 1243. 14. Canadian Task Force on the Periodic Health Examination. Periodic health examination, 1994 update: 3. Primary and secondary prevention of neural tube defects. Can Med Assoc J 1994; 151:159–166. 15. Health and Welfare Canada, Nurition. Canada Food Consumption Patterns Report, Ottawa, 1977. 16. Better news on population. Lancet 1992; 339:1600. 17. Blishen BR, Carroll WK, Moore C. The 1981 socioeconomic index for occupations in Canada. Can Rev Soc Anth 1987; 24:465–486. 18. Clark NAC, Fisk NM. Minimal compliance with the Department of Health recommendation for routine folate prophylaxis to prevent fetal neural tube defects. Br J Obstet Gynecol 1994; 709–710. 19. Centers for Disease Control. Knowledge and use of folic acid by women of childbearing age— United States, 1995. MMWR 44:716–718. 20. Becker MH, Maiman LA, Kirscht JP, et al. Patient perceptions and compliance: recent studies of the Health Belief Model. In: RB Haynes, DW Taylor, DL Sackett, eds. Compliance in Health Care. Baltimore: Johns Hopkins University Press, 1979, pp 78–120. 21. Koren G, Pastuszak A. Prevention of unnecessary pregnancy terminations by counselling women on drug, chemical and radiation exposure during the first trimester. Teratology 1990; 41:657–661.

11 Pregnancy Outcome Following Maternal Exposure to Corticosteroids A Prospective Controlled Cohort Study and a Meta-Analysis of Epidemiological Studies Laura Park-Wyelie, Paolo Mazzotta, Myla E. Moretti, Anne Pastuszak, Lizanne Beique, Laura Hunnisett, Mark H. Friesen, Sheila Jacobson, Sonja Kasapinovic, Debra Chang, Irena Nulman, and Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

Orna Diav-Citrin The Hebrew University, Jerusalem, Israel

C. Laskin and K. Spitzer The Toronto Hospital and The University of Toronto, Toronto, Ontario, Canada

Thomas R. Einarson The University of Toronto, Toronto, Ontario, Canada

INTRODUCTION Prednisone—a glucocorticoid used in the treatment of asthma, in transplantation, and in collagen-vascular and other disorders—has been shown to cross the human placenta (1– 3). Despite the well-documented effects of short-term, antenatally administered steroids on improving pulmonary function in the human fetus (4), experimental studies have shown that prolonged, prenatal exposure to prednisone may cause a reduction in fetal growth and birth weight (5). Large doses administered to pregnant mice, rats, and rabbits during organogenesis have caused cleft palate in the exposed offspring (6–8). The same teratogenic effect appears as that seen with the naturally occurring glucocorticoid, cortisone, on the developing palate of mice (9). Direct extrapolation from animals to humans is provisional. In case reports where women were treated during the first trimester, prednisone was prescribed for a plethora of diseases [Hodgkin’s disease (10), leukemia (11,12), renal transplantation (13–15), systemic lupus erythematosus (SLE) (16–24), antiphospholipid antibodies (25–27), asthma (28–30), rheumatoid arthritis (31), regional enteritis (32), glomerulonephritis (33), subchorionic hematomas (34), Takayasu’s arteritis (35), idiopathic thrombocytopenic purpura (36–38), pemphigus vulgaris (39), ulcerative colitis (40), recurrent fetal loss (41), ne151

152

Park-Wyelie et al.

phrotic syndrome (42), and chronic renal failure (43)]. Although a variety of birth defects in their offspring were reported, there was no consistent pattern of fetopathy. The increased incidence of low birth weight and stillbirths may be attributable to the underlying maternal condition for which the steroids were administered. There are two case reports of women treated with cortisone during the period of organogenesis [for SLE (44) and idiopathic steatorrhea (45)]. In the first case, the infant was born with a cleft palate of the soft tissue and, in the second, a stillborn male infant was born with a palate that had a large postalveolar cleft. In both cases, cortisone exposure had occurred during the critical window of palate development, between the 45th and 84th days postconception. The most rigorous epidemiological data come from those observational and interventional studies in which women with unexplained recurrent fetal loss (46–51), or autoimmune thrombocytopenia (52) were either randomized or allocated to receive prednisone throughout pregnancy. In these publications, there was no evidence to suggest that the baseline risk for cleft palate was elevated. In light of the paucity of controlled data, we investigated the relative fetal safety of maternal prednisone therapy in a prospective, controlled, observational study. In addition, we conducted a meta-analysis to determine the risk, if any, of steroid use on the fetus with respect to major malformations and, more specifically, oral clefts.

SUBJECTS AND METHODS Prospective Study This was a prospective, controlled, observational cohort study in which the exposed patients (group 1) were women who voluntarily telephoned the Motherisk Program for information about the fetal safety/risk of maternal prednisone exposure during pregnancy. Women were included in the exposed cohort if they had taken systemic prednisone for any indication in the first trimester and if their exposure details and medical history had been collected prospectively by one of our teratology information counselors and documented on the clinic’s standardized intake forms. A second corticosteroid-exposed group consisted of pregnant women who were enrolled in a randomized double-blind clinical trial of aspirin and prednisone (ASA/P) for recurrent fetal loss at the Toronto Hospital (group 2) (53) plus additional women treated with corticosteroids with the same protocol and indication (see below). Women were included if they were between 18 and 39 years of age, medically healthy, had at least two consecutive pregnancy losses prior to 32 weeks, and were positive for at least one of the following on two of three occasions at screening: lupus anticoagulant, anticardiolipin IgG, antilymphocyte IgM antibody, anti–double stranded DNA antibody, activated partial thromboplastin time antinuclear antibody, or anti–single stranded DNA antibody. After confirmation of pregnancy, each woman was randomized to receive either the investigative treatment or placebo. Women in the treatment arm received prednisone, initially as 0.8 mg/kg/day for 4 weeks, which was then reduced to 0.5 mg/kg/day until delivery or fetal loss, and aspirin, 100 mg/day until 36 weeks gestation. Women treated with prednisone who became pregnant after they completed the ASA/P trial, and women who received prednisone at their physician’s discretion were also included. Although considered openlabel patients, they were treated with the same protocol, assessed in an identical fashion, and followed up prospectively.

Pregnancy Outcome After Corticosteroid Exposure

153

The original control group for this randomized controlled trial (RCT) (53) could not be used because of the poor quality of data for pregnancy outcome owing to a large portion of missing data. Therefore, as the unexposed control group, we chose a cohort of pregnant women who voluntarily contacted the Motherisk Program for safety/risk information about either topical retinoic acid for uncomplicated acne or oral astemizole for seasonal allergies (group 3), neither of which has been associated with an increased risk for major malformations (54,55). This comparison group was intended to represent the baseline population of pregnant women. We prospectively collected the following information from every caller: dose, indication and dates of initiation and discontinuation of the medication of concern, and the obstetric, medical, genetic and drug exposure history of the mother. Approximately 1 year after the expected date of delivery, as calculated by the date of the last menstrual period, all patients were telephoned by a Motherisk team member who collected details about the outcome of pregnancy, birth weight, and presence or absence of birth defects, and perinatal and neonatal complications. Follow-up details were subsequently corroborated by written documentation from the child’s physician. Similar maternal and outcome data of those patients from the University of Toronto clinical trial were abstracted from the clinical trial study forms and recoded on the Motherisk follow-up forms prior to statistical analysis. The primary outcome of interest was the rate of major birth defects attributable to corticosteroids, which was compared among the three groups. If other etiologies were identified which were proven to cause the specific malformation, these cases were not counted as malformations likely caused by corticosteroids (see Table 5 for details). The reports of all anomalies were reviewed and classified as major or minor malformations. The relative risk (RR) and 95% confidence intervals (CI 95) for major defects were calculated for each intergroup comparison. Secondary outcome measures included pregnancy outcome and offspring characteristics (gestational age at birth and birth weight). All data were entered into a spreadsheet for statistical analysis using Statview SE1Graphics (Abacus Concepts, Berkeley, California, 1987) for Macintosh computers. For the three-group comparisons of continuous data, a single-factor factorial, nonrepeated analysis of variance (ANOVA) was used and the Fisher’s Least Significant Difference post hoc test for multiple comparisons was used to identify mean differences between groups. Categorical data were compared using chi-square analysis (χ 2 ) or Fisher’s exact test, when appropriate, and significance for multiple categorical comparisons was determined by calculating an adjusted p value ( p adj ) for each computed observed p value ( p obs ) using the following formula: p adj 5 1 2 (1 2 p obs ) n comparisons (56). Meta-Analysis The databases Medline (1966–December 1999), EMBASE (1988–October 1999), and Current Contents (Jan–July 1997) were searched using the following criteria. The search items abnormalities, drug-induced, teratogen, and birth defect were combined using the Boolean operator OR. The search item glucocorticosteroid was combined with the previous search using the Boolean operator AND. A preliminary review of the titles (and abstracts when available) was made to determine whether the article was relevant to our topic. Abstracts of meetings published in the journals Teratology and Pediatric Research for the years 1995–1998 were also reviewed to cover new studies potentially not published yet as full papers. Bibliographies were reviewed for retrieved articles to identify any additional relevant articles. The reviewers were blinded by eliminating all reference to the

154

Park-Wyelie et al.

authors, journal, and study location. This was done to reduce the potential bias in the selection of articles, data extraction, and quality assessment. Controlled studies that examined first-trimester human systemic exposure for any corticosteroids, any doses, all indications, any duration, and all languages were included. Both case-control and cohort studies were acceptable for analysis; however, studies consisting of less than 10 corticosteroid-treated patients were excluded. Major malformations were defined using the criteria described by Heinonen et al. (1977) (57). Trials examining topical or inhaled steroids were excluded. A third author was used as an arbitrator in the case of discrepancies between raters. Inclusion of studies was agreed upon by mutual consent between reviewers and the reasons for exclusion were identified. The same procedure was followed during the data extraction and quality assessment scoring. Accepted studies had to report on the rates of major malformations in the corticosteroid and control groups. Data from the accepted studies were extracted in the form of 2 3 2 tables. The following four values were extracted from each study: number of neonates exposed to corticosteroids exhibiting major malformations, number of neonates exposed to corticosteroids without major malformations, number of neonates not exposed to corticosteroids with malformations, and number of neonates not exposed to corticosteroids without malformations. In the case of a woman with multiple pregnancies, each pregnancy was considered to be an independent event. Stillbirths and abortions were excluded from data extraction unless the study specifically mentioned assessment of malformations. The following demographic data of all the accepted studies were recorded where available: disease of the mother, number of neonates, types and doses of corticosteroid used, and type of birth defect. Abstracted data were entered into 2 3 2 tables using the Cochrane Review Manager version 4.0.3. software for IBM-compatible computers. For each of the studies, a MantelHaenszel summary and cumulative OR (with a 95% CI) was calculated. The null hypothesis (of no variation in outcome between n studies) was not rejected (and the pooling of outcomes was considered valid) if the chi-square test of homogeneity was less than the upper 95th percentile of the chi-square distribution with n 2 1 degrees of freedom. An OR of 1 reflected no effect of exposure on fetal outcome, whereas an OR . 1 reflected an increased risk for malformations among corticosteroid exposed groups. A sensitivity analysis was performed to determine the association between corticosteroids exposure and cleft palate. In addition, we calculated an incidence rate of major malformations from the included studies in addition to the studies that were rejected only because they had no control groups but which otherwise met the inclusion criteria. Assessment of the quality of the studies was performed using a quality assessment score for epidemiological studies published by us recently (58). Interrater agreement was measured for inclusion of articles, data extraction, and quality assessment of articles. The correlation between the quality of scores for our studies and the OR was tested in order to determine whether the quality of the studies was biasing our results. The mean, raw mean, and the meta-analytic mean were calculated for the incidence rate of major malformations.

RESULTS Prospective Study During 1985–1995, a total of 184 women met inclusion criteria and completed the postnatal follow-up. In the Motherisk cohort, prednisone was used for Crohn’s disease (by 34 women, 18%), asthma (by 30 women, 16%), ulcerative colitis (by 28 women, 15%), rheu-

Pregnancy Outcome After Corticosteroid Exposure

155

matoid arthritis (by 18 women, 10%), Bell’s palsy (by 7 women, 4%), transplant (by 8 women, 4%), lupus (by 20 women, 11%), sarcoidosis (by 4 women, 2%) and other indications (by 35 women, 19%). Although duration, dose and route of prednisone exposure varied, 138 (75%) women were exposed in the first trimester of pregnancy (Table 1). In the University of Toronto clinical trial, all women randomized to receive oral prednisone and aspirin took an average daily dose of 45 mg 6 14 mg prednisone during pregnancy. Compared to groups 1 and 3, women in group 2 were older; they had more pregnancies that ended in fetal loss and fewer live-born children (Table 2). Compared to the baseline controls (group 3), women in group 1 were more likely to be primigravidas and to have identified themselves as smokers. The rates of previous terminations of pregnancy and reported patterns of alcohol consumption among the three groups were not statistically different (Table 2). In group 1, a total of 184 women delivered 157 infants (three sets of twins); in group 2, a total of 123 women delivered 86 infants (one set of triplets, and two sets of twins); group 3 included 188 women who delivered 171 live-born infants. Women in group 2 were less likely to deliver live-born infants and more likely to experience fetal loss in the first 26 weeks (Table 3). The rate of live-born infants was similar between groups 1 and 3 (157/184 versus 171/188, p 5 0.2) however, the number of elective pregnancy terminations was higher in group 1 (16/184 versus 2/188, p 5 0.01). There was one stillbirth in group 3 and a fetal death (at 26 weeks) in each of groups 1 and 3. The proportion of male to female infants was similar among all three groups. When compared to babies delivered to mothers in groups 1 and 3, babies in group 2 were smaller, born earlier, and more likely to be premature (Table 3). When compared to babies in group 3, babies born to mothers in group 1 were smaller (mean 3112 versus 3428 g, p 5 0.0001), born earlier (mean 38 versus 39.5 weeks, p 5 0.0001) and more likely to be premature (27/158 versus 9/172, p 5 0.0003). Despite these differences, the majority of infants in groups 1, 2, and 3 were appropriate for gestational age (AGA) (139/ 157, 72/78, 158/168, respectively, p 5 0.2) and there was no preponderance of small for

Table 1

Characteristics of Prednisone Exposure in Groups 1 and 2

Therapy duration (weeks) Daily dose (mg) Route PO IV PO and IV Exposure #13 weeks #26 weeks 13–26 weeks Throughout .13 weeks .26 weeks Polytherapy Prednisone plus amniosalicylic acid Prednisone plus azathioprine Prednisone plus other

Group 1 (n 5 184)

Group 2 (n 5 123)

21 6 16 (161) 27 6 29 (173)

21 6 12 (100) 45 6 14 (122)

165/167 (99%) 1/167 (0.5%) 1/167 (0.5%)

126/126 (100%) 0 0

38/184 16/184 11/184 84/184 22/184 13/184 122/182 38/124 13/124 73/124

0 0 0 126/126 (100%) 0 0 124/124 (100%) 0 0 124/124 (100%)

(21%) (9%) (6%) (45%) (12%) (7%) (67%) (31%) (10%) (59%)

156

Table 2

Baseline Characteristics of Mothers in the Study Program p values Group 1 (n 5 184)

Maternal age (years) Gravidity G1 G2 Parity P0 P1 Prior miscarriages SA 0 SA 1 Previous abortions TA 0 TA 1 Smoking None Alcohol None

Group 3 (n 5 188)

30 6 5 (186)

33 6 4 (119)

31 6 5 (186)

75 (41%) 112 (61%)

7 (6%) 119 (97%)

86 (46%) 102 (54%)

0.0003

0.6

0.0003

83 (45%) 104 (56%)

91 (74%) 35 (28%)

51 (27%) 137 (73%)

0.0003

0.0003

0.002

153 (83%) 34 (18%)

7 (6%) 119 (97%)

151 (80%) 37 (20%)

0.0003

0.99

0.0003

173 (94%) 14 (8%)

110 (89%) 16 (13%)

175 (93%) 13 (7%)

0.3

0.99

0.02

154/185 (83%)

70/184 (83%)

175 (93%)

0.99

0.01

0.04

152/185 (82%)

7/84 (88%)

141 (75%)

0.97

0.3

0.8

Indicates significance at p 5 0.05. SA Spontaneous abortion. TA Therapeutic abortion.

1 vs. 2

1 vs. 3

2 vs. 3

a

a

Three groups 0.0003

Park-Wyelie et al.

a

Group 2 (n 5 123)

Pregnancy Outcome Characteristics p values

Outcome Live-born infants Miscarriage ,26 weeks Medical abortion Fetal death at 26 weeks Stillbirth Neonatal death Male: female Gestational age (weeks) Premature (,37 weeks) Birth weight (g) AGA: SGA : LGA a Vaginal delivery

Group 1 (n 5 184)

Group 2 (n 5 123)

Group 3 (n 5 188)

157 (85%)b 13 (7%) 16 (9%) 1 (0.5%) 0 0 86 : 68 38 6 3 (157) 27/158 (17%) 3112 6 684 (157) 139: 11 : 7 112/155 (72%)

86 (70%)c 38 (31%) 0 0 0 1 (0.8%) 35 : 43 35 6 3 (87) 63/87 (72%) 2614 6 716 (76) 72 : 3 : 1 47/68 (69%)

171 (91%) 13 (7%) 2 (1%) 1 (0.5%) 1 (0.5%) 0 81 : 84 39.5 6 2 (127) 9/172 (5%) 3428 6 578 (172) 158: 4: 6 131/172 (76%)

1 vs. 2

1 vs. 3

2 vs. 3

0.01 0.0003 0.01

0.2 1 0.01

0.0001 0.0003 0.96

0.99

0.7

0.9

d

d

0.9 0.7

d

d

0.3 0.9

Three groups

d

0.0001

d

0.0001

Pregnancy Outcome After Corticosteroid Exposure

Table 3

0.99 0.4

a

AGA 5 appropriate for gestational age; SGA 5 small for gestational age ,3rd percentile, LGA 5 large for gestational age .97th percentile. Includes three sets of twins (six infants). c Includes two sets of twins and one set of triplets (seven infants). d Indicates significance at p 5 0.05. b

157

158

Park-Wyelie et al.

Table 4 Classification of Birth Defect Status Between Groups p values

No anomalies Major anomalies a

Group 1 (n 5 111)

Group 2 (n 5 65)

Group 3 (n 5 172)

105 (94%) 6 (6%)

63 (92%) 2 (3%)

169 (98%) 3 (2%)

1 vs. 2

1 vs. 3

2 vs. 3

Three groups

0.71

0.16

0.62

0.23

Only major malformations that may be attributable to corticosteroids are included (see Table 5).

gestational age (SGA) or large for gestational age (LGA) babies (Table 3). There was no statistical difference in the rate of vaginal or cesarean section deliveries among the groups. When pregnancies resulting in multiple births were temporarily excluded from analysis of gestational age at birth and weight, mean differences remained statistically different. There was no statistical difference in the rate of major anomalies between the groups (group 1, 6/111; group 2, 2/65; group 3, 3/171, p 5 0.2) (Table 4). The eight major defects reviewed by the dysmorphologist and presented in Table 5 show no consistent phenotype when examined by group of exposure. The RR for major anomalies between groups 1 and 2 was 2.3 [95% CI 0.3,20.5] ( p 5 0.7); between groups 1 and 3, it was 2.0 [95% CI 0.5,9.0] ( p 5 0.4); and between groups 2 and 3, it was 0.9 [95% CI 0.1, 8.3] ( p 5 1.0). None of these was statistically significant. Table 5 Description of all Recorded Major Anomalies in Infants in the Prospective Study Abnormality

Group

Prednisone exposure

Polydactylya

1

Hirschprung’s disease Double-outlet right ventricle, valvar and subvalvar pulmonary stenosis, hypothyroid, hypospadias Multiple birth defects, congenital toxoplasmosisa Undescended testicle (required intervention)

1 1

5 mg/day throughout plus 50 mg/day azathioprine 80 mg/day from 2–2 3/7 weeks 30 mg/day from 0–4.5 weeks

1 1

Cleft palate, hypospadias

1

Severe cerebral palsy, one ventricle enlarged Erb’s palsy a (arm nerve damage due to delivery, hip click, neurological defects) Aortic valve stenosis Pyloric stenosis Dysplastic kidney (stillbirth)

2

a

2 3 3 3

unknown dose from 12–12.5 weeks plus cotrimoxazole 25 mg/day throughout plus 80 mg/day ASA b and 50 mg progesterone in first trimester 15 mg/day throughout plus 200 mg/day carbamazepine, 40 mg/day nifedipine, 100 mg/day atenolol 37.5–60 mg/day from 4–35 weeks plus 100 mg/day ASA until 35 weeks 37.5–22.5 mg/day from 5.5–35 weeks plus 100 mg/day ASA until 35 weeks None None None

These three malformations were not attributable to corticosteroids (polydactyly with family history; mental retardation following maternal toxoplasmosis; Erb’s palsy following birth trauma). These were excluded from the calculation of RR. b ASA 5 acetylsalicylic acid.

Summary of Patient, Drug, and Size of Study for Studies Included in the Meta-analysis

Study and year

Pathology of mother

Study type

# of cases

Popert ’62

Cohort

Warrell and Taylor ’68

Cohort

Heinonen et al. ’77 Mogadam et al. ’81

Cohort Cohort

Mintz et al. ’86 Robert et al. ’94 Czeizel and Rockenbauer ’97

Cohort SLE Case-control N/A Case-control Asthma, hay fever, rheumatoid arthritis, Addison’s disease, subfertility

204 1448 56,557

Rodriguez-Pinilla & Martinez-Frias, ’98 Carmichael & Shaw, ’99

Case-control N/A

12,304

Park-Wyllie et al., 2000 (present study)

Rheumatoid arthritis, SLE, ankylosing spondylitis, psoriatic arthropathy Asthma, eczema, ulcerative colitis, SLE, uticaria, sarcoidosis N/A Inflammatory bowel disease

Case-control Crohn’s disease, asthma, lupus Cohort

Crohn’s disease, ulcerative colitis, rheumatoid arthritis, SLE & other

Corticosteroid(s)

Dosage (prednisolone equivalent)

22

Prednisolone, cortisone, corticotropin

2.5–27.5 mg/day

69

Prednisolone

2.5–40 mg/day

Corticosteroid and/or corticotropin Corticosteroid or corticosteroid and sulfasalazine Prednisone Corticosteroids Dexamethasone, prednisone, cortisone, betamethasone, methylprednisolone, triamcinolone Prednisolone, hydrocortisone, prednisone, triamcinolone Prednisone, cortisone, dexamethazone, triamcinole Prednisone

N/A N/A

50,282 521

1396 372

10 mg/day N/A 5–100 mg/d

Pregnancy Outcome After Corticosteroid Exposure

Table 6

10–30 mg/d N/A 5–80 mg/d

SLE 5 systemic lupus erythematosis. N/A 5 not available.

159

160

Park-Wyelie et al.

Meta-analysis From the search of the literature, 455 articles were identified. Ten studies met our inclusion criteria and were entered into the meta-analysis (Table 6) (57,59–67). The Motherisk study (1997) described at the beginning of this manuscript was considered as a single study including only group 1 and 3 (67). The main reasons for excluding studies were absence of a control arm, inadequate reporting of fetal outcome, and inability to verify first-trimester exposure. Eight studies were excluded from the analysis because they lacked a control group; however, these studies reported sufficient information to calculate an incidence rate but not OR (68–75). Three studies were excluded because the number of treated patients was less than 10. Of the 10 accepted studies, 6 were cohort studies (57,59–62,67) and 4 were casecontrol studies (63–66). The time frame of study publications was 1962–1999. The studies included women with varying underlying diseases including rheumatoid arthritis, systemic lupus erythematosus, ankylosing spondylitis, asthma, inflammatory bowel disease, and arthropathy. Study sample sizes ranged from 22 to over 50,000 neonates. The specific corticosteroids and dosage regimens that were used by the mothers were not detailed in 4 of the 10 studies (57,61,63,66). Of the total, 4 studies examined the effects of fetal exposure to corticosteroids and other medications (61,65–67), while the remaining 6 studies followed women who were presumably only on corticosteroids. Of the 4 case-control studies, one focused solely on the presence of oral clefts as pregnancy outcome for women exposed to corticosteroids in the first trimester (65). When performing the Mantel-Haenszel Summary OR of cohort studies examining major malformations, statistical significance was not achieved [cumulative OR 5 1.45 (95% CI 0.81,2.60)] and the trials were homogeneous (χ 2 5 4.16 with five degrees of freedom) (Fig. 1) However, of the 6 trials included, Heinonen et al. (1997) did not separate major and minor malformations in the exposed group. Since one of our primary outcomes was major malformation rates in corticosteroid-exposed pregnancies compared with nonteratogen controls, analysis was repeated with and without Heinonen’s study, yielding marginally significant results: the Mantel-Haenszel Summary OR was 3.03 (95% CI 1.08,8.54) with a χ 2 5 2.40 with four degrees of freedom (Fig. 1).

Figure 1 Individual and Cumulative Mantel-Haentzel Summary Odds Ratio for Corticosteroid Exposed Cohort Studies for Major Malformations with and without Heinonen (1977).

Pregnancy Outcome After Corticosteroid Exposure

161

Table 7 Malformations Reported in Exposed and Control Subjects from the Cohort Studies Included in the Meta-analysis Corticosteroid-treated group Cleft palate* (3 cases) Ancephalic (2 cases) Congenital deafness Left ventricular atresia (died within 1 month) Hirschprung disease Double-outlet right ventricle; valvar and subvalvar pulmonary stenosis; hypospadias Undescended testicles Other major malformation (undefined) (6 cases)

Control group Aortic valve stenosis Pyloric stenosis Dysplastic kidneys Spina bifida Other major malformation (undefined)

* I also had microglossia and one with hypospadias.

Specific malformations reported by the cohort studies are listed in Table 7. The pooled sample of malformations in fetuses exposed to corticosteroids revealed that cleft palate was the most commonly reported anomaly with three cases being identified, compared to none among the controls. There were 4 case control studies examining the risk of oral clefts. All 4 had a significantly increased OR, with an overall OR of 3.35 [95% CI 1.97,5.69] (Fig. 2). For both cohort and case-control studies, Spearman’s rho did not show a significant correlation between the quality of the studies and their OR (r 5 20.32, p 5 5.38). An incidence rate of 3.5% was calculated using the meta-analytic means and the means of 15 studies. The raw average showed a very similar incidence rate of 3.9%.

DISCUSSION Although the list of epidemiologically proven human teratogens is small, labeling a medication as the cause of a major birth defect has serious clinical implications. In situations where known teratogens are drugs of choice (carbamazepine, warfarin, isotretinoin, lithium, methotrexate, phenytoin, valproic acid), counseling women of childbearing potential to use effective birth control to avoid drug exposure in gestation can possibly avoid fetal

Figure 2 Individual and Cumulative Mantel-Haentzel Summary Odds Ratio for Corticosteroid Exposed Case-Control Studies Focusing on Oral Clefts.

162

Park-Wyelie et al.

exposure. When exposure postconception is inevitable, however, evidence-based risk assessment is important for the woman in order to let her make an informed decision on the future of the pregnancy. Prospective Study While prednisone has never been proven to be a human teratogen, there are clinicians who extrapolate animal studies to suggest that the drug can cause cleft lip or palate. They therefore counsel their patients to avoid prednisone therapy during pregnancy. This may have been a factor for some of the 16 women in group 1, who chose to terminate their pregnancies after prednisone exposure. Prednisone is used for a variety of indications by women of reproductive age, making it very likely that risk assessment will be sought by many who intend to conceive while on prednisone therapy. The positive predictive value of animal testing for teratogenicity in humans is not high (76). Interspecies pharmacodynamic differences may account for the difference in susceptibility between humans and animals. Glucocorticoids may mediate their teratogenic effect in animals through a receptor that does not exist in the human embryonic palate (77). In light of the limited number of published pregnancy outcomes following prednisone use for recurrent fetal loss, our study reports outcome for 314 women exposed to prednisone and is the first and largest controlled study to specifically compare the rate of major defects between exposed women (111 with various diseases and 65 with a history of unexplained fetal loss) and a nonexposed cohort. The observed rates of anomalies (6, 3, and 2%, respectively) are within the expected baseline rate and none of the RR from the intergroup comparisons achieved statistical significance. The RR of 2.0 with confidence limits that span 1 may reflect inadequate power. For 2.0 to be statistically significant at an α of 0.05, 90% power, and a baseline incidence of major defects of 5%, we would need at least 300 infants in each group. As likely however, is that this RR of 2.0 reflects potential confounding by the underlying maternal diseases in our Motherisk exposed cohort. When women with prednisone exposure and relatively uncomplicated medical histories were compared to healthy women without prednisone exposure, the RR was 0.9 ( p 5 0.9). Combining groups 1 and 2 in order to pool all cases of prednisone exposure and contrasting this rate of major birth defects to that in the baseline group, we again report no association between prednisone exposure and major birth defects. It is not surprising that the women in group 2 delivered fewer live-born infants, as the association between autoantibodies and lupus anticoagulant in women with a history of recurrent fetal loss is well established (78,79). Recurrent fetal loss is an expected manifestation of numerous autoimmune diseases and an otherwise likely expectation for this population of women. It is most likely that the reduced birth weight and lower gestational ages at delivery of the babies in group 1 compared to baseline controls is due to the underlying maternal diseases. Among the six major defects that could be attributed to corticosteroids in groups 1 and 2, there was no apparent pattern. There was, however, one reported case of an infant born with an isolated cleft palate and impaired neuro-development. His mother took prednisone along with carbamazepine (not known to elevate the risk for cleft lip and/or palate) as well as nifedipine and atenolol throughout pregnancy. Although this does not confirm prednisone as a teratogen, in utero exposure did occur during the critical window of palate

Pregnancy Outcome After Corticosteroid Exposure

163

development (between the 45th and 84th days postconception) and prednisone is metabolized to prednisolone, which can be oxidized by placental enzymes to cortisone. There are two previous case reports of maternal cortisone use and cleft palate (44,45). We cannot, however, exclude the fact that this case may have arisen de novo, as the baseline incidence for cleft palate alone is 1 in 2500 (80). In group 1, there was one case of Hirschprung’s disease, which may be familial or may be associated with other syndromes. Although the mother of this infant took prednisone during pregnancy, prednisone may not have caused this outcome.

Meta-analysis To our knowledge, no systematic analysis examining the relationship between first-trimester exposure to corticosteroids and rates of major malformations has been published. Lack of well-designed trials that specifically address major malformations compound the controversy surrounding corticosteroid-induced malformations. The animal studies and case studies suggesting an association with cleft palate prompted us to critically review the existing evidence. The cumulative OR for cohort and case-control studies included in the meta-analysis showed a non-significant increased risk for major malformations associated with corticosteroid exposure, with the Heinonen study and a significant risk without this study (Fig. 2). However, a subanalysis of the cohort studies that specified the malformations revealed three cases of cleft palate among 390 corticosteroid exposures compared with no cases of cleft palate among 708 unexposed fetuses. Some of the studies could not be used because they did not contain any details as to the nature of the major malformations. Of the four case-control studies identified, one focused on the association between corticosteroid use in the first trimester and the incidence of oral clefts. Separately and when combined, the four studies produced a significant summary OR for oral clefts and the trials were homogeneous. Although our meta-analysis could only detect an increased risk for major malformations without the largest trial, it did show a greater than threefold increase in the risk for oral clefts when the fetus was exposed to corticosteroids during the first trimester. As important, it forwards evidence of clustering of cleft palate cases among cohort studies, which is consistent with the existing animal experience. Taken together, these data make epidemiological sense, as oral clefts make up only a small part of all major malformations. The apparent increased risk of oral clefts caused by corticosteroids has to be balanced against potentially serious implications for the mother and indirectly to the fetus if steroid therapy is discontinued or not initiated. Our study may allow clinicians to make a more informed decision for the use of corticosteroid therapy in pregnancy, showing a small but significantly increased risk of oral clefts after first-trimester exposure.

REFERENCES 1. Beitins IZ, Bayard F, Ances IG, et al. The transplacental passage of prednisone and prednisolone in pregnancy near term. J Pediatr 1972; 81:936–945. 2. Levitz M, Jansen V, Dancis J. The transfer and metabolism of corticosteroids in the perfused human placenta. Am J Obstet Gynecol 1978; 132:363–366.

164

Park-Wyelie et al.

3. Pasqualini JR, Nguyen BL, Uhrich F, et al. Cortisol and cortisone metabolism in the human placental unit at midgestation. J Steroid Biochem 1970; 1:209–219. 4. Crowley P. Antenatal corticosteroid therapy: a meta-analysis of the randomized trials, 1972 to 1994. Am J Obstet Gynecol 1995; 173:322–335. 5. Renisch JM, Simon JN, Karow WG, Gandelman R. Prenatal exposure to prednisone in humans and animals retards intrauterine growth. Science 1978; 202:436–438. 6. Walker B. Induction of cleft palate in rats with anti-inflammatory drugs. Teratology 1971; 4: 39–42. 7. Pinsky L, DiGeorge AM. Cleft palate in the mouse: a teratogenic index of glucocorticoid potency. Science 1965; 147:402–403. 8. Fainstat T. Cortisone induced congenital cleft palate in rabbits. Endocrinology 1954; 55:502– 508. 9. Fraser FC, Fainstat TD. The production of congenital defects in the offspring of pregnant mice treated with cortisone: a progress report. Pediatrics 1951; 8:527–533. 10. Schilsky RL, Sherins RJ, Hubbard SM, et al. Long-term follow-up of ovarian function in women treated with MOPP chemotherapy for Hodgkin’s disease. Am J Med 1981; 71:552. 11. Dara P, Slater LM, Armentrout SA. Successful pregnancy during chemotherapy for acute leukemia. Cancer 1981; 47:845. 12. Sinykin MB, Kaplan H. Leukemia in pregnancy. Obstet Gynecol 1962; 83:220–224. 13. Coulam CB, Zincke H, Sterioff S. Pregnancy after renal transplantation: estrogen secretion. Transplantation 1982; 33:556. 14. Nolan GH, Sweet RL, Laros RK, Roure CA. Renal cadaver transplantation followed by successful pregnancies. Obstet Gynecol 1974; 43:732. 15. diMalatesta F, Pierucci PF, Paolucci A, et al. Pregnancy and kidney transplantation: clinical problems and experience. Transplant Proc 1993; 25:2188–2189. 16. Tozman ECS, Urowitz MB, Gladman DD. Systemic lupus erythematosus and pregnancy. J Rheumatol 1980; 7:624–632. 17. Jones MM, Lidsky MD, Brewer EJ, et al. Congenital cytomegalovirus infection and maternal systemic lupus erythematosus. Arthritis Rheum 1986; 29:1402–1404. 18. Zulman JI, Talal N, Hoffman GS, Epstein WV. Problems associated with the management of pregnancies in patients with systemic lupus erythematosus. J Rheumatol 1980; 7:37–49. 19. Lockshin MD. Pregnancy does not cause systemic lupus erythematosus to worsen. Arthritis Rheum 1989; 32:665–670. 20. Meehan RT, Dorsey K. Pregnancy among patients with systemic lupus erythematosus receiving immunosuppressive therapy. J Rheumatol 1987; 14:252–258. 21. Branch DW, Scott JR, Kochenour NK, Hershgold E. Obstetric complications associated with the lupus anticoagulant. N Engl J Med 1985; 13:1322–1326. 22. Ros JO, Tarres MV, Baucells MV, et al. Prednisone and maternal lupus anticoagulant. Lancet 1983; 2:576–577. 23. Lubbe WF, Palmer SJ, Butler WS, Liggins GC. Fetal survival after prednisone suppression of maternal lupus anticoagulant. Lancet 1983; 1:1361–1363. 24. Cederqvist LL, Merkatz IR, Litwin SD. Fetal immunoglobulin synthesis following maternal immunosuppression. Am J Obstet Gynecol 1977; 129:687–689. 25. Tabbutt S, Griswold WR, Ogino MT, et al. Multiple thromboses in a premature infant associated with maternal phospholipid antibody syndrome. J Perinatal 1994; 14:66–70. 26. Sanchez-Guerrero J, Alarcon-Segovia D. Course of antiphospholipid antibodies in patients with primary antiphospholipid syndrome before, during and after pregnancy treated with low dose aspirin. J Rheumatol 1992; 19:1083–1088. 27. Ordi J, Barquinero J, Vilardell M, et al. Fetal loss treatment in patients with antiphospholipid antibodies. Ann Rheum Dis 1989; 8:798–802.

Pregnancy Outcome After Corticosteroid Exposure

165

28. Snyder RD, Snyder D. Corticosteroids for asthma during pregnancy. Ann Allergy 1978; 41: 340–341. 29. Fitzsimmons R, Greenberger PA, Patterson R. Outcome of pregnancy in women requiring corticosteroids for severe asthma. J Allergy Clin Immunol 1986; 78:349–353. 30. Schatz M, Patterson R, Zeitz S, et al. Corticosteroid therapy for the pregnant asthmatic patient. JAMA 1975; 233:804–807. 31. Wright CG, Rouse RC, Weinberg AG, et al. Vaterite otoconia in two cases of otoconial membrane dysplasia. Ann Otol 1982; 91:193–199. 32. Kraus AM. Congenital cataract and maternal steroid ingestion. J Pediatr Ophthalmol 1975; 12:107–108. 33. Cote´ CJ, Meuwissen HJ, Pickering R. Effects on the neonate of prednisone and azathioprine administered to the mother during pregnancy. J Pediatr 1974; 85:324–328. 34. Baxi LV, Pearlstone MM. Subchorionic hematomas and the presence of autoantibodies. Am J Obstet Gynecol 1991; 165:1423–1424. 35. Del Corso L, De Marco S, Vannini A, Pentimone F. Takayasu’s arteritis: low corticosteroid dosage and pregnancy. Angiology 1993; 44:827–831. 36. Al-Mofada SM, Osman MEE, Kides E, et al. Risk of thrombocytopenia in the infants of mothers with idiopathic thrombocytopenia. Am J Perinatol 1994; 11:423–426. 37. Kaplan C, Daffos F, Forestier F, et al. Fetal platelet counts in thrombocytopenic pregnancy. Lancet 1980; 336:979–982. 38. Rubins JM, Woll JE. Immune thrombocytopenic purpura. N Y State Med J 1981; 81:1743– 1747. 39. Hup JM, Bruinsma RA, Boersma ER, de Jong MCJM. Neonatal pemphigus vulgaris: transplancental transmission of antibodies. Pediatr Dermatol 1986; 8:468–472. 40. Zwi LJ, Becroft DMO. Intrauterine aplastic anemia and fetal hydrops: a case report. Pediatr Pathol 1986; 5:199–205. 41. Parke A, Maier D, Hakim C, et al. Subclinical autoimmune disease and recurrent spontaneous abortion. J Rheumatol 1986; 13:(6) 1178–1180. 42. Uribe LG, Thakur VD, Krane NK. Steroid-responsive nephrotic syndrome with renal insufficiency in the first trimester of pregnancy. Am J Obstet Gynecol 1991; 164:568–569. 43. DeWitte DB, Buick MK, Cyran SE, Maisels MJ. Neonatal pancytopenia and severe combined immunodeficiency associated with antenatal administration of azathioprine and prednisone. J Pediatr 1984; 105:625–628. 44. Doig RK, Cox OM. Cleft palate following cortisone therapy in early pregnancy. Lancet 1956; 2:730. 45. Harris JWS, Ross IP. Cortisone therapy in early pregnancy: relation to cleft palate. Lancet 1956; 1:1045–1047. 46. Lockshin MD, Druzin ML, Qamar T. Prednisone does not prevent recurrent fetal death in women with antiphospholipid antibody. Am J Obstet Gynecol 1989; 160:439–443. 47. Out HJ, Bruinse HW, Christiaens G, et al. A prospective, controlled multicenter study on the obstetric risks of pregnant women with antiphospholipid antibodies. Am J Obstet Gynecol 1992; 167:26–32. 48. Cowchock FS, Reece EA, Balaban D, et al. Repeated fetal losses associated with antiphospholipid antibodies. Am J Obstet Gynecol 1992; 166:1318–1323. 49. Silveira LH, Jara LJ, Saway S, et al. Prevention of anticardiolipin antibody–related pregnancy losses with prednisone and aspirin. Am J Med 1992; 93:403–411. 50. Silver RK, MacGregor SN, Sholl JS, et al. Comparative trial of prednisone plus aspirin vs aspirin alone in the treatment of anticardiolipin antibody-positive obstetric patients. Am J Obstet Gynecol 1993; 169:1411–1417. 51. Carp HJ, Menashe Y, Frenkel Y, et al. Lupus anticoagulant: significance in habitual firsttrimester abortion. J Reprod Med 1993; 38:549–552.

166

Park-Wyelie et al.

52. Karpatkin M, Porges RF, Karpatkin S. Platelet counts in infants of women with autoimmune thrombocytopenia. N Engl J Med 1981; 305:936–939. 53. Laskin CA, Bombardier C, Hannah ME, et al. Prednisone and aspirin in women with autoantibodies and unexplained recurrent fetal loss. N Engl J Med 1997; 337:148–153. 54. Shapiro L, Pastuszak A, Curto G, Koren G. Safety of first trimester exposure to topical tretanoic: a prospective cohort study. Lancet 1997; 350:1143–1144. 55. Pastuszak A, D’Alimonte D, Koren G. Astemizole use and pregnancy outcome. J Allergy Clin Immunol 1996; 98:124–126. 56. Cox D, Hinkley DV. Theoretical Statistics. London: Chapman & Hall 1974. 57. Heinonen OP, Sloan D, eds. Birth Defects and Drugs in Pregnancy: Maternal Drug Exposure and Congenital Malformations. Littleton MA: Publishing Sciences Group, 1977. 58. Seto AH. Meta-analysis of adverse neonatal effects due to maternal exposure to antihistamines. Thesis. Faculty of Pharmacy, University of Toronto, 1993. 59. Popert AJ. Pregnancy and adrenocortical hormones. BMJ 1962; 5283:967–972. 60. Warrell DW, Taylor R. Outcome for the foetus of mothers receiving prednisolone during pregnancy. Lancet 1968; 1(534):117–118. 61. Mogadam M, Dobbins WO, Korelitz BI, Ahmed SW. Pregnancy in inflammatory bowel disease: effect of sulfasalazine and corticosteroid on fetal outcome. Gastroenterology 1981; 80(1): 72–76. 62. Mintz G, Niz J, Gutierrez G, et al. Prospective study of pregnancy in systemic lupus erythematosus: results of a multidisciplinary approach. J Rheumatol 1986; 13:732–739. 63. Robert E, Vollset SE, Botto L, et al. Malformation surveillance and maternal drug exposure: the MADRE project. Int J Risk Safety Med 1994; 6:75–118. 64. Czeizel AE, Rockenbauer M. Population-based case-control study of teratogenic potential of corticosteroids. Teratology 1997; 56:335–340. 65. Rodriguez-Pinilla E, Martinez-Firas ML. Corticosteroids during pregnancy and oral clefts: a case-control study. Teratology 1998; 58:2–5. 66. Carmichael SL, Shaw GM. Maternal corticosteroid use and risk of selected congenital anomalies. Am J Med Genet 1999; 86:242–244. 67. Park-Wyllie L, Mazzotta P, Pastuszak A, et al. Birth defects following maternal exposure to corticosteroids: A prospective cohort study, and a meta-analysis of epidemiological studies. Teratology. 2000. In press. 68. Yackel DB, Kempers RD, McConahey WM. Adrenocorticosteroid therapy in pregnancy. Am J Obstet Gynecol 1966; 96(7):985–989. 69. Walsh DS, Clark FR. Pregnancy in patients on long-term corticosteroid therapy. Scot Med J 1967; 12:302–306. 70. Hack M, Brish M, Serr D, et al. Outcome of pregnancy after induced ovulation follow-up of pregnancies and children born after clomiphene therapy. JAMA 1972; 220(10):1329–1333. 71. Schatz M, Patterson R, Zeitz S, et al. Corticosteroid therapy for the pregnant asthmatic patient. JAMA 1975; 233(7):804–807. 72. Morris WI. Pregnancy in rheumatoid arthritis and systemic lupus erythematosus. Aust NZJ Obstet Gynecol 1969; 9(3):136–144. 73. Wells CN. Treatment of hyperemesis gravidarum with cortisone. Am J Obstet Gynecol 1953; 89(3):598–601. 74. Kenny FM, Preeyasombat C, Spaulding JS, et al. Cortisol production rate. Pediatrics 1966; 37(6):960–966. 75. Mercado AB, Wilson RC, Cheng KC et al. Extensive personal experience, prenatal treatment and diagnosis of congenital adrenal hyperplasia owing to steroid 21-hydroxylase deficiency. J Clin Endocrinol Metabol 1995; 80:2014–2020. 76. Jelovsek, FR, Mattison DR, Chen JJ. Prediction of risk for human developmental toxicity: how important are animal studies for hazard identification? Obstet Gynecol 1989; 74:624– 636.

Pregnancy Outcome After Corticosteroid Exposure

167

77. Pratt, RM. Receptor-dependent mechanisms of glucocorticoid and dioxin-induced cleft palate. Environ Health Perspect 1985; 61:35–40. 78. Gleicher N. Pregnancy and autoimmunity. Acta Hematol 1986; 76:68–77. 79. Rai RS, Regan L, Clifford K, et al. Antiphospholipid antibodies and β 2-glycoprotein-I in 500 women with recurrent miscarriage: results of a comprehensive screening approach. Hum Reprod 1995; 10:2001–2005. 80. Harper PS. Practical genetic counselling, 4th ed. Oxford, England: Butterworth-Heinemann, 1993.

12 Use of the Retinoid Pregnancy Prevention Program in Canada: Patterns of Contraception Use in Women Treated with Isotretinoin and Etretinate Anne Pastuszak and Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

Michael J. Rieder The Children’s Hospital of Western Ontario, London, Ontario, Canada

INTRODUCTION When thalidomide was demonstrated to be a potent human teratogen, it was immediately removed from the market. Conversely, isotretinoin and etretinate were introduced to the market with full knowledge of their animal teratogenicity, which had been established in preclinical animal studies (1). Isotretinoin (13-cis retinoic acid; Accutane) is used in the therapy of severe, cystic acne that has proven to be recalcitrant to standard therapy, and etretinate (Tegison) is prescribed for patients with psoriasis and other disorders of keratinization. Both are currently contraindicated in pregnancy. Based on the half-lives of elimination, women discontinuing isotretinoin are advised not to conceive before 1 calendar month has elapsed, and those stopping etretinate are frequently advised that the safe time to wait before conceiving has not yet been established. In some cases, the prescribing physician will define a waiting period of 2 to 3 years before the patient should conceive. Tragically enough, despite clear labeling of isotretinoin as contraindicated during pregnancy, birth defects consequent to in utero exposure were reported soon after its distribution. The pattern of isotretinoin embryopathy is now well known and includes serious craniofacial, central nervous system, cardiovascular, and thymic malformations. Ear abnormalities include microtia, low-set ears, and anotia; central nervous system defects include microcephalus, hydrocephalus, and reduction malformations of the brain; cardiovascular defects include transposition of the great arteries, tetralogy of Fallot, ventral septal defects, and aortic arch abnormalities (2).

Reprinted from Reproductive Toxicology 1994; 8(1)63–68. Copyright  1994 Elsevier Science Ltd.

169

170

Pastuszak et al.

Acknowledging that routine labeling of isotretinoin as teratogenic was not sufficient to prevent fetal exposure, the manufacturer, together with the Food and Drug Administration in the United States, developed a Pregnancy Prevention Program (PPP) for isotretinoin, which contains printed material to be used by the prescribing physician in educating the female patient about the serious risk for birth defects (3). Etretinate has not yet been conclusively labeled as a major human teratogen in the medical literature. However, because it is a synthetic retinoid and a derivative of vitamin A, like isotretinoin, extreme caution has been exercised when counseling women of reproductive age about conception during and after etretinate therapy. This unprecedented program contains an information brochure for patients, a qualification checklist, and a line drawing of a hypothetical isotretinoin-exposed fetus. In addition, female patients are asked to sign a consent form acknowledging that they have been instructed through the PPP, that they are aware of the need to use contraception during isotretinoin therapy, and that they will undergo pregnancy testing before, during, and after isotretinoin therapy. The PPP stresses and suggests the simultaneous use of two forms of contraception in an attempt to further reduce the likelihood of fetal exposure. The PPP for isotretinoin was distributed to all isotretinoin-prescribing physicians in Canada in October 1988, and a similar PPP for etretinate followed in 1989. Despite such a campaign to better inform women about the teratogenic risks of retinoids, fetal exposure to these drugs still occurs. Special attention needs to be paid to the potential reasons that lead to fetal exposure, because any strategy that aims to reduce such adverse occurrences must begin with an understanding of the mechanisms that lead to them. We prospectively analyzed a group of isotretinoin- and etretinate-exposed women who contacted the Motherisk Program for counseling about reproductive risks. Using a standardized questionnaire, we investigated and report maternal demographic characteristics, degree of use of the PPP, and pattern of contraception use by such women after they were seen by the physician who prescribed the synthetic retinoids.

PATIENTS AND METHODS Study Subjects The Motherisk Program is a teratogen information service in Toronto that counsels health professionals and members of the lay public about the reproductive safety of exposure to drugs, chemicals, radiation, and infections in pregnancy and lactation. A detailed description of the program can be found elsewhere (4,5). The majority of the callers are from Ontario; however, the service is known across Canada. In the case of isotretinoin and etretinate, the manufacturer, Hoffmann-La Roche Ltd., informed inquiring individuals of the services that Motherisk may provide. All women who voluntarily contacted the Motherisk Program from November 1, 1988, to January 30, 1991, requesting counseling in regard to maternal exposure to isotretinoin or etretinate were included in our study and seen in our clinic at their earliest convenience. If the patient was from a remote area of Ontario or another part of Canada, she was interviewed by telephone. These patients can be categorized as pregnant women who had exposed their fetuses in utero, women inquiring about the safety of a future pregnancy given a previous retinoid exposure, and women who conceived before the recommended waiting period had elapsed.

The Retinoid Pregnancy Prevention Program

171

A standardized intake questionnaire was designed and used to collect the following information for each patient: maternal age, ethnicity and marital status, gravidity, parity, previous elective and spontaneous abortions, gestational age at first visit to Motherisk, socioeconomic status, and pattern of ethanol and tobacco exposure. We also questioned about the method of birth control used during retinoid therapy and inquired about which contents of the PPP the patient recalled seeing in her physician’s office before she filled her retinoid prescription. Age and contraceptive use of women exposed to etretinate is purely observational, due to the small sample size. To determine whether these women were notably different from the typical patients using our service, isotretinoin-exposed cases were compared to a matched control group. These were patients counseled on the same day as the isotretinoinexposed case about nonteratogenic nonretinoids (drugs proven in large studies to be safe in pregnancy). This procedure controlled for month and year of visit to Motherisk because all cases are entered into our computerized database in chronologic order.

STATISTICAL ANALYSIS We compared the ages, mean gestational ages at first consultation, number of cigarettes smoked per day among smokers, gravidity, parity, and the mean number of spontaneous and therapeutic abortions between the two groups using a paired Student’s t-test. Socioeconomic status for both groups was estimated with the use of the occupational status titles and the Blishen and Carroll scale for women (6) and compared using the nonparametric Wilcoxon signed rank test. Distribution of marital status, ethnicity, and range of alcoholic drinks consumed per day were compared using the Fisher’s exact test.

RESULTS During the defined time period, 26 isotretinoin-exposed women and 7 etretinate-exposed women were counseled. Of the 7 etretinate-exposed women (mean age, 27.5 years ⫾ 4.1 years), 2 (29%) used no form of contraception either concurrent with therapy or for the number of months after discontinuation that was recommended by the prescribing physician, while 4 women (57%) used some form of contraception. One woman conceived 15 months after cessation of etretinate therapy despite having been instructed to wait for 3 years. By her own will, she stopped her oral contraceptives 1 month after discontinuing etretinate. Serum HPLC analysis in our laboratory at 12 weeks gestation detected no etretinate (lowest limit of sensitivity, 5 ng/mL), and none of the metabolites, acitretin and 13cis acitretin. Serial ultrasound examinations during pregnancy failed to detect malformations, and the patient gave birth to a morphologically normal baby in April 1991. The remaining six women did not conceive after etretinate therapy. The age of isotretinoin-exposed women was significantly lower than of their matched controls (25.2 ⫾ 6.7 years vs. 28.9 ⫾ 5.1 years, p ⫽ 0.03) (Table 1). They were also younger than women contacting Motherisk for other exposures and indications (29.9 ⫾ 3 years, p ⬍ 0.01). Nine of the isotretinoin-exposed women (34.6%) were less than 20 years of age (which we defined as adolescent), compared to only one woman (3.8%) in the matched control group ( p ⫽ 0.014). Women exposed to isotretinoin were more advanced in their pregnancy when they first contacted Motherisk than their matched

172

Pastuszak et al.

Table 1 Comparison of Demographic Characteristics and Secondary Exposure Patterns Between Isotretinoin-Exposed Women and a Matched Control Group Counseled by the Motherisk Program in Toronto

Mean age at first consultation Number of adolescents (ⱕ20 years) Marital status Single Married Data not available Ethnic background White Black Oriental Persian East Indian Data not available Mean gestational age (wks) at first consultation Referral to Motherisk Physician Self Number of alcoholic drinks/day 0 ⱕ1 1–2 Data not available Mean number of cigarettes/day among admitted smokers Mean socioeconomic (SE) index (0, lowest 100, highest) Mean gravidity Mean parity Mean number of previous therapeutic abortions Mean number of previous spontaneous abortions a

Isotretinoin cases (n ⫽ 26)

Matched controls (n ⫽ 26)

p value

25.2 ⫾ 6.7 9

28.9 ⫾ 5.1 1

P ⫽ 0.03 P ⫽ 0.014

11 12 3

8 18 —

NS a

22 3 1 — — — 10.2 ⫾ 8

18 0 1 1 2 4 6 ⫾ 4.2

NS

P ⫽ 0.01

20 6

21 5

NS

15 6 — 5 1.3 ⫾ 4.4

17 7 2 — 3.6 ⫾ 7.3

NS

NS

44.3 ⫾ 16.1

43.9 ⫾ 23.6

NS

⫾ ⫾ ⫾ ⫾

NS NS NS NS

1.7 0.5 0.1 0.2

⫾ ⫾ ⫾ ⫾

1.5 0.6 0.4 0.7

1.9 0.6 0.2 0.4

1.4 0.7 0.4 0.7

NS: P value ⬎ 0.05; mean ⫾ SD.

controls (10.1 ⫾ 8 weeks versus 6 ⫾ 4.2 weeks, p ⫽ 0.01). The two groups did not differ in their ethnic or marital status distributions, obstetrical or gestational histories, or in their use of cigarettes or alcohol (Table 1). Of the 26 isotretinoin-exposed women, 10 (38.5%) did not use any method of contraception during therapy, 6 (23.1%) failed their method of contraception (3 experienced condom failure, 2 failed the oral contraceptive pill, and the partner of 1 patient experienced failure of a vasectomy), and 2 women (8%) elected to stop their method of birth control and continue isotretinoin therapy. Only 8 women (30.8%) used some method of contraception (4 oral contraceptive pills, 3 abstinence, and 1 rhythm) during therapy.

The Retinoid Pregnancy Prevention Program

173

One might submit that those patients who used birth control were those who knew the drug was teratogenic. This, in fact, is not the case. Twenty of the 26 women using isotretinoin (77%) acknowledged receiving some warning about the drug’s teratogenic potential and the consequent need for effective birth control during therapy. Two women stated that they did not remember receiving such information, one refused to answer this question, and for three cases, data were not available because contact was made with a physician not responsible for the original isotretinoin prescription, not the patient herself. In the worst-case scenario, if we combined the 10 who used no birth control, the 6 who failed the method, and the 2 who discontinued the birth control and assumed that in each case the patient conceived during therapy, we potentially might have observed 18 cases of fetal isotretinoin exposure. Of these 18, only 1 (5.5%) saw all components of the PPP; the remainder were counseled with only single components of the program. Conversely, of the 8 who used birth control, 5 (62.5%) saw all components of the PPP; the remainder were counseled with only single components. These ratios are statistically significant (5.5% versus 62.5%, p ⫽ 0.004) and the clinical significance is clear. Of women who received etretinate, 100% acknowledged receiving some warning about fetal risk consequent to in utero exposure to etretinate.

DISCUSSION Our study focused on the patterns of use of the PPP for isotretinoin and etretinate and the pattern of contraception use in those women who voluntarily contacted the Motherisk Program. We are aware that we are contacted by women who use effective contraception during retinoid therapy and that those women who do ultimately contact us may not represent all retinoid users. Similarly, the frequency of contraception failure in our study is likely not representative of true failure statistics for all retinoid users. However, we feel that our preliminary results have identified an extremely important public health issue. Fetal exposure to isotretinoin cannot be eliminated and will, in fact, continue to occur in those patients who do not use contraception and in those who experience method failure. Physicians prescribing retinoids need to be aware that this is a clinical area of concern. More aggressive contraception counseling and perhaps, rigorous follow-up may be warranted for these patients. More than 38% of the isotretinoin-exposed patients did not use any form of contraception during isotretinoin therapy despite admitting that they were informed about serious fetal risks. While this figure may, at the outset, reflect a selection bias, there is a populationbased study that has reported alarmingly similar numbers. Using 1988 data from the Saskatchewan Database, which records all prescriptions of isotretinoin, Hogan et al. (7) observed that 35% of all women prescribed isotretinoin in that province did not use contraceptive measure even though the drug was known to be teratogenic at that time. Similarly, Lammer (2) has found in a retrospective analysis that 33% of U.S. women on isotretinoin therapy did not use concurrent contraception. Of this same group, 18% failed their method of contraception. The higher prevalence of young women in our group may explain the higher rate of contraception failure, and ineffective contraception in sexually active adolescents may explain the study rise in teen pregnancies (8). Our figure for contraception failure is not entirely surprising, because a large proportion of these women did not use a method with a low failure rate. Barrier methods have high failure rates (spermicides up to 12%, condoms to 6%, intrauterine devices up to

174

Pastuszak et al.

2.6%), when compared to oral contraceptive steroids (up to 2.6%), female sterilization (up to 0.6%), injectable depomedroxyprogesterone (0.3%), or implanted Norplant capsules (0.04%) (9–11). Pregnancy during isotretinoin therapy is counseled as an unwanted event by physicians and may be a possible factor contributing to contraception failure in our patients. This issue has been addressed by Grady and colleagues (11), who demonstrated that 4% of women seeking to prevent an unwanted pregnancy failed their method of birth control compared to 8.3% of those who sought to delay a wanted pregnancy. Our data suggest that young women are at a higher risk for unwanted fetal exposure to isotretinoin, and this trend may be consistent with the higher trend of unplanned pregnancies in adolescents in general. Hence, any strategy that aims to prevent pregnancy during isotretinoin therapy will need to address the unique difficulties in assuring effective contraception in young women. Women exposed to isotretinoin are likely to seek advice from a counseling service at more advanced stages of pregnancy than women exposed inadvertently to other medications during pregnancy. Assuming these patients have truly been warned by the prescribing physician, one may speculate that in addition to the unplanned nature of these pregnancies, denial, fear, and guilt may play a role in delaying appropriate steps by these pregnant women. It may also be argued that young women suffering from severe, cystic acne may represent a unique group, with long-standing difficulties in body image and sexuality. However, more studies are needed in order to prove that these characteristics have an impact on their approach to contraception. To reduce the chance of fetal exposure to isotretinoin, some countries have restricted dispensing of the drug to men only or prescription writing ability to dermatologists only (12). In Canada, however, the manufacturer has opted to structure the educational process of female patients who will receive isotretinoin. Our data indicate that merely informing women about the enormous teratogenic risk of isotretinoin is not sufficient to modify their sexual behavior toward using effective contraception. Although more than 75% of the isotretinoin-exposed women were warned a priori about the fetal risks associated with in utero exposure, 38% elected to use no contraceptive measures, and almost 25% experienced method failure with no second method being used as backup in case the first failed. We have observed that adherence to instructed contraceptive use tended to increase in women who were counseled with more components of the PPP for isotretinoin as compared with those who saw less of the package; only one woman who saw the whole program used no contraception. Ensuring better physician use of the PPP may help to decrease fetal exposure to retinoids; however, definitive conclusions will require more experience. Beyond the specific impact of gestational exposure to isotretinoin and etretinate, experience from current medical and regulatory initiatives that aim to prevent fetal damage from these drugs will help in crystallizing societal approaches to future, unique molecules that may be fetotoxic. Thalidomide, which was banned in 1961, is now used in groups of women with leprosy, who receive a monthly supply following injection of a systemic contraceptive hormone (13). The risk of contraception failure in sexually active women may prompt the necessity for more careful use and prescription of retinoids and the consideration of monthly prescription refills subsequent only to injection of a birth control agent. Society is painfully divided in its attitudes toward abortion. Presently, in Western countries, three babies are born for every aborted fetus (14). In young women who accept abortion as an option, early counseling after gestational exposure to retinoids is crucial. Our present data indicate that isotretinoin-exposed women tend to seek advice later than women exposed to other medications, thus shortening the period available for informed decisions. For those who do not accept abortion under any circumstances, an even stronger

The Retinoid Pregnancy Prevention Program

175

emphasis should be focused on well-counseled, stringent use of contraception. The current PPP does not address the woman’s and family’s attitudes to a pregnancy termination in case of fetal exposure to retinoids, despite this being a crucial consideration. This may reflect an unwillingness to confront a controversial and thorny issue, yet for such a preventative program to succeed, incorporation of this important determinant of pregnancy outcome should be considered. In summary, analysis of contraception use during isotretinoin therapy in patients contacting Motherisk reveals that an alarmingly high percentage of women do not use any contraception, and a similarly large group have failed their method despite their having been informed about the fetal risks. Our study shows that this group tends to be young and includes a high number of adolescents, which may suggest that preventative measures must address the special problems associated with unplanned pregnancies in this population of female patients. Our data demonstrate reasonable use of the PPP for isotretinoin in this first years of use and document a trend toward less inappropriate use of contraception in those patients who were counseled with the entire PPP, although larger numbers will be necessary to confirm this observation. ACKNOWLEDGMENT This work was supported by Hoffmann-La Roche Ltd. REFERENCES 1. Kamm JJ. Toxicology, carcinogenicity and teratogenicity of some orally administered retinoids. J Am Acad Dermatol 1982; 6:652–659. 2. Lammer EJ, Chen DT, Hoar RM, et al. Retinoic acid embryopathy. N Engl J Med 1985; 313: 837–841. 3. Pastuszak AL, Koren G. The retinoid pregnancy prevention program. In: Koren G, ed. Retinoids in Clinical Practice: The Risk-Benefit Ratio. New York: Marcel Dekker, 1993; pp 147–175. 4. Koren G, Feldman Y, Shear N. Motherisk—a new approach to antenatal counselling of drug/ chemical exposure. Vet Hum Toxicol 1986; 28:563–565. 5. Koren G, Pellegrini E. The Motherisk program. Drug Protocol 1990; 5:23–32. 6. Blishen BR, Carroll WK. Sex differences in a socioeconomic index for occupations in Canada. Can Rev Soc Anth 1978; 15:352–371. 7. Hogan DJ, Strand LM. Lane PR. Isotretinoin therapy for acne: a population-based study. Can Med Assoc J 1988; 138:47–50. 8. Russel JK. Sexual activity and its consequences in the teenager. Clin Obstet Gynecol 1974; 1:683–698. 9. Mishell DR. Contraception. N Engl J Med 1989; 320:777–787. 10. Sophocles AM, Brozovich EM. Birth control failure among patients with unwanted pregnancies: 1982–1984. J Fam Pract 1986; 22:45–48. 11. Grady WR, Hirsch MB, Keen N, Vaughan B. Contraceptive failure and continuation among married women in the United States, 1970–1975. Stud Fam Plann 1983; 14:9–19. 12. Lancaster PAL. Teratogenicity of isotretinoin. Lancet 1988; 1:1255. 13. Lane PR, Hogan DJ, Moreland AA. Making the use of isotretinoin safer. Can Med Assoc J 1989; 141:376. 14. Koren G, Bologa M, Pastuszak A. The way women perceive tertogenic risk: the decision to terminate pregnancy. In: Koren G ed. Maternal-Fetal Toxicology: A clinician’s Guide. New York: Marcel Dekker 1990, pp 373–381.

13 Drugs and Breast-Feeding Anna Taddio and Shinya Ito The Hospital for Sick Children, Toronto, Ontario, Canada

Clinical Case A woman who is breast-feeding her 2-month-old infant asks you if she can use fluoxetine during breast-feeding. What would you advise her to do?

INTRODUCTION This chapter reviews the safety of commonly used medications in lactating ambulatorycare patients. These medications include antibiotics, analgesics, antihypertensives, and many others. Breast-feeding mothers may require drug therapy and must make informed decisions about the safety of possible drug consumption by their nursing infants. We hope to provide nursing mothers and their health care providers with some general guidelines for these situations. The mother’s decision to breast-feed during drug therapy should be based on a reasonable understanding of any risks to her infant. The choice to avoid breastfeeding during drug therapy should be made where there is an unacceptable risk to the baby. In many cases, mothers may continue to breast-feed while on drug therapy with little risk to the child. In some situations of short-term courses of drug therapy, a mother may wish to suspend breastfeeding temporarily. It is not known how many women stop breast-feeding due to their concern for drugs in their breast milk. The use of medication, however, is believed to contribute to a mother’s choice to bottle-feed. Other factors that affect this decision are the mother’s parity, education level, social class, psychosocial factors, and socioeconomic class. The type of delivery, and the presence of illness in mother or baby are also contributory factors (1,2). Studies have shown that drug use in the postpartum period is high. One study reviewed medication use in 970 postpartum women in Norway in 1980. According to medical records, almost 98% of these women were breast-feeding when discharged from hospital. Ninety percent received at least one drug during the early postnatal period. The most frequently used drugs were analgesics (mainly codeine and dextropopoxyphene), hypnotics (mainly nitrazepam), and methylergometrine or ergometrine (3). A similar study examined medication use in 2004 postpartum women in Northern Ireland in 1982. Of these mothers, 33% were breast-feeding at the time of discharge from

177

178

Taddio and Ito

hospital. More than 99% of all postpartum women received at least one drug during the first week after delivery. The drugs used encompassed the following drug classes; antibiotics, analgesics, psychotropics, antiepileptics, gastrointestinal drugs, cardiovascular drugs, antihistamines, steroids, hematinics and vitamins, and miscellaneous drugs. The most commonly prescribed medications included iron and vitamin preparations and analgesics. The pattern of prescribing was not much different between breast- or bottle-feeding mothers except for sedative and iron and vitamin use. The mothers that did not breast-feed used sedatives and iron/vitamin preparations more often (4). The use of medication by breast-feeding women implies that nursing infants may be exposed to the medication through ingestion of breast milk. The magnitude of the risk to the baby from this practice determines the drug’s safety, and it must be evaluated in each patient’s case. To assist in this decision, this review summarizes determinants of drug excretion into milk and the methods used to estimate a nursing infant’s dose or exposure. An extensive summary of data obtained from selected studies on drug excretion into milk is provided. Some conclusions are drawn regarding the safety of these medications during lactation. The health professional reading this chapter should note that we do not recommend the use of any drug without the advice that all nursing infants be monitored for any adverse effects during maternal drug therapy. The intensity of this monitoring will vary with the potential risk and with the mother’s own concerns.

BACKGROUND Impact of Breast-Feeding on Infant Health The most recent statement by the American Academy of Pediatrics states that breastfeeding is the optimal mode of nutrition for infants in the first 12 months of life (5). The benefits of using human milk for infant feeding are improved infant nutrition, health, immune function, development, and intellectual performance (5). Breast-feeding has benefits for mothers, families, and society also. These include improved health, psychological, social, economic, and environmental benefits (5). Despite the overwhelming evidence of the advantages of breast-feeding, relatively few women breast-feed their infants for as long as 12 months (5). Many factors lead women to discontinue breastfeeding their infants prematurely; one is maternal use of medications. The benefits of breast-feeding should be considered whenever medications are required by nursing women, for, in many cases, breast-feeding need not be interrupted during maternal drug use. Breast-Milk Production Breast milk is synthesized in the mammary tissue. The alveolar cells make up the functional units of this gland and are responsible for secreting newly formed milk. A lactating breast can be compared to a cluster of grapes, where each grape consists of a cluster of alveolar cells and a central lumen (6). Capillaries surround the alveolar cells (7). These cells discharge milk products into the lumen, where they are transported to the duct system. The ducts meet in channels of increasing size until the nipple is reached (6). The process of lactogenesis is under the control of many hormones; principal among these are estrogen, progesterone, prolactin, and oxytocin. Estrogen and progesterone are

Drugs and Breast-Feeding

179

necessary for the maturation of the mammary gland during pregnancy. These hormones are secreted under the control of the pituitary and placental follicle stimulating hormone (FSH) as well as luteotropic hormone (LH). They have an inhibitory effect on milk production. At the end of pregnancy, the alveolar cells are not completely developed and milk production is relatively slow. Just prior to parturition, the levels of estrogen and progesterone decrease rapidly. The inhibitory effect of these hormones on prolactin are thereby removed (8). Prolactin is the major stimulus for milk production and secretion, and it is released from the pituitary gland under the stimulus of infant suckling (9). Regular milk production is stimulated by the actual nursing period by the infant and continues after the nursing episode. Breast emptying during nursing is prompted by oxytocin, which promotes the contraction of the myoepithelial cells surrounding the alveoli and ducts and causes the ‘‘milk ejection reflex’’ (8). Other supportive hormones include insulin, cortisol, thyroidparathyroid hormone, and growth hormone (10).

Composition of Breast Milk Breast milk contains water, proteins, electrolytes, lipids, carbohydrates (mainly lactose), vitamins, minerals, and immune factors. Its composition changes over time during the course of lactation, which can be divided into three periods according to these changes. The first milk or colostrum, produced by a lactating mother is formed in the first 5 days postpartum. Transitional milk is produced after colostrum. Mature milk follows transitional milk, approximately 2 to 3 weeks postpartum (11). Mature milk contains higher concentrations of lactose and fat than colostrum does, but its protein and immunoglobulin content is lower (12). Water constitutes about 88% of mature milk, while lactose, lipids, protein, and mineral constituents make up 7, 4, 1, and 0.2%, respectively (13). Although the overall composition of mature milk is not thought to vary significantly (14), concentrations of lipid, fat-soluble, and water-soluble vitamins in milk do change according to the mother’s nutrient intake (15). There is a correlation between blood flow and milk production (16). Milk volume is usually low during the first 2 days postpartum but increases rapidly and levels off by 5 days (17). Milk yields reach approximately 800 mL/day or more by 6 months (17). The average daily intake by infants is variable and may determine milk yield (18). The average pH of milk is lower than that of plasma. Morriss (19) has shown that the average pH of milk also changes with time; it is 7.45 in colostrum and 7.0–7.1 for transitional and mature milk. The pH gradually increases during lactation until it reaches 7.4 at 10 months. Milk composition varies throughout the day. There is diurnal variation in fat content, the highest fat levels being found in the morning (20–22). Infants will generally feed 10–15 minutes on each breast, and milk composition is variable during that time. The hindmilk contains more fat than foremilk (21). The lipid content of mature milk also increases with time (23). The differences between milk from premature and term deliveries have been summarized by Anderson (24). Mother’s milk from early lactation in preterm deliveries is higher in fat and protein, lower in lactose, and more varied in its composition compared with milk from mothers with full-term babies. Many of the differences between preterm and term milk, however, become less prevalent as the mother continues to nurse. Drugs ingested by nursing mothers are not known to affect the maturation or pH of milk. They may, however, affect the quantity of milk produced or its composition.

180

Taddio and Ito

Bromocriptine, metoclopramide, and oral contraceptives represent a few examples of such drugs. The implications of the use of these medications are discussed later in this chapter.

DRUG EXCRETION INTO BREAST MILK Determinants of Infant Exposure Drugs taken by lactating women may be present in their infants through their consumption of breast milk. The extent of this exposure will depend on the pharmacokinetic properties of the drug in the child and mother. Breast physiology, milk composition, and infant suckling pattern also determine the extent of drug exposure in the infant. Infant Drug Handling There are differences between infants and adults in each of these factors: drug absorption, drug distribution, drug metabolism, and drug excretion. In a review of these factors, Besunder (25) concluded that neonatal gastric acid secretion and gastric emptying are both decreased compared to those in adults; the values for total body water of a newborns are higher than those for adults; fat content is lower, proportionally, than in adults; there is decreased protein binding in neonatal plasma; and there is decreased oxidative metabolism of drugs at birth. The capacity of this metabolizing pathway, however, increases in the first few weeks of life and supersedes adult capacity in childhood. The infant’s capacity for drug conjugation and glucuronidation of drugs is lower at birth. Sulfation, however, is relatively well developed in the newborn (26). Kidney blood flow, glomerular filtration, and tubular function are all reduced in the neonatal period. The infant’s capacity does not reach adult levels until approximately 6 months (25). The decreased clearance rate in neonates has been used to provide a summary of drugs incompatible with breast-feeding based on likely infant plasma levels achieved during maternal therapy (27). The susceptibility of an infant to adverse effects from drugs also depends on the pharmacokinetic characteristics of the drug and the sensitivity of the infant. Adverse reactions can be divided into two groups; those that are dose-related and reflect the pharmacological actions of the drug and those that are idiosyncratic reactions. Many of the adverse reactions reported in breast-fed infants consuming drugs via milk are dose-related, where relatively large doses of the drug have been delivered to the infant via breast milk. The rest of the reports reflect an exaggerated response to the agent by the infant. Idiosyncratic reactions are not dose-related, and it is often difficult to quantify the risk of such reactions occurring. Fortunately, they are not common in breast-fed infants. Summary of the Literature Regarding Drug Excretion in Breast Milk and the M/P Ratio Drugs must pass from maternal mammary capillaries to the alveolar cells before they can be transferred into milk. During this process, drugs must cross a series of membrane barriers (6). Drug molecules may pass through membranes in two ways—passive diffusion and active transport. Lipid-soluble drugs may dissolve in the lipid phase of the membrane and diffuse through it. Water-soluble drugs may diffuse through water-filled pores. Carrier proteins may also transport drugs across membranes (28). The rate of drug passage de-

Drugs and Breast-Feeding

181

pends on the degree of ionization of the drug, lipid solubility, size, and concentration gradient (28). A drug transferred into milk may exist in association with the various milk components (29), such as the aqueous layer, the proteins, or fat. The protein-bound portion of a drug is not available for transfer into milk. Thus, a drug that is highly bound to maternal plasma protein may not appear in breast milk to as great an extent as a drug that is not protein-bound. Propranolol and warfarin are examples of highly protein-bound drugs that do not cross into breast milk in appreciable amounts. The milk-to-plasma ratio (M/P ratio) is a ratio of the concentration of a drug achieved in milk compared to maternal plasma. It is utilized as an estimate of the dose of a drug delivered in milk as a function of the amount in maternal plasma. Different models have been described to explain and predict how drugs are transferred into breast milk (30); however, a two-compartment model is usually sufficient to describe drug transfer in milk. It assumes a central compartment and a peripheral compartment that includes milk. In this model, there is a time dependency of the M/P ratio as a function of distribution characteristics. A method for deriving other mammary models has been described (31). Drug diffusion between plasma and milk is a dynamic and reversible process. Thus, most drugs are not permanently ‘‘trapped’’ in milk. For nursing mothers, this implies that they need not ‘‘pump and dump’’ milk in an attempt to promote drug elimination. This practice may be better suited for other purposes, such as alleviating the pain from engorged breasts or maintaining an adequate milk flow. Numerous clinical trials have measured the M/P ratios of different drugs. The values from these trials have provided health professionals with some measurements as a basis for assessing the safety of drug usage during breast-feeding. However, some reported values have questionable validity because they were obtained from studies with imperfect designs. Namely, investigators have often failed to assure steady-state conditions, to measure metabolites, to account for infant suckling pattern, or to study an adequate number of patients. This topic has been reviewed by Wilson (30); a summary is provided. Many trials report an M/P ratio obtained from only one simultaneously obtained milk and plasma drug level. Clearly, if the M/P ratio varies with time, this will not produce an accurate estimate of the amount of drug the infant may consume. The M/P ratio assumes that a constant ratio is maintained between the maternal plasma drug level and the milk drug level. For drugs in which the plasma and milk concentrations parallel each other over time, the M/P ratio is ideally calculated by taking the area under the curve (AUC) time-course profile in milk and plasma and equating these values in a ratio. One of the basic assumptions for calculating an accurate M/P ratio is that the mother is at steady state. This is because drugs may accumulate in breast milk as a function of dosing. The M/P ratio calculated before steady state is achieved may be an underestimate of the M/P ratio at steady state. Another factor that is overlooked in calculating the infant dose from drugs in breast milk is the contribution of metabolites with additional pharmacological activity (i.e., active metabolites). Metabolites may be formed in the maternal circulation or in breast tissue and transferred to the infant via breast milk. The M/P ratio of drug metabolites is not always measured. Case reports that measure metabolites offer valuable information on the overall transfer of a drug in breast milk. However, one of the pitfalls of many case reports is that they provide sparse details of the analytical methods used to achieve this. The effect of infant suckling on the M/P ratio is not always known. Clinical trials often do not incorporate infant feeding between measurements. It is known that a long

182

Taddio and Ito

feeding time is associated with an increase in the percentage of lipid in the milk. Thus, for lipid-soluble drugs, the M/P ratio may increase throughout a feeding interval, and this may not be detected in the trial. The calculation of an M/P ratio may also vary depending on the type of milk sample obtained. Some trials analyze the first few drops of milk expressed, while others average the amount collected over an entire feeding interval. The M/P ratio may vary according to the maturity of the milk as well. Trials should report the exact stage of lactation when drug analyses were made. Finally, many of the trials involving measurement of drugs in breast milk utilize small sample sizes. This may be due to factors such as insufficient number of women taking medications while breast-feeding, difficulty in retrieving consent and meeting study criteria, unavailability of assay methods, and perhaps a perception that drug consumption by infants through breast milk is negligible. Results often show considerable variability within and between subjects. It is clear that conclusions drawn from such studies cannot be extrapolated to all lactating women with a measurable degree of certainty. Estimating M/P Ratios and Drug Excretion into Breast Milk Studies of drug transfer into milk are not available for many agents, and the M/P ratio is now known. Another method, however, can be employed to estimate a nursing infant’s drug consumption. It involves calculating an M/P ratio from information known about the physicochemical properties of the drug. Much research has been done on drug transfer into milk in animal species. The principles of these findings are often used to estimate the M/P ratios of drugs if they are otherwise unavailable. Animal studies of drug transfer into milk have shown that the distribution of drugs between maternal plasma and milk can partially be explained using pH partition theory (32–35). Drug transfer occurs by passive diffusion of the un-ionized species; the degree of ionization at plasma pH determining the amount available for diffusion into milk. This can be determined using the pH partition theory (36) and the pKa of the drug: un-ionized ionized ionized Basic drugs: pk a ⫽ pH ⫹ log un-ionized Acidic drugs: pk a ⫽ pH ⫹ log

(1)

Because the pH of milk is generally lower than that of plasma, the concentration that an electrolyte achieves in milk will depend on the pK a of the drug. When the M/P ratio has not been measured, the milk-to-plasma ultrafiltrate ratio (M u /Pu ) (36) may be calculated instead by using the Henderson-Hasselbalch equation. This ratio describes drug partitioning into ultrafiltrate fluids devoid of lipids or proteins. For acidic drugs: M u /Pu ⫽ For basic drugs: M u /Pu ⫽

1 ⫹ 10(pH m ⫺ pKa ) 1 ⫹ 10(pHp

⫺ pK a )

1 ⫹ 10(pK a ⫺ pH m) 1 ⫹ 10

(2)

(pK a ⫺ pH p )

where pH p ⫽ pH of plasma, pH m ⫽ pH of milk. As most drugs are weak acids or weak bases, they will partition into breast milk according to their degree of ionization in the two mediums. Most acids are present in

Drugs and Breast-Feeding

183

dissociated and undissociated forms in milk and plasma. Since only the undissociated form can pass from plasma to milk, an equilibrium is established across the mammary membrane and the concentration of the undissociated drug is equal in both ultrafiltrates. However, since the plasma pH is slightly higher, a larger portion of drug will be present in plasma than milk in the dissociated state. Thus the M u /Pu ratio is less than 1. The situation is reversed for weak bases. Bases exist in dissociated and undissociated forms in plasma and milk. Due to the lower pH in milk, a larger portion of drug will be ionized. Thus, the drug concentration in the milk ultrafiltrate will be higher than in the plasma and weak bases can achieve higher concentrations in the milk ultrafiltrate than in the plasma. This phenomenon is called ion trapping. Un-ionized drugs will have M u /Pu ratios of approximately 1. The amount of drug available to cross into milk is dependent on protein-binding. This is because only the unbound portion is able to diffuse into milk. Binding to milk proteins may also occur, and predictions of M/P ratios without considering protein-binding may mislead conclusions (37). The Henderson-Hasselbalch equation can be modified to allow for correction of protein-binding of drugs in plasma and milk. The calculated milkto-plasma ratio then reflects the ratio of drugs in skim milk to plasma (M skim /P) (38). In this case, the milk is devoid of lipid constituents. M skim M u plasma f u ⫽ ⫻ P Pu milk f u

(3)

where f u ⫽ 1 ⫺ fraction protein-bound. Milk also contains other constituents that can alter drug concentration, such as lipids. Atkinson and collegues (39) have designed a detailed model accounting for drug octanol : water partition coefficient (or lipophilicity) and plasma protein-binding. This model thus estimates a whole milk to plasma ratio (M/P) instead of a M u /Pu or M skim /P ratio. The specific model is: For basic drugs: ln M/P ⫽ 0.025 ⫹ 2.3 ln(M u /Pu ) ⫹ 0.9 ln(Fup ) ⫹ 0.5 ln K For acidic drugs: ln M/P ⫽ ⫺0.405 ⫹ 9.4 ln(M u /Pu ) ⫺ 0.7 ln(Fup ) ⫺ 1.5 ln K where K ⫽ (0.955/f um ) ⫹ (0.045 ⫻ milk: lipid Papp ), f up ⫽ fraction of drug unbound in plasma (1 ⫺ plasma protein-binding), Papp ⫽ apparent partition coefficient at pH 7.2. An explanation of the derivation and use of the model is found in the cited paper. As mentioned, experimentally or theoretically derived M/P ratios are used to estimate the amount of drug excretion in milk. For example, the drug concentration in milk (C m ) can be calculated as follows: C m ⫽ M/P ⫻ C ss

(4)

where C ss is the average maternal plasma level at steady state, which is usually obtained from reference texts. If it is not known, the C ss can be estimated using the following wellknown equation: C ss ⫽

R⫻F CL

(5)

where R ⫽ dose rate, F ⫽ bioavailability, CL ⫽ clearance. These variables are also obtained from reference texts.

184

Taddio and Ito

Once the C m is calculated, it can then be used to estimate infant drug exposure. This is done by estimating the volume of milk (Vm ) that is ingested by a breast-feeding infant per day. An average value of 150 mL/kg/day (29) is often used. The following equation can then be used to calculate the infant dosage (D inf ) per unit time (29). D inf ⫽ C m ⫻ Vm

(6)

In addition to this, one can compare the infant dose with the maternal dose or with the therapeutic dose of the drug in infants. The former option involves calculating the percentage of maternal dose on a mg/kg basis, or the fraction of the dose that the infant receives (Finf ). It is obtained in the following way: Finf ⫽

D inf ⫻ 100 maternal dose

(7)

Another approach used to estimate infant drug exposure takes into account the M/P ratio and the clearance of the drug by the infant. The formula calculates the infant exposure index (40), which is defined as the percentage of the infant’s therapeutic dose per body weight that would be ingested through breast milk. The formula is: Exposure index (%) ⫽ (A ⫻ M/P) / Cl i

(8)

where A is a coefficient (10 mL/min/kg) representing milk intake ⫻ 100, M/P is the milk-to-plasma drug concentration ratio, and Cl i is the clearance of the infant. When Cl i is unknown, then the adult Cl is used as an estimate of Cl i . It is apparent from this formula that drugs with a high Cl i (even in the presence of high M/P) result in relatively low infant exposure. Drugs with low Cl i and low M/P will result in significant infant exposure. These calculations have been used by many investigators to estimate the drug consumption of nursing infants. It should be noted, however, that they are simply estimates of infant exposure to drugs through breast milk. The values obtained by these equations are not sufficient to estimate risk to the infant. Other variables will affect the risk-benefit analysis for drug exposure during breast-feeding. These include the pharmacology of the drug, the presence of active metabolites, the dose and length of therapy, the disposition of the drug in infants, infant age and clinical status, quantity of milk consumed, utilization of the drug in pediatric medicine, and possible effects of the drug on lactation. Most importantly, the probability of developing adverse reactions must be considered when risk is assessed. Breast-feeding mothers should not be encouraged to take medications unless it is necessary. When drug therapy is required during lactation, the safest in a therapeutic class should be chosen and the lowest dose that will achieve the therapeutic goal should be used. If the milk-plasma drug profile has been clearly defined, mothers can also tailor their dose times to avoid peak milk levels during infant feedings. Information Sources on Drugs and Lactation There are many reviews on the topic of drug excretion in breast milk. The most extensive review is from the World Health Organization (WHO) Working Group. In the text Drugs and Human Lactation (41), the WHO Working Group reviews breast physiology,

Drugs and Breast-Feeding

185

mechanisms of drug excretion in milk, clinical trials, and case reports involving drugs and breast milk. They also make recommendations on the safety of the drugs that are reviewed. Another source of information is the text Drugs in Pregnancy and Lactation (42). Here some of the findings from clinical trials and case reports are summarized. The recommendations of the American Academy of Pediatrics (AAP) Committee on drugs are also given; the AAP publishes a summary of drugs considered compatible with breastfeeding (43). Guidelines for studies on drug passage into milk have also been developed by the WHO Working Group (41). They address factors such as sample size of study, drug assay methods, and general study design. Clinical trials of drug excretion in milk are difficult to conduct and the case report makes up much of the information that is available in this area. A brief discussion of the clinical usefulness of case reports is therefore warranted. Obviously, case reports often provide information that is otherwise unavailable. They identify possible adverse effects of drugs from exposure through breast milk. For example, acetaminophen is generally regarded as safe during lactation. However, Matheson (44) reported on a 2-month-old infant who developed a rash after maternal ingestion of acetaminophen. Upon rechallenge at a later date, a similar type of rash developed. The milk concentrations found were similar to those previously reported. Accordingly, it is apparent from this case report that recognizing potential drug-related reactions is important during breast-feeding. Case reports can also verify that some drugs whose safety is suspect may indeed cause side effects in the breast-feeding neonate. For example, Atkinson (27) recommended that atenolol be avoided by nursing mothers with infants below 44 weeks postconceptual age. He based his recommendation on the estimated infant concentrations that would be achieved. Countering this concern, atenolol has been studied in puerperium lactating women without adverse effects. In a case report, however, Schmimmel (45) reported a neonate who suffered from signs of β-adrenergic blockade while her mother was taking atenolol. This was the first report of toxic effects caused by atenolol through breast-feeding and it lends credibility to the models that predicted it could occur. Case reports are often simpler to produce than a clinical trial; only 1 patient may be involved and the time committment may be proportionately smaller. Results are also rapidly available to the scientific community because the analysis of one patient is shorter than for more patients. Case reports may also be useful in introducing analytical methods that can be used by others to analyze drugs in breast milk. The potential disadvantage of case reports is that the events they describe may be interpreted by health professionals as common phenomena. It must be remembered that a single case report does not reflect the risk or safety from drug exposure for all nursing infants. In addition, it does not always prove a cause-and-effect relationship between drug exposure and effect. The information should be extrapolated to other settings with extreme caution. Whenever possible, it is best to incorporate the case report into an overall risk assessment together with other available clinical data. This includes information such as investigations of other available treatment modalities, consideration of infant age and clinical status, and observations of the baby’s breastfeeding pattern. Table 1, which includes data obtained from clinical trials and case reports, shows a summary of the excretion of selected drugs in milk. Recommendations on the safety of selected drugs during nursing follows thereafter.

Summary of Data Obtained from Studies of Drugs in Breast Milk Max. Finf

Standard adult dose

2–22 months

0.91

4.15

325–650 mg q4–6h

SD

2 months

1.02

1.28

4

SD

2–8 months

1.1

1.85

119 120

3 1

SD SD

? 7 weeks

— 0.40 (0.45)

1.58 1.29 (1.35)

as above

1

SD

13 weeks

121

2

MD

8–72 hours

0.59 (0.61) —

122

5

MD

1 day

1.47 (1.5) 0.61 (1.27) 1.57 (2.2)

123

5

SD

⬎1 month

—

6.0

122

5

MD

1 day

—

0.65 (0.89)

Yes

124

1

MD

3 weeks

—

2.25

Yes

125

6

MD

Postcesarean

—

10.17

Ref.

No. pts.

Dose

Analgesics Acetaminophen

117

11

SD

44

1

118

Meperidine

Morphine

Oxycodone

Lactation stage

M/P

M/S

t 1/2 (h)

t max (h)

BF

Comments

2.3

1–2

Yes

No adverse reaction in infants Acetaminophen and metabolite not detected in urine of infants (limit of detection 0.5 µg/mL) Case report of infant rash on two occasions No metabolites detected in milk (limit of detection 100 ng/mL)

Yes 1.24

10–60 mg q4–6h

0.76 2.16

No

2.7 2.5

2 1

No No

Yes 50–150 mg q3–4h

1.17

? Yes

5–30 mg q4h PO

5 mg q6h

2.45

3.4

?

?

M/P morphine ⫽ 2.46 ( ) ⫽ including active metabolite morphine ( ) ⫽ including active metabolite morphine ( ) ⫽ including active metabolite normeperidine Intravenous infusion Neurobehavioral depression observed in neonates during 3rd day of maternal therapy ( ) ⫽ including active metabolite normeperidine Epidural/intramuscular/ intravenous injection Intravenous infusion No adverse effects in neonates ( ) ⫽ including metabolite morphine-3-glucuronide No adverse reaction in neonate Neonatal plasma level 4 ng/mL Maternal dose unspecified; calculation based on 5 mg q6h Peak milk level 226 ng/mL

Taddio and Ito

Mean Finf

Drug

Codeine

186

Table 1

Anti-infectives Acyclovir

1

MD

4 months

0.77

1.17

Amoxicillin

127 128 129

1 1 6

MD MD SD

1 year 7 months 3 days

0.94 1.22 0.08

1.28 1.45 0.23

Ampicillin

130

10

MD

?

—

0.75

Cephalexin

131 132 129

2–3 6 6

SD MD SD

5–7 days 1–8 days 3 days

0.05 0.19 0.06

0.072 1.17 0.15

Cefadroxil

131 129

2–3 6

SD SD

5–7 days 3 days

0.13 0.16

0.29 0.43

Chloramphenicol

131 133

2–3 5

SD MD

5–7 days First week

— 1.88

0.18 5.23

as above 5 134 4 as above 5 131 2–3 135 1

MD SD MD SD SD

Puereperium 4 days 1 week 5–7 days 17 days

1.94 0.56 0.71 1.38 —

3.63 1.49 2.28 1.51 1.08

Ciprofloxacin

Clindamycin

136

1

MD

3 months

1.76

—

137 138

10 5

MD MD

? 1–2 weeks

0.91 —

2.27 6.2

139 131 140

2 2–3 1

MD SD MD

? 5–7 days 1–5 days

0.93 0.81 —

1.35 1.44 —

200 mg tid– 200 mg 5x/day

Yes

0.6–4.1

3.24

2.8

0.013–0.043

250–500 mg q8h 250–500 mg q6h

4–5

? Yes ?

0.23

Yes

250–1000 mg q6h

0.008–0.14

4–5

? 5 of 6 ?

500–1000 mg bid

0.009–0.019

6–7

? ? ? ?

50 mg/kg/ day divided q6h

0.6 0.47–0.59

1.8

1.4

250–750 mg q12h

1.6–2.14

2

No adverse reactions in infant

No difference in monitored parameters in infants compared with control group Administered as pivampicillin

Chemical method of analysis used measures antimicrobially effective fraction along with ineffective metabolites

? ? ? ? No Yes

150–450 mg q6h

One infant urine sample contained 1.08 µg/mL

Drugs and Breast-Feeding

126

One milk level reported Infant serum level below detection limit (0.03 µg/mL)

? No ? ? Yes

187

Intravenous therapy No milk levels obtained Report of neonatal bloody stools, but neonate previously treated with gentamicin and ampicillin

Table 1

Continued Dose

Cloxacillin

131

2–3

SD

5–7 days

0.072

Erythromycin Fluconazole

131 141

2–3 1

SD MD

5–7 days 8 days, 20 days

0.4 —

0.504 16.8

142

1

SD

12 weeks

12.5/day a

17.6/day a

Isoniazid

143

1

SD

?

Ketoconazole Metronidazole

144 145

1 12

MD MD

146

11

as above

147

Lactation stage

Max. Finf

Standard adult dose

0.144

250–500 mg q6h 250 mg qid 100–200 mg daily

10.0 (12.2)

5–10 mg/kg/ day

1 month 1 week

0.4 9.0 (13.2)

1.4 11.64 (15.9)

20 mg daily 250 mg bid– 750 mg tid

MD

First month

8.55 (11.7)

4

MD

First month

10.8 (13.4)

13.5 (18.2)

3

SD

6–14 weeks

2.78

5.04

M/S

t 1/2 (h)

t max (h)

BF ?

0.9 0.75

—

18.3 (24)

M/P

26.9

? ?

30

No

5.9

3

No

0.91

Yes Yes

1

Yes

1

Yes

2–4

No

Comments Intramuscular injection

a

Dose calculation based on infant consumption on a daily basis due to fluconazole’s long half-life in milk Peak acetylisoniazid milk level 5 h postdose; milk half-life 13.5 h ( ) ⫽ including metabolite acetylisoniazid Neonatal stool changes and Candida isolated in more cases than controls; no serious adverse reactions in neonates Neonatal plasma levels 1.27–2.41 µg/mL; metabolite levels 1.1– 2.41 µg/mL (in 7 neonates) M/P metabolite ⫽ 0.77 ( ) ⫽ including active hydroxymetabolite No adverse reactions in neonates Neonatal plasma metronidazole level 0.3–1.4 µg/mL; metabolite level 0.1–0.8 µg/mL M/P hydroxymetabolite ⫽ 1.2 ( ) ⫽ including active hydroxymetabolite No adverse reactions in neonates Neonatal plasma metronidazole level 0.6–4.9 µg/mL; metabolite level 0.4–2.3 µg/mL M/P hydroxymetabolite ⫽ 1.2

Taddio and Ito

No. pts.

188

Mean Finf

Ref.

Drug

148 149

13 1

SD MD

3–8 days 2 weeks

150

9

MD

151

?

Ofloxacin

152 as above 137

Penicillin

0.04 —

0.12 —

1000 mg qid

?

—

1.13

50–100 mg qid

MD

1–4 days

—

—

3 3 10

MD MD MD

2–5 days 2–5 days ?

— — 1.46

4.2 6.54 2.71

153

7

SD

2–5 days

—

0.054

Penicillin V

154

18

SD

?

0.04

0.21

Sulfamethoxazole

155

50

MD

First 10 days

2.0–2.5

—

Sulfisoxazole

156

6

MD

?

—

—

Tetracycline

157

5

MD

?

0.51

1.16

Trimethoprim

131 155

2–3 50

SD MD

5–7 days First 10 days

0.88 3.75–5.51

1.44 —

Vancomycin

158

1

MD

Puerperium

5.7

—

Nitrofurantoin

0.061

No Yes

0.27–0.31

No No

200–400 mg q12h 1–10 million units per day divided q4– 6h 125–500 mg q6–8h

400–1000 mg q12h 1000 mg qid

2.2 2.3 0.98–1.66

1–8

0.1 7.2

2–8

Intramuscular injection

14 infants 0.52 mg/L detected in urine of one infant; undetectable in another infant tested 3 infants with looser stools Yes 1 of 6

Yes

1.25

None detected in 20 milk samples (detection limit 2 µg/mL)

No No ? ?

250–500 mg q6h

A mean of 0.45% of the 24-h maternal dose was recovered in milk over a 48-h period (including metabolite) Detected in urine of the breastfeeding infant M/S N-acetyl sulfisoxazole ⫽ 0.22; milk half-life 8.9 h No adverse reactions in infants Infants’ serum ⬍0.07 mg/L

? Yes ?

Intravenous therapy Single milk level obtained 4 h postdose

189

80–160 mg q12h 125–500 mg q6–8h PO

2

0.06

Case report of hemolytic anemia in neonate negative for G6PD deficiency

Drugs and Breast-Feeding

Nalidixic acid

190

Table 1

Continued Max. Finf

2–30 days

3.0 (4.3)

3.32 (5.27)

MD

3–28 days

2.44 (2.97)

4.5 (5.0)

19

MD

2–35 days

2.72 (4.35)

3.83 (6.4)

162

1

MD

5 weeks

2.07

—

163

3

MD

2 days–5 weeks

3.78 (5.48)

4.66 (7.41)

164

?

MD

3–32 days

—

—

0.39

?

165

?

MD

?

—

—

0.41

?

166

1

MD

3 weeks

—

—

Ref.

No. pts.

Dose

Anticonvulsants Carbamazepine

159

1

MD

160

4

161

Lactation stage

Standard adult dose

M/P

M/S

10–20 mg/ kg/day

0.39

0.36

t 1/2 (h)

t max (h)

BF

Comments

Yes

No adverse reaction in infant Infant serum level 1.1–1.8 mg/L Mother also taking phenytoin ( ) ⫽ including active metabolite 10,11-epoxide Steady-state serum level in neonates 1.0 µg/mL Neonatal sedation and hyperexcitability noted M/S 10,11-epoxide ⫽ 0.49 ( ) ⫽ including 10,11-epoxide No abnormalities noted in breast-fed infants Neonatal serum levels ⬍1.5 µg/mL in 3 tested neonates M/P 10,11-epoxide ⫽ 0.53 ( ) ⫽ including 10,11-epoxide One milk level obtained No adverse reaction in infant Mother also taking primidone No adverse reaction in infants Infant plasma level 0–1.8 µg/mL M/P 10,11-epoxide ⫽ 1.05 ( ) ⫽ including 10,11-epoxide Mean milk level 1.9 µg/mL (0.8–3.8 µg/mL) Mean milk level 1.8 µg/mL (0.5–3.8 µg/mL) Neonatal cholestatic hepatitis after exposure throughout pregnancy and lactation

Yes

?

Yes

0.6

Yes

Yes

Taddio and Ito

Mean Finf

Drug

Phenobarbital

Phenytoin

167

1

MD

3–5 days

164

?

MD

3–32 days

168

1

MD

3 days–5 months

68.6

99

169

5

MD

3–28 days

51.2

81.3

165

?

MD

?

43.2

164

?

MD

3–32 days

170

6

MD

1–3 months

164

?

MD

165

?

171

172

48.9

53.8

—

—

297

1000–1500 mg/day

0.94

No

0.79

?

0.8

30–100 mg/ day

Yes

0.86

Yes

0.36

?

0.46

?

—

—

4.8

8.8 (9.84)

3–32 days

—

—

0.18

?

MD

?

—

—

0.19

?

1

MD

1 week

0.58

0.68

2

MD

1–33 days

4.8

7.8

100 mg tid

Yes

0.13

0.45

Yes

Yes

Mean milk level 21.3 µg/mL (18–24 µg/mL) No adverse reaction in infant Peak infant plasma level 29.5 mg/L 4 of 6 breast-feeding neonates hyperexcitable or sedated; 5 of these mothers taking additional antiepileptics Neonatal serum levels 15–40 µg/mL in 5 breast-fed neonates Maternal dose unspecified; calculation based on 100 mg/ day Mean milk level 4.8 µg/mL (0–33 µg/mL) Mean milk level 10.4 µg/mL (0.5–33 µg/mL) No adverse reactions in infants 2 of 6 infants with measurable levels of 0.12 and 0.18 µg/mL ( ) ⫽ including conjugated and unconjugated 4-OH phenytoin Mean milk level 0.8 µg/mL (0.5–1.4 µg/mL) Mean milk level 0.7 µg/mL (0–2.2 µg/mL) 5 infants breastfed nursing did not affect the elimination of phenytoin from exposure in utero No adverse reactions in infants

Drugs and Breast-Feeding

Ethosuximide

191

Continued

Drug Primidone

Valproic acid

Max. Finf

3–32 days

—

—

MD

?

—

?

MD

First month

—

174

4

MD

4–27 days

175

1

MD

176 177 178

11 1 6

MD MD MD

No. pts.

Dose

164

?

MD

165

?

173

Standard adult dose

M/P

t 1/2 (h)

t max (h)

BF

Comments

0.81

?

7.47–9.96

0.71

?

—

0.72

Yes

8.49 (18.3)

21.2 (37.8)

0.72

Yes

5–29 days

2.88

4.06

Mean milk level 2.3 µg/mL (0.5–6.7 µg/mL) Maternal dose unspecified; calculation based on standard adult dose Mean milk level 2.1 µg/mL (0–8.3 µg/mL) Withdrawal reaction in 1 of 6 infants breast-fed Peak neonatal serum level in 2 neonates tested; primidone, phenobarbital, PEMA levels 2.5, 13, and 1.4 µg/mL, respectively M/S metabolite phenobarbital ⫽ 0.36 M/S metabolite PEMA ⫽ 0.64 2 of 5 breast-fed neonates with poor feeding for first 5 days of life Serum levels of 1 neonate; primidone, phenobarbital, PEMA 0.8–1, 1.5–3, and 0.5–0.6 µg/mL, respectively M/S phenobarbital ⫽ 0.41 M/S PEMA ⫽ 0.76 ( ) ⫽ including metabolites phenobarbital and PEMA No adverse reaction in infant Undetected in infant serum at 1 month

3–6 days 62–130 h 3–82 days

1.22 0.58 1.12 (1.25)

3.8 0.85 6.98 (7.63)

Lactation stage

250 mg tid– qid

M/S

Yes

15–60 mg/ kg/day divided tid 0.05 0.01–0.02 0.027

? ? Yes

M/S 3-keto metabolite ⫽ 0.074 ( ) ⫽ including 3-keto metabolite; other metabolites not detected (⬍0.1 µg/mL)

Taddio and Ito

Mean Finf

Ref.

192

Table 1

1

MD

3 months

180

1

MD

4–6 months

1.06 (1.89)

1.2 (2.1)

181

1

MD

6–8 weeks

1.29 (1.79)

1.36 (1.89)

Clomipramine

182

1

MD

4–35 days

2.4

3.75

Desipramine

183

1

MD

10 weeks

0.84 (1.74)

1.38 (3.0)

Doxepin

184

1

MD

1–4 months

0.38 (1.09)

—

185

1

MD

8 weeks

0.19 (0.29)

0.31 (0.43)

Antidepressants Amitriptyline

50–150 mg/ day

Yes

Infant thrombocytopenia purpura, anemia and reticulocytosis after exposure throughout pregnancy and lactation; infant serum valproic acid 6.6 µg/mL

Yes

No adverse reaction in infant Infant serum amitriptyline ⬍5 ng/mL; nortriptyline ⬍15 ng/mL ( ) ⫽ including active metabolite nortriptyline Infant plasma levels below limit of detection 10 ng/mL ( ) ⫽ including active metabolite nortriptyline No adverse reaction in infant

Yes

75–150 mg/ day 75–150 mg/ day

0.76–1.62

75–150 mg/ day

1.08–1.66

Yes Yes

Yes

Yes

193

No adverse reaction in infant Infant plasma desipramine ⬍1 ng/mL; hydroxydesipramine ⬍5 ng/mL ( ) ⫽ including metabolite 2-hydroxydesipramine No adverse reaction in infant Infant plasma doxepin ⬍5 µg/L; desmethyldoxepin 15 µg/L M/P desmethyldoxepin ⫽ 1.02–1.53 ( ) ⫽ including metabolite Ndesmethyldoxepin Case report of infant sleepiness Infant serum doxepin 3 µg/L; desmethyldoxepin 58 and 66 µg/L on 2 occasions ( ) ⫽ including metabolite Ndesmethyldoxepin

Drugs and Breast-Feeding

179

194

Table 1

Continued Max. Finf

20–747 days

3.1 (6.5)

6.77 (17.18)

MD

5 months

1.3 (3.17)

—

Yes

1

MD

17 weeks

1.89 (3.35)

3.02 (5.36)

Yes

189

1

MD

14 weeks

0.65

—

190

1

MD

6 weeks

0.07 (0.19)

0.13 (0.29)

191

2

MD

1 month

—

1.64 (4.97)

192

1

MD

6–7 days

1.3

2.9

Sertraline

193 as above

1 1

MD MD

3 weeks 3 weeks

— 0.24

— 0.39

Trazodone

194

6

SD

3–8 months

0.27

0.43

Zopiclone

195 196

3 12

SD SD

? 2–6 days

0.5 0.26

1.27 1.37

No. pts.

Dose

186

10

MD

187

1

188

Fluvoxamin Imipramine

Fluoxetine

Nortriptyline

Lactation stage

Standard adult dose 20–40 mg/ day

M/P

M/S

t 1/2 (h)

0.88

t max (h)

BF

Comments

3

Yes

No adverse reactions in 11 infants ( ) includes active metabolite norfluoxetine Plasma fluoxetine and norfluoxetine ⬍1 ng/mL in 1 infant tested Urine fluoxetine 2–17 ng/mL in 4 of 5 infants tested Urine fluoxetine 11–13 ng/mL in 2 of 5 infants tested One milk level reported ( ) ⫽ including active metabolite norfluoxetine No adverse reaction in infant ( ) ⫽ including active metabolite norfluoxetine No adverse reaction in infant One milk level reported No adverse reaction in infant ( ) ⫽ including active metabolite desipramine M/P desipramine ⫽ 0.91 ( ) ⫽ including active metabolite desipramine No adverse reaction in infant; normal motor development at 4 months Infant serum ⬍10 ng/mL No adverse reactions in infant Infant serum ⬍0.5 ng/mL

Yes

100–200 mg/ day 75–150 mg/ day

30–100 mg/ day

50–200 mg/ day 150–400 mg/ day 7.5 mg prn

Yes

0.91

Yes

0.87–3.17

Yes

0.142 0.6 0.51

5.3

5–9

Yes Yes

2

No

2.4

? No

Taddio and Ito

Mean Finf

Ref.

Drug

6

SD

1–12 months

—

0.18 (0.34)

10 mg/day

1.2

198, 199

4

MD

5–12 months

(0.16)

(0.45)

60 mg q12h

(0.21)

200

3

SD

14 weeks–18 months

0.5

0.84

2.5 mg q4– 6h

0.5–0.56

?

Antihypertensives Acebutolol

201

3

MD

3–9 days

1.55 (5.31)

3.09 (8.07)

400–800 mg/ day

1.9–9.2

?

Atenolol

202

1

MD

1–6 weeks

19

25

50–200 mg/ day

2.9

203

7

MD

Puerperium

9.44

15.43

204 205

4 11

MD MD

4–12 days ?

9.8 6.39

25 —

206 45

5 1

MD MD

Puerperium 1 week

5.67 4.22

9.36 —

Captopril

207

11

MD

?

0.009

0.014

12.5–25 mg tid

Clonidine

208

9

MD

1–5 days

4.14

—

0.2–1.2 mg/ day divided qid

as above

9

MD

10–14 days

7.85

—

Yes

as above

9

MD

45–60 days

6.70

—

Yes

Terfenadine

Tripolidine

(14.2)

2

No

(4.3)

No

8

4.5 1.1–3.1

Yes

#? 2–6

Yes Yes

Yes Yes

3.8

No

Yes

Peak metabolite level 5.3 h postdose M/P metabolite ⫽ 0.8 ( ) ⫽ including active metabolite descarboethoxyloratadine ( ) ⫽ including carboxylic acid metabolite Terfenadine ⬍10 ng/mL

Hypotension, bradycardia in 1 neonate M/P diacetolol ⫽ 2.3–24.7 ( ) ⫽ including active metabolite diacetolol No adverse reaction in infant Infant plasma level, below detection, 10 ng/mL Serum level 0.07 mg/L in 1 neonate; no adverse reaction Infant plasma level ⬍3 µg/L No adverse reactions in infants Detected in urine of 3 tested infants No adverse reactions in infants One milk level reported Neonatal bradycardia, cyanosis Neonatal serum level 2010 ng/ mL on one occasion Several mothers nursed their infants during therapy with no adverse reports in infants M/B AUC ⫽ 0.031 Mean neonatal serum level 0.65 ng/mL Mean neonatal serum level 0.5 ng/mL Mean infant serum level 0.25 ng/mL

195

197

Drugs and Breast-Feeding

Antihistamines Loratadine

196

Table 1

Continued Max. Finf

Standard adult dose

0.69

0.83

3–45 days

—

—

30–60 mg q6–8h 2.5–40 mg/ day

SD

Puerperium

0.016 (0.03)

0.05 (0.07)

1

MD

8 weeks

—

0.8

213

25

MD

3 days

0.10

0.45

214

3

MD

6–9 days

0.30

0.59

215

1

MD

17 days

1.22

1.75

216

4

MD

30–60 hours

—

217

3

MD

1–8 weeks

—

0.09 (0.72) 1.19 (3.2)

218

8

MD

?

No. pts.

Dose

Diltiazem

209

1

MD

18 days

Enalapril

210

3

SD

211

5

Hydralazine

212

Labetolol

Methyldopa

Lactation stage

0.2 (0.51)

—

M/P

M/S

t 1/2 (h)

t max (h)

BF

Comments

No ?

0.005–0.043

No

Enalapril not measured; enalaprilat milk level ⬍0.2 ng/ mL Normal angiotensin-converting enzyme activity in milk M/S enalaprilat ⫽ 0.021–0.031 ( ) ⫽ including active metabolite enalaprilat

Yes

10–50 mg qid 200–600 mg bid

24 of 25 0.8–2.6

2–3

Yes

Yes

250 mg bid– 500 mg qid

? 0.19–0.34

Yes

0.46

Yes

No adverse reactions in neonates Neonatal plasma level above detection (21 µg/L) in 1 of 2 neonates tested No adverse reaction in neonate Neonatal serum level ⬍0.2 µg/mL ( ) ⫽ including conjugated (and free) methyldopa No adverse reactions in infants 1 of 3 infants with detectable plasma level of 0.09 µg/mL (detection limit 0.05 µg/mL) ( ) ⫽ including conjugated (and free) methyldopa Maternal dose unspecified; calculation based on 750 mg/ day ( ) ⫽ including conjugated (and free) methyldopa

Taddio and Ito

Mean Finf

Ref.

Drug

219

9

MD

?

220

8

MD

3–5 days

Minoxidil

203 204 221

3 3 1

MD MD MD

4–6 months 4–60 days 2 months

Nadolol

222

12

MD

Nifedipine

223

13

224

1.0

3.6

0.75–1.5

2.95–5.9

— 1.26 1.34 (1.54)

3.0 2.2 5.0 (5.41)

⬎1 month

4.1

5.0

MD

1–100 days

0.1

—

1

MD

?

—

0.53

225

1

MD

10 days

—

Nitrendipine

226

2

MD

⬎3 months

Oxprenolol

228

9

MD

3–6 days

0.1 (0.26) 0.45

1.04 (1.53) 0.25 (0.54) 1.5

Propranolol

229 230

12 3

MD MD

⬍4–⬎8 days First week

0.72 0.3 (1.0)

2.1 1.0 (2.0)

231 232 206

1 1 5

MD MD MD

2–3 months 3–6 days Puerperium

0.19 0.26 0.30

0.24 0.45 0.41

Sotalol

233

5

MD

First week

2.18

41.95

Timolol

234 235 229

1 1 11

MD MD MD

5, 105 days 5–7 days ⬍4–⬎8 days

— 21.8 1.1

23.6 28.1 3.3

50–100 mg bid

3.7

Yes

2.8

Yes 3.6

No Yes Yes

2–3.1 10–40 mg/ day 4.6

80–240 mg/ day 20 mg q12h

20 mg bid– 16 mg/day divided bid–tid

6

3.3

0.2–0.4

1

No

1

Yes

3–4

No

0.29 0.33–1.65

? ?

6.5

2.4–5.6 5–10 mg bid

Yes Yes Yes Yes

5.4

2.75–3.57 0.8

Sustained released nifedipine No adverse reactions in infants

( ) ⫽ including pyridine metabolite ( ) ⫽ including inactive pyridine metabolite

?

0.45

3

80–160 mg bid

Infant plasma level ⬍3 µg/L No adverse reaction in infant ( ) ⫽ including glucuronide metabolite

No Yes ⫽ 4

10–20 mg q8h

10–40 mg/ day 60–320 mg/ day divided tid

21.8

Peak neonatal plasma level 0.5–45 µg/L

Drugs and Breast-Feeding

Metoprolol

Yes No ?

M/P naphthoxylactic acid 0.19–0.42; milk half-life 4.2 h ( ) ⫽ including propranolol glucuronide and naphthoxylactic acid No adverse reaction in infant No adverse reaction in neonate No adverse reactions in neonates No adverse reactions in neonates No adverse reaction in infant

197

Continued

Drug Verapamil

Anti-inflammatory Acetylsalicylic acid

198

Table 1

Mean Finf

Max. Finf

3–5 days

0.09

0.14

MD

8 weeks

0.38 (0.51)

0.51 (0.68)

1 1

MD MD

13 days 3 months

0.55 0.10 (0.13)

0.91 0.30 (0.38)

120

1

SD

7 weeks

0.45

0.55

as above 240

1 1

SD MD

13 weeks 6 months

— 1.8–2.6

0.74 3.2–4.7

241

1

MD

9 weeks

—

—

Yes

242

1

MD

2 weeks

—

—

Yes

Ref.

No. pts.

Dose

236

1

MD

237

1

238 239

Lactation stage

Standard adult dose

M/P

M/S

t 1/2 (h)

t max (h)

0.23

80 mg tid– qid 0.64

Comments No adverse reaction in neonate Neonatal plasma level 2.1 ng/mL No adverse reaction in infant Infant plasma verapamil and norverapamil ⬍1 µg/L ( ) ⫽ including metabolite norverapamil

Yes

0.6

325–650 mg q4h antipyretic dose

BF Yes

4.29

2 1

No Yes

No

0.05

0.03–0.34 2.6–3.9 g/ day in divided doses–antiinflam. dose

Infant plasma verapamil and norverapamil ⬍1 µg/L Norverapamil peak milk level 2 h postdose; milk half-life 1.34 h; M/P ⫽ 0.16 ( ) ⫽ including metabolite norverapamil

0.04–0.08

3

Yes

Yes Maternal dose unspecified; calculation based on standard adult dose Peak milk level 1 mg/dL

Taddio and Ito

No milk levels obtained No adverse reaction in infant; infant serum level 0.47 mmol/L (infant 50% breastfed) No milk levels obtained Neonatal metabolite acidosis; serum salicylate level 24 mg/dL

243

6

SD

?

—

—

Flurbiprofen

244 245 246

12 10 1

MD SD MD

3–5 days ⬎1 month Mature milk

— — —

0.29 0.17 —

247

12

MD

3–5 days

—

—

248

16

MD

⬍10 days, 10 months

—

1.21

249

1

MD

4–6 days

—

—

Ketorolac

250

10

MD

2–6 days

—

0.18

Mefenamic acid

251

10

MD

2–4 days

0.2 (1.2)

0.79 (2.4)

250 mg q6h

Naproxen

252

1

MD

5 months

1.79

2.21

250 mg bid– qid

as above

1

MD

6 months

2.45

2.80

253 as above 254

1 1 4

MD MD MD

13 months 8 months 3—4.5 months

3.6 3.6 3.51

7.65 4.95 6.345

Ibuprofen

Indomethacin

Piroxicam

?

25–50 mg tid 50 mg q6h 0.019

3

400–800 mg q4–6h

No No Yes

? 25 mg bid– 200 mg/ day

Yes

0.37

Yes

10 mg qid

Yes

4

Mean infant level 0.08 µg/mL 1 hour after nursing ( ) ⫽ including metabolites

Yes Yes

0.01–0.03

Ibuprofen and major metabolites not detected in milk (limit of ibuprofen detection 0.5 µg/mL) No ibuprofen detected in milk (limit of detection 1 µg/mL) Rectal administration No adverse reactions in neonates; plasma levels in 6 of 7 neonates tested ⬍20 µg/L; level 47 µg/L in 1 neonate Neonatal seizures No milk or neonatal serum levels obtained

No

0.015– 0.037

20 mg/day

Intramuscular injection No drug detected in milk

Drugs and Breast-Feeding

Diclofenac

Yes No Yes

0.47 mg of naproxen and conjugated naproxen excreted by infant over 12 h after maternal dose of 375 mg (0.26% of cumulative maternal value) Not detected in infant serum No adverse reaction in infants Piroxicam and conjugate not detected in one infant’s urine

199

200

Table 1

Continued Max. Finf

—

0.44–2.94

6 weeks

0.15

—

MD

3–4 weeks

1.87

2.1

3 1 1

MD MD MD

? 1 month 5 days

3.72 0.12 16.6–33.3

9.60 0.17 —

261

4

MD

1–4 weeks

—

—

Yes

262

1

MD

0–10 weeks

—

27.02

Yes

263

1

MD

70 days

4.0

—

Yes

Ref.

No. pts.

Dose

Antipsychotics Chlorpromazine

255

4

MD

?

256

1

Md

257

1

258 259 260

Haloperidol

Perphenazine Lithium

Lactation stage

Standard adult dose

M/P

t 1/2 (h)

t max (h)

BF

Comments

10–50 mg tid–qid

2 of 4

Maternal dose unspecified; calculated based on standard adult dose Milk levels 7–98 ng/mL 1 of 2 breast-fed infants was drowsy and lethargic

0.5–2 mg bid–tid

No

M/S

Yes

2–8 mg tid 900 mg–1.2 g/day divided bid– qid (as carbonate)

0.7–1.1

? Yes Yes

No adverse reaction in infant; developmental milestones achieved at 6 months and 1 year 1.5 µg/L detected in infant’s urine No adverse reaction in infant Neonatal cyanosis, lethargy 0.6 mEq/L detected in infant plasma and in milk sample Infant exposed to lithium in utero Milk level range 0.16–0.56 mmol/L Infant serum level range 0.1– 0.3 mmol/L No adverse reaction in infant Infant serum level ⬍0.2 mmol/L throughout breastfeeding One milk level obtained No adverse reaction in infant; lithium level 0.04 mEq/L

Taddio and Ito

Mean Finf

Drug

Propylthiouracil

Anxiolytics Alprazolam

Clonazepam

Diazepam

264

1

MD

2–6 months

2.1

4.5

265 266 267

1 5 9

MD SD SD

? 2–6 weeks 1–8 months

7.17 — 0.32

11.52 1.34 —

268

1

SD

?

—

—

269

1

MD

7 days

—

—

270

8

SD

6–28 weeks

—

6.96/day b

271

1

MD

2–4 days

1.89

2.41

272

1

MD

3–14 days

—

—

273

1

MD

1 year

4.2 (7.36)

4.7 (8.6)

0.3–0.7

Yes

1.16 0.98

5–15 mg/day

1

50 mg bid– tid

? ? ?

No

Yes

0.25 mg bid– tid

0.36

8–10 mg/day divided tid 0.3

2 mg bid–10 mg qid

0.2

14.46

1.1

No

4

Yes Yes

Yes

Given as carbimazole No adverse effects in breastfeeding twins; methimazole plasma levels 0–156 ng/mL (peak level 2–4 h postdose) Given as carbimazole 1 infant breast-feeding while mother on long-term treatment showed no adverse effects 0.077% of a radioactive dose of propylthiouracil excreted in 500 mL of milk over 24 h

Drugs and Breast-Feeding

Antithyroid Methimazole

No levels obtained Neonatal restlessness and irritability; neonatal exposure in utero also b Dose calculation based on infant consumption on a daily basis due to alprazolam’s long milk half-life Alpha-hydroxy metabolite ⬍4 ng/mL Mother also taking phenytoin

201

Milk level range 11–13 ng/mL Infant neurodevelopmental exam normal at 5 months No adverse reaction in infant; low levels of metabolites detected in infant plasma M/P desmethyldiazepam ⫽ 0.13 M/P oxazepam ⫽ 0.10 M/P temazepam ⫽ 0.14 ( ) ⫽ including active metabolites n-desmethyldiazepam, oxazepam, temazepam

Continued Mean Finf

Max. Finf

2.76–4.6 (6.07– 10.12)

7.8–13.0 (14.8– 24.6)

No. pts.

Dose

Lactation stage

274

1

MD

1 week–3 months

275

4

MD

3–9 days

276

1

MD

8 days

—

—

Yes

277

3

MD

6 days

2.34 (3.9)

—

Yes

Lorazepam

278

1

MD

5 days

2.16 (6.1)

—

2–3 mg/day divided bid–qid

Yes

Midazolam

279

12

MD

2–7 days

—

—

15 mg prn

Yes

as above

2

SD

2–3 months

Nitrazepam

279

10

MD

2–7 days

0.06 (0.10) 2.52

0.11 (0.15) 3.60

Oxazepam

280

1

MD

⬎7 months

0.68

0.75

2.1 (5.4)

Standard adult dose

M/P

0.16

3.87 (11.5)

5–10 mg prn 10–15 mg tid–qid

M/S

t 1/2 (h)

BF

Comments

Yes

Normal infant development; infant sedation if feeding occurred within 8 h of maternal dose Infant diazepam level 0.7 ng/ mL and n-desmethyl-diazepam level 46 ng/mL at 32 days of age M/P n-desmethyldiazepam ⫽ 0.27; oxazepam not in milk ( ) ⫽ including n-desmethyldiazepam Neonatal lethargy and weight loss; urine positive for oxazepam No adverse reactions in infants Mean infant level on 2 occasions for diazepam 74 and 172 ng/mL; 31 and 243 ng/mL for n-desmethyl-diazepam Oxazepam not detected in any sample ( ) ⫽ including n-desmethyl-diazepam One milk level obtained No adverse reaction in neonate ( ) ⫽ including conjugated and free lorazepam Morning milk level after evening dose ⬍3 µg/L ( ) ⫽ including hydroxymidazolam Plasma level ⬍3 µg/L in 1 infant that was tested

Yes

0.15

?

0.27

Yes ?

Taddio and Ito

Ref.

Drug

t max (h)

202

Table 1

10

MD

⬍15 days

—

1.31

Cardiovascular Amiodarone

282

1

MD

6–9 weeks

—

36.9 (51.5)

283

1

MD

2–3 days

—

284

1

MD

First month

5.67 (7.7)

8.1 (11.7) 8.21 (11.0)

285

3

MD

0–6 weeks

10.35 (15.3)

13.7 (21.8)

Bretylium

286

1

MD

First 4 months

Digoxin

287

2

MD

288 289

11 1

290

11

—

—

14 days

2.14

3.5

MD MD

3–7 days 7 days

2.3 2.28

— —

SD

?

—

—

30 mg HS

⬍0.2

200–600 mg/ day

Yes

No adverse reactions in neonates; plasma levels of temazepam and oxazepam below detection in 2 neonates that were tested Milk oxazepam levels ⬍5 µg/L for all patients

Yes

Infant amiodarone and desethylamiodarone plasma level at 9 weeks 0.4 mg/L and 0.15 mg/L, respectively ( ) ⫽ including metabolite desethylamiodarone ( ) ⫽ including metabolite desethylamiodarone No adverse reaction in infant Infant serum amiodarone level ⬍0.1 mg/L; desethylamiodarone ⬍0.05 mg/L ( ) ⫽ including metabolite desethylamiodarone One infant stopped breast-feeding due to hypothyroidism Desethylamiodarone M/P ⫽ 1.9 ( ) ⫽ including metabolite desethylamiodarone No milk levels obtained No adverse events in infant

No Yes

7.8

100 mg tid– 600 mg q6h 0.125–0.25 mg/day

Yes

Yes

4–6

0.8–0.9

0.59 0.9

Yes

No Yes

0.62

No

No adverse reaction in neonates; neonatal plasma levels ⬍0.1 ng/mL One milk level obtained Neonatal serum level 0.2 ng/mL Intravenous injection

203

281

Drugs and Breast-Feeding

Temazepam

204

Table 1

Continued Max. Finf

Standard adult dose

5.85 (10.7)

8.55 (15.8)

100–200 mg qid

2–16 days

2.75

3.05

MD

2 months

1.75

2.21

1

MD

5–7 days

3.70

4.92

294 295

11 1

MD MD

1–6 days 2 days, 6 weeks

3.7 1.1

7.0 1.2

Procainamide

296 297

1 1

MD MD

2–5 days 2 days

1.04 2.43 (4.0)

1.58 4.59 (6.84)

Propafenone

298

1

MD

3 days

Quinidine

299

1

MD

4–5 days

0.03 (0.08) 5.58

6.27

300–600 mg q8–12h (as sulfate)

Diuretics Acetazolamide

300

1

MD

10–11 days

1.53

1.89

250 mg daily–qid

Yes

Chlorothiazide

301

11

SD

⬎3 months

—

500–2000 mg/day

No

No. pts.

Dose

291

1

MD

4 days–1 month

292

1

MD

293

1

Flecainide

235

Mexiletine

Disopyramide

Lactation stage

⬍0.36

M/P

M/S

t 1/2 (h)

t max (h)

0.9

0.4

100–150 mg q12h

Comments

Yes

No adverse reaction in infant; infant plasma level 0.5 mg/L at 28 days M/P n-monodesalkyl metabolite ⫽ 5.6 ( ) ⫽ including n-monodesalkyl metabolite No adverse reaction in neonate; neonatal plasma level ⬍0.5 mg/L Infant disopyramide level 0.1– 0.14 mg/L

Yes

0.46–0.53

Yes

1.57–2.18

No

2.6–3.7

14.7 1.14–2

200–300 mg q8h

50 mg/kg/ day divided q3h– q6h 150 mg q8h

BF

1.45 4.3

3–6

No Yes

Yes Yes

No 0.71

No

No adverse reaction in infant; infant serum level ⬍0.05 mg/L M/P NAPA ⫽ 3.8 ( ) ⫽ including active metabolite NAPA ( ) ⫽ including metabolite 5OH-propafenone Administered as quinidine sulfate

No adverse reaction in neonate; neonatal plasma level 0.2– 0.6 µg/mL Milk levels ⬍ mg/L; calculation assumes milk level of 1 mg/L

Taddio and Ito

Mean Finf

Ref.

Drug

302

7

MD

3 days

Furosemide

303

120

MD

First week

Hydrochlorothiazide

304

1

MD

Spironolactone

305

1

Gastro-intestinal Cimetidine

306

6.66

15.5

?

25–100 mg/ day 20–40 mg bid

—

—

28 days

1.94

2.92

50–100 mg/ day

Yes

MD

17 days

0.84

1.15

25–200 mg/ day

Yes

1

MD

6 months

4.42

4.77

800–1200 mg/day

No

as above 307 308

1 12 10

SD SD MD

6 months 6–45 weeks 1–5 days

0.92 — 0.08

1.67 5.66 0.26

309

2

MD

3–6 days

0.08

—

Famotidine

310 as above 311

10 4 8

SD MD SD

2–8 days 1–3 months ?

0.006 0.07 —

0.01 — 0.32

Loperamide

312

6

MD

First 3 days

0.024

Metoclopramide

313

5

MD

1–3 weeks

3.3

0.07 (0.09) 4.7

18 10 3

MD SD MD

9–14 weeks 7–10 days 3–8 months

1.44 2.26 1.47

3.75

Nizatidine

as above 314 315

Ranitidine

316

6

SD

6–10 days

1.4

4.4

317

1

MD

54 days

4.86

8.6

Cisapride Domperidone

3.6

No

5.77 5–10 mg tid–qid 10 mg tid– qid

20–40 mg/ day 2 mg prn

2.5

3.3

0.045

Furosemide and fluid restriction successfully suppressed lactation No adverse reaction in neonate; neonatal blood level ⬍20 ng/mL No adverse reaction in neonate Canrenone levels measured; spironolactone levels not measured M/S canrenone ⫽ 0.51–0.72

No No No No No ? No

0.41–1.78

No Yes

5–10 mg tid–qid

150 mg/day bid 150 mg/day– bid

M/B ⫽ 0.06

Drugs and Breast-Feeding

Chlorthalidone

Yes ? ?

⬍2

?

1.92–6.7 5.5

No

205

6.8–23.8

( ) ⫽ including loperamide and loperamide oxide Metoclopramide detected in plasma from 1 of 5 neonates tested; peak level 20.9 ng/mL

206

Table 1

Continued Mean Finf

Max. Finf

Standard adult dose

—

—

800 mg–4000 mg/day

?

0.07 (7.5)

—

7, 11 days

0.03 (3.59) —

(4.27) —

Drug

Ref.

No. pts.

Dose

Miscellaneous 5-aminosalicylic acid

318

1

MD

6 weeks

319

1

MD

320

2

MD

Chloroquine

Lactation stage

3

SD

2–5 days

322

11

SD

?

6.6/day c

11.25/day c

Hydroxychloroquine

323 324

5 1

SD MD

2–2.5 months 9 months

4.71/day 2.81

11.91/day 3.67

Methylergometrine

325 326

1 8

MD MD

8 weeks 5 days

— 2.4

⬍0.05 3.38

M/S

t 1/2 (h)

t max (h)

Yes

Rectal administration Milk levels not reported Case report of watery diarrhea in infant One milk level reported M/P acety-5-ASA ⫽ 5.1 ( ) ⫽ including metabolite acetyl 5-ASA ( ) ⫽ including acetyl-5-ASA

Yes

0.13

Yes

6.6

0.2–0.4 mg q6-12h

Comments

0.27

No

500 mg/week 1.96–4.26 for malaria (as phosphate)

200–600 mg/ day arthritis (as sulfate)

BF

5.5

8.8 days

14.4

Yes

5.5 days

3

No Yes

No ?

Mean amount excreted in milk over 9 days was 0.19% of maternal dose M/P metabolite desethylchloroquine ⫽ 0.54–3.89 c Dose calculation based on infant consumption on a daily basis due to chloroquine’s long milk half-life 4 infants of 4 tested had positive urine samples for chloroquine and desethylchloroquine M/P desethylchloroquine ⫽ 1.5

Taddio and Ito

321

M/P

327

6

MD

3–152 days

328

1

SD

329

7

Pseudoephedrine

200

Sulphasalazine

Sumatriptan

—

4.51

?

0.51

—

Yes

SD

?

—

—

No

3

SD

14 weeks–18 months

1.6

2.6

330

12

MD

5–6 weeks

—

0.9 (11.1)

331

1

MD

4.5–6.5 months

332

8

MD

2–24 weeks

—

333

1

MD

2 months

—

—

334

5

SD

10.8–28.4 weeks

—

4.2

(2.25)

5–60 mg/day (prednisone)

60 mg qid prn 2.4 g/day divided qid

1

0.1–0.2

2.2–2.8

4.2–7

1–1.5

?

? Yes

(5.85)

Yes

1.42 (11.7)

Yes

Yes

4.9

2.2

2.6

No

Sulfasalazine milk level ⬍1 µg/mL in 26 of 31 samples M/S sulfapyridine ⫽ 0.4 ( ) ⫽ including metabolite sulfapyridine No sulfasalazine, 5-ASA, acetyl-5-ASA detected in milk No adverse reaction in infant; infant sulfapyridine and metabolites urine levels 3–4.1 µg/mL M/P sulfapyridine ⫽ 0.6–0.63 ( ) ⫽ including sulfapyridine and metabolites Sulfasalazine undetected in 6 of 7 milk samples and 6 of 8 infants (⬍1 mg/L); peak infant plasma sulfapyridine level 4.8 mg/L M/S sulfapyridine ⫽ 0.48 ( ) ⫽ including sulfasalazine and sulfapyridine Bloody diarrhea in infant; infant blood sulfapyridine level 5.3 µ/mL Mother found to be a poor acetylator Subcutaneous injection

207

100 mg PO (6 mg SQ ⫽ subcutaneous)

One milk level reported Prednisolone and prednisone measured Mean total recovery per 1 L of milk during 48 h from radiolabeled prednisolone was 0.14%

Drugs and Breast-Feeding

Prednisolone

Table 1

Continued

Theophylline

Warfarin

Mean Finf

Max. Finf

Standard adult dose

?

—

8.14–13.48

400–900 mg/ day PO divided bid

SD

9 months

—

2.82

13

MD

3–10 days

—

—

2

MD

?

—

—

Ref.

No. pts.

335

3

SD

336

1

337

338

Dose

Lactation stage

M/P

M/S

t 1/2 (h)

t max (h)

0.67

0.7

2–10 mg/day

1–3

BF

Comments

No

Intravenous injection

Yes

Agitation in infant; 5 other nursing infants with no adverse effects No adverse reactions in neonates Milk and neonatal plasma levels below limit of detection 25 ng/mL No adverse reactions in infants No warfarin detected in milk

7—Yes

Yes

208

Drug

Note: This table reviews the findings of clinical trials of drugs taken during lactation. It is inevitable that some of the details sought by the reader will be missing. Readers are encouraged to review the individual research papers cited and other resources for additional information. Study design: All drugs were taken orally unless otherwise specified. Where data were unavailable, a question mark (?) appears. Ref ⫽ reference. No. ⫽ number of subjects (taken from the number of women donating milk samples for drug analysis or from the number of women whose data appear in the paper). MD ⫽ multiple dose, SD ⫽ single dose. Lactation stage: The period of lactation is specified. Whether the infant was nursing or not during maternal drug therapy is also noted (BF ⫽ breast-fed infants). The mothers donating milk samples for study analysis were not always the same mothers who nursed their infants during drug therapy. Dosage to infant: The percent maternal dose available to the infant (Finf ) was calculated as a percentage of the maternal dose in milligrams per kilogram. The following formula was used: Finf(%) ⫽

C m ⫻ Vm ⫻ 100 D

D inf (mg/kg/day) ⫽ Finf ⫻ standard adult daily dose/60 kg Other findings: M/P ⫽ milk-to-plasma ratio; M/S ⫽ milk-to-serum ratio, M/B ⫽ milk-to-blood ratio. These ratios have been derived from investigators in different ways (e.g., arithmetic mean, area under the curve). The t 1/2 ⫽ half-life in milk, t max ⫽ time to maximum drug concentration in milk. Comments: Other findings of the study are noted.

Taddio and Ito

where C m ⫽ milk concentration of drug, Vm ⫽ volume of milk ingested by infant (150 mL/kg/day for MD studies and 30 mL/kg/dose for SD studies) (41). It is assumed that an infant nurses five times daily during MD therapy and once during a SD study. Maternal dose: D is the maternal dose (mg/kg). A maternal weight of 60 kg was used to calculate infant exposure if maternal weight was not provided by the paper. The bioavailability of the drug was assumed to be 100%. Mean Finf ⫽ mean percentage of maternal dose obtained through milk. This value was calculated using the average milk concentration reported, or the arithmetic average calculated from data shown; arithmetic means were calculated using milk levels reported in the first 8–16 hours after dosing if appropriate. Max Finf ⫽ maximum percentage of maternal dose obtained through milk. This value was calculated using the highest milk level observed in one subject or the average maximum concentration achieved. Standard adult dose: The standard adult dose, taken from standard reference books (339, 340), is provided so that an infant’s dose (Dinf) may be estimated using the relationship:

Drugs and Breast-Feeding

209

THERAPEUTIC IMPLICATIONS OF DRUG EXCRETION IN BREAST MILK Assessing the Risk of Drug Exposure Through Breast Milk The dose of a drug ingested by an infant through breast milk provides an estimate of the risk to the infant. For most drugs that have been studied, the infant dose ingested through breastmilk is less than 5% of the maternal dose on a weight-adjusted basis (Finf ). Because the therapeutic dose is similar in both adults and children on a milligram-per-kilogram basis, the dose ingested by the infant will not be sufficient to exert a pharmacological effect. Occasionally, however, the Finf may approach the recognized infant therapeutic dose. In such cases, the clinician assumes that the risk for side effects is not negligible and may be similar to the risk of side effects for a drug given to the infant directly. For the convenience of risk assessment, drugs can be ranked and categorized according to the proportion of the maternal dose (41). However, care should be exercised when these data are used in clinical settings. First, the risk to the infant is not entirely proportional to the Finf . Even if the infant dose should be a small fraction of a maternal dose, potent pharmacologic actions or potential for severe side effects may make the drug incompatible with breast-feeding. In contrast, a drug that causes no known adverse effects may be compatible with breast-feeding even if the infant ingests close to a therapeutic dose via milk. Assessing potential risks is a crucial step in counseling a mother about whether to breast-feed or formula-feed while taking medication. The potential risks of developing adverse reactions in the breast-fed infants exposed to drugs in milk should be weighed against the risks associated with formula-feeding if the mother does not breast-feed. Bottlefeeding is associated with higher mortality rates for infants than breast-feeding. In industrialized countries, bottle-feeding is associated with mortality rates of 1 to 5 per 1000 live births. In developing countries, the mortality rate is 300 per 1000 live births (46). Furthermore, recent studies have shown associations between premature infants’ exposure to cow’s milk (which most infant formulas are based on) to immune system disorders such as insulin-dependent diabetes mellitus (47) and Crohn’s disease (48). While there appears to be a real risk associated with formula based on cow’s milk, virtually nothing is known about the probability of adverse reactions developing in infants of breast-feeding mothers on medications. When one assesses the risk associated with breast-feeding during maternal medication, any case report of adverse reactions in breast-fed infants should be considered. One should also bear in mind that the incidence of such events may well be extremely low. General guidelines have been proposed by Berlin (6) based on consistent findings from studies that have evaluated drug excretion in breast milk: 1. Most drugs (and environmental chemicals) are capable of crossing from plasma to milk. 2. The concentrations of drugs in milk parallel the levels achieved in plasma over a similar time course, and milk-to-plasma concentration ratios vary from 50– 100%. 3. Following a single maternal dose of a drug, the observed half-lives are similar for both plasma and milk. 4. The total amount of drug available for infant absorption is usually less than 1% of the maternal dose.

210

Taddio and Ito

5. 6.

There are few studies or reports based on steady state-maternal dosing. For most drugs, the risk of side effects in the infant can be assessed to be small or negligible.

In the Motherisk Program in Toronto, we have been prospectively collecting infant outcome data in a cohort of breast-fed infants whose mothers were on medications. Mothers initially contacted Motherisk seeking advice on the safety of drug consumption during breast-feeding. They had not yet started drug therapy at this time. Mothers were subsequently contacted for a follow-up interview (about 6 weeks after the initial consultation). The mothers who ingested drugs while breast-feeding were asked if their children experienced any adverse reactions during their drug therapy. This study included 838 motherinfant pairs and took place between 1985 and 1991. No major adverse events (defined as requiring medical attention) were identified. Overall, 11% of the mothers reported symptoms in the infants during their drug therapy. These symptoms did not require medical attention. Table 2 shows a partial list of the minor adverse effects on the breast-fed infants exposed to drugs commonly inquired about in our program. Long-term effects due to infants’ drug consumption in breast milk are not available. Studies of this nature are difficult to conduct. Large cohorts are necessary and patients must be followed for long periods of time.

REVIEW OF SELECTED DRUG CLASSES This section offers recommendations for the use of selected drugs (mostly taken from Table 1) and breast-feeding. Infant monitoring guidelines are also recommended.

Analgesics Acetaminophen is excreted into milk in small amounts. Narcotic analgesics such as codeine and morphine have also been detected in milk in small amounts. Therapeutic doses of codeine are generally considered safe to use. One recent case report found morphine plasma concentrations in the analgesic range in a 3-week-old infant whose mother was being tapered off high doses of morphine (see Table 1). No adverse effects were reported in this infant. Nursing infants, however, may be sensitive to the effects of narcotics and should be observed for signs of sedation or unexplained episodes of cyanosis and apnea during maternal therapy (49).

Anti-infectives Penicillin and cephalosporin antibiotics are considered safe to use while breast-feeding. Infants should be observed for allergic reactions. Infants with glucose-6-phosphate dehydrogenase (G6PD) deficiency should not be exposed to sulfonamides, nalidixic acid, nitrofurantoin, or other agents capable of causing hemolysis. Sulfonamides should be avoided if newborn infants have jaundice because of a theoretical risk of displacement of bilirubin from binding sites and kernicterus. The use of metronidazole during lactation is controversial. Experimental studies have revealed that it has carcinogenic and mutagenic properties (50). However, human

Drugs and Breast-Feeding

211

Table 2 Main Maternal Drug Exposures and Associated Minor Adverse Reactions in the Breast-Fed Infants (Data from Motherisk Cohort)

Drug

No. of mother– infant pairs

Antibiotic Amoxicillin 25 Erythromycin 17 Sulfamethoxazole ⫹ 12 trimethoprim Cloxacillin 10 Cephalexin 7 Nitrofurantoin 6 Ampicillin 5 Cefaclor 5 Analgesic/narcotic Acetaminophen 43 Acetaminophen ⫹ codeine 26 ⫹ caffeine Ibuprofen 21 Naproxen 20 Aspirin 15 Antihistamine Terfenadine 25 Diphenhydramine 12 Astemizole 10 Dimenhydrinate 7 Chlorpheniramine 5 Sedative/antidepressant/antiepileptic Carbamazepine 6 Alprazolam 5 Miscellaneous Oral contraceptive 16 Pseudoephedrine 10 5-Aminosalicylic acid 8 Prednisone (prednisolone) 6 Levothyroxine 5 Permethrin 5 Warfarin 5 a

Number of infants with reactions Diarrhea

Drowsiness

Irritability

Others

3 2 0

0 0 0

0 2 0

0 0 2a

2 2 2 1 1

0 0 0 0 0

0 0 0 0 0

0 0 1a 0 0

0 0

0 5

0 1

0 2a

0 0 0

0 2 0

0 0 0

0 1a 0

0 0 0 0 0

0 1 0 0 0

3 0 2 1 0

0 0 0 0 0

0 0

0 1

0 0

0 0

0 0 1 0 0 0 0

1 0 0 0 0 0 0

2 2 0 0 0 0 0

4a 0 0 0 0 0 0

Other reactions included poor feeding (two cases with sulfamethoxazole/trimethoprim), decreased milk volume (one with nitrofurantoin and three with oral contraceptives), constipation (two with acetaminophen ⫹ codeine, one with oral contraceptives), and vomiting (one with naproxen).

data have not substantiated the carcinogenic risk (51). Breast-feeding may be continued during metronidazole therapy. Close supervision of the nursing infant is warranted, since significant amounts may be present in milk and the infant may be at risk for adverse effects such as diarrhea. If a single dose of metronidazole has been prescribed, breastfeeding may be temporarily withheld for 12–24 hours while the drug is being excreted

212

Taddio and Ito

by the body. During this time period, the breasts can be emptied manually to alleviate engorgement and maintain milk flow. Ciprofloxacin and ofloxacin have been measured in milk in small amounts. According to the drug monograph, norfloxacin was not detected in the milk of lactating women after a single dose of 200 mg (52). The use of quinolones during breastfeeding, however, has not gained wide acceptance in the medical community. The quinolones were shown to cause cartilage damage in juvenile animals (53) and were subsequently not recommended for use in children, pregnant women, and nursing women. Many pediatric patients, however, have received quinolone antibiotics. The current published data on over 1000 pediatric patients treated with quinolones has revealed no documentation of unequivocal quinolone-induced arthropathy (54,55). Even though the risk for quinolone-induced arthropathy in the nursing infant is unknown, it is probably very low. Nevertheless, quinolones are not first-line therapy for most infections that nursing women may encounter; other antibiotics are generally preferred. If maternal therapy with a quinolone such as ciprofloxacin is warranted, the data suggest that the baby may continue to nurse. As a precaution, the infants should be monitored closely for signs of arthropathy (e.g., tenderness, stiffness). Infant exposure to acyclovir through breast milk is small. Topical use of acyclovir is compatable with breast-feeding. Oral administration is unlikely to expose the infant to significant plasma levels because of the low bioavailability of the drug. Chloramphenicol should be avoided because of risks of idiosyncratic reactions in the newborn, such as bone marrow suppression. The use of tetracycline during breastfeeding has not been documented to cause problems. However, there is some controversy regarding its use. Even though tetracycline crosses into breast milk in low amounts, its bioavailability in nursing infants may be low due to complexation with calcium or magnesium ions in milk. However the potential risk for tooth mottling in the breast-feeding infant due to tetracycline is unknown. The AAP has recommended that tetracycline is safe to use during lactation (43). In most cases, safer antimicrobials can be substituted for both chloramphenicol and tetracycline. One case report showed that isoniazid is excreted in large amounts in breast milk. Its use during lactation has not been recommended due to concerns for infant liver toxicity.

Anti-inflammatory Agents Salicylic acid has been shown to cross into breast milk in small amounts following maternal acetylsalicylic acid (ASA) ingestion. Occasional use of ASA is not likely to harm the infant. Chronic high-dose ASA therapy, however, has not been adequately studied. Inevitably, more salicylate will be present in milk and the infant may be at risk for salicylate toxicity (56). Other nonsteroidal anti-inflammatory drugs (NSAIDs)—such as diclofenac, ibuprofen, naproxen, flurbiprofen, ketoprofen, ketorolac, and mefenamic acid—have been shown to be present in small amounts and are considered safe to use.

Antihypertensive Drugs The special need for antihypertensive therapy in postpartum women has promoted the study of this class of drugs.

Drugs and Breast-Feeding

213

The transfer of different β-adrenergic blockers into breast milk has been well studied. The primary determinant of such passage for this drug class is protein-binding (57). Drugs that are less lipophilic and less protein-bound, such as atenolol or sotalol, are present in higher quantities. Any infant’s exposure to these drugs will be further increased by poor elimination processes. For example, drug half-lives may be doubled or tripled for drugs excreted renally (58) in the postnatal period. Another consideration is that β-adrenergic blockers used throughout pregnancy may cause hypotension, bradycardia, and respiratory difficulties in newborns. Since a newborn’s heart rate is lowest in the first few days postpartum, the added use of these drugs may be worrisome, especially if the newborn is premature (58). Other antihypertensive agents—such as methyldopa, captopril, enalapril, nifedipine, diltiazem, and verapamil—have been measured in breast milk in small amounts. Adverse effects have not been reported and the use of these drugs is considered compatible with breast-feeding. Anticonvulsants The anticonvulsants carbamazepine, valproic acid, and phenytoin are excreted into breast milk in low amounts. There are case reports of hematological and hepatic adverse effects in infants during maternal use of valproic acid and carbamazepine, respectively. However, breast-feeding is generally considered compatible with their use. Phenobarbital, primidone, and ethosuximide, on the other hand, are excreted in much higher amounts, and infants may be exposed to nearly therapeutic doses. Phenobarbital is usually contraindicated during nursing (Table 3). Infants exposed to these medications via milk may be at risk for adverse effects such as lethargy, sedation, and poor feeding. They should be closely supervised. Infant anticonvulsant plasma concentration monitoring is also advisable. Antihistamines This class of drugs should not stop the nursing mother from breast-feeding. Antihistamines are present in many over-the-counter medications. Although quantitative data on the excre-

Table 3 Drugs Usually Contraindicated During Breast-Feeding Antineoplastics Bromocriptine a Ergotamine Gold Iodine-containing substances Lithium Oral contraceptives a Phenobarbital Radiopharmaceuticals Social drugs and drugs of abuse a

May decrease milk supply.

214

Taddio and Ito

tion of most antihistamines in milk are lacking, their use is generally considered safe during nursing. Specific data on drug excretion in milk is present for loratadine, terfenadine, and triprolidine, all of which have been measured in milk in small amounts. There is one case report in the literature of sedation and irritability in an infant following maternal ingestion of clemastine (59). The Motherisk cohort data includes 54 mother-infant pairs, with 7 (13%) mothers reporting minor adverse effects in their children (Table 1). Antihistamines may be used during lactation; however, infants should be monitored for potential adverse effects. Cardiovascular Drugs Infant consumption of digoxin, flecainide, procainamide, or mexiletine from breast milk is small. Amiodarone and disopyramide, however, are present in larger amounts, which may affect the infant. Whenever possible, drugs levels in breast milk and/or infant’s plasma should be monitored during maternal therapy. During amiodarone therapy, infant thyroid function tests should be monitored as well since amiodarone can cause hyper- or hypothyroidism. Diuretics Diuretics have been found in breast milk in low amounts. The theoretical concern that they may suppress lactation has been raised. However, lactogenesis itself has not be been shown to be influenced by fluid intake (60), and the impact of taking diuretics is unknown. Drugs Acting on the Central Nervous System Many puerperial women may require these medications to alleviate symptoms of depression, anxiety, or other psychiatric conditions. Most of the information available for these drugs, however, is in the form of case reports. Generalizations regarding ‘‘risk’’ are difficult to make. Case reports and cohort studies of various tricyclic antidepressants and selective serotonin reuptake inhibitors (SSRIs) measured in breast milk show that infant exposure is consistently low for drugs of this class. Breast-feeding may be continued during antidepressant use, keeping in mind that the effects of long-term exposure of these drugs on the developing central nervous system are not known. Similarly, excretion of some neuroleptics into breast milk has been studied. Only small amounts have been found, but potential long-term behavioral effects are unknown. Lithium therapy during breastfeeding warrants concern because the drug freely passes into breast milk and significant plasma levels have been detected in nursing infants (Table 2). Lithium is contraindicated by many health professionals during lactation because of this (Table 3). If a woman chooses to nurse during lithium therapy, however, infant plasma lithium levels should be monitored. Benzodiazepines should be used with caution by breast-feeding women. Metabolizing capacity of these drugs may be limited in the neonatal period and accumulation may occur. If benzodiazepine therapy is necessary, short-term treatment with a short-acting agent devoid of pharmacologically active metabolites, such as lorazepam or alprazolam, is preferable (61).

Drugs and Breast-Feeding

215

Thyroid and Antithyroid Drugs Levothyroxine is compatible with breast-feeding. Thyroid hormones cross into breast milk in low amounts. Their presence is not likely to affect the infant’s thyroid (62–65). Propylthiouracil crosses into breast milk in small amounts and is the antithyroid drug of choice during breast feeding. Infants have been safely breastfed on maternal doses of 50–300 mg/day (66). Thyroid function tests should be performed in the infant on a regular basis (67,68), if warranted, to monitor for toxicity. Gastrointestinal Drugs and Laxatives H 2-receptor blockers such as cimetidine and ranitidine should be used with caution by breast-feeding mothers. To date, adverse effects have not been reported in nursing infants, but the potential for inhibition of gastric acid secretion and enzyme metabolizing capability exist. Aluminum and magnesium salts are considered safe to use (69,70). Sucralfate is poorly absorbed from the gastrointestinal tract and is not likely to cause adverse effects in the nursing infant (70). Metoclopramide and domperidone are excreted into milk in small amounts. Both increase prolactin levels (71–73), and metoclopramide has increased milk production in women with faltering milk production (72–74). Several investigators studied the clinical effects on nursing infants after maternal ingestion of laxatives. Phenolphthalein (75) mineral oil, magnesia, and senna glycosides (76,77) have all been associated with normal infant bowel patterns after short-term maternal therapy. No harmful effects from docusate have been observed (78). Nonetheless, lactating women are generally advised to use drugs that are poorly absorbed from the gastrointestinal tract, such as psyllium or other bulk-forming agents, instead. Miscellaneous 5-Aminosalicylic acid and sulfasalazine can be used by the lactating mother. The infant should be monitored for adverse reactions, such as diarrhea. Chloroquine and hydroxychloroquine are both excreted in breast milk. Adverse effects have not been reported with their use. Once-weekly dosing of chloroquine for malaria is unlikely to expose the nursing infant to toxic amounts. There are no reports on the safety of daily chloroquine administration for arthritis. Since the milk half-life of chloroquine is approximately 1 week (Table 1), daily administration would inevitably lead to accumulation of the drug in milk. Prednisone and prednisolone have been detected in milk in small amounts. Breastfeeding need not be interrupted. Because systemic absorption following topical and inhalational steroids are usually not substantial, they are also safe to use during lactation. Theophylline is excreted in milk. Since clearance of theophylline is low in the neonatal period, close monitoring of the infant is necessary to detect adverse effects. Warfarin is highly protein-bound. Levels in milk have been consistently below detection limits and no adverse effects have been reported in nursing infants. Heparin therapy is also compatable with breast-feeding. It does not pass into milk (79). Topical Agents The excretion of many topically applied drugs into breast milk has not been studied. This fact does not, however, preclude their use in nursing women. For example, the excretion

216

Taddio and Ito

of inhalational β-adrenergic agonists (e.g., salbutamol) in breast milk have not been studied, but bioavailability of these agents is low. Breast-feeding during their use is considered safe. Similarly, ipratropium and sodium cromoglycate are unlikely to cause adverse effects. Many other topical agents (eyedrops, creams, ointments, etc.), with low bioavailability and a low order or toxicity may be used during the lactation period as well. Examples of these agents are lidocaine ointment, nystatin cream, or clotrimazole cream.

DRUGS USUALLY CONTRAINDICATED DURING BREAST-FEEDING Table 3 lists drugs commonly avoided during lactation. The following section reviews the reasons why these drugs are not used. Some drugs that fall into this category are discussed under ‘‘Review of Selected Drug Classes,’’ above. Antineoplastics The use of antineoplastic drugs is generally contraindicated during lactation. Even if the absolute dose the infant is exposed to is small, the possibility for immediate or long-term toxicity precludes breast-feeding during their use. Azathioprine and methotrexate are two of the more commonly used antineoplastic agents in ambulatory-care patients. Both have been measured in breast milk. The peak milk concentration of 6-mercaptopurine, the active metabolite of azathioprine, was 18 ng/mL in a woman taking 25 mg of azathioprine daily and 4.5 ng/mL in a woman taking 75 mg daily (80). There are three cases of infants breastfed during maternal azathioprine therapy without adverse effects (80,81). Methotrexate secretion in milk was measured in a patient receiving 22.5 mg daily. The peak milk methotrexate level was only 2.6 ng/mL and the highest M/P ratio was 0.08 (82). Based on this data, some have argued that methotrexate may be safe to use in rheumatic diseases (83), where it is administered only once weekly. Further study seems warranted. Lactating mothers requiring these medications should be well informed of the risks to the baby. If these drugs are used during lactation, the following parameters should be strictly monitored: milk drug levels, infant plasma drug levels, and infant hematological parameters. Radiopharmaceuticals The use of radiopharmaceuticals usually requires temporary cessation of breast-feeding; this is done to avoid infant exposure to excessive radioactivity. The time necessary before most radioactivity is excreted varies with the radioisotope used; reference texts or nuclear departments can be consulted for this information. While the mother is awaiting clearance of the isotope, she may use breast pumping to preserve nursing function. Since most diagnostic testing is planned in advance, mothers may express milk (and keep it frozen) before the procedure. This milk will be given to the baby during the time that it is not nursing. Ergot Alkaloids Bromocriptine, an ergot alkaloid, is contraindicated during breast-feeding. It possesses prolactin-suppressing activity and is used to prevent lactation (84,85). Ergotamine is

Drugs and Breast-Feeding

217

avoided due to risks of milk suppression (86) and ergotism in the infant (87). Methylergonovine, however, is used to assist uterine involution. Short-term therapy with methylergonovine is not associated with adverse effects. Social Drugs and Drugs of Abuse Social drugs such as alcohol are not recommended for use during lactation. Ethanol freely distributes into milk and levels in milk are similar to those in plasma. Elimination from milk is also similar to elimination from blood (88,89). Occasional alcohol consumption exposes the nursing infant to small amounts of ethanol and is not considered harmful, but the infant may be at risk for sedation with higher doses (90). At higher doses, ethanol presents a risk to successful lactation also, as oxytocin release is blocked (91). Infant motor development is affected by maternal alcohol ingestion in a dose-dependent fashion (92). Interestingly, both alcoholic and nonalcoholic beer has been shown to increase prolactin secretion (93). In summary then, alcohol use should be limited during lactation. Smoking during lactation is also not recommended. A nursing mother should be encouraged to stop smoking, as there are well-documented health risks to herself and her nursing infant. Nicotine has been measured in milk (94). Cotinine, the major metabolite of nicotine, has also been measured in milk (95,96). In fact, infant urine cotinine levels have been correlated with cotinine levels in maternal breast milk (97). Maternal smoking has been linked to a higher incidence of respiratory tract infections in infants (97). Smoking has also been shown to have a negative effect on breastmilk production (98). Smoking may be a contributing factor to infant colic (99). In addition, infants may also be exposed to other chemicals from cigarette smoking, such as carcinogens. As such, passive smoking by the infant (100,101) may lead to long-term adverse effects. The use of ‘‘street drugs’’ by lactating women is contraindicated. The first concern should be that it may be difficult for nursing women using these drugs to care for their infants properly if they are experiencing an altered state of mind. Also, there exists the potential for drug exposure through breast milk (or passive smoking) and, therefore, adverse effects on the infant. These drugs are typically very potent, and smaller doses can have pharmacological effects. In a case report, cocaine and its metabolite were present in a mother’s milk after intranasal use and in the urine of her infant, who showed signs of cocaine toxicity (102). Phencyclidine has been detected in breast milk (103). THC (delta-9-tetrahydrocannabinol), considered to be the main psychoactive component of marijuana, is also excreted into breast milk (104). In a study of infant development after marijuana exposure during pregnancy and lactation, maternal use of marijuana in the first month postpartum was associated with lower motor development scores at 1 year of age in exposed infants compared to unexposed infants (105). Marijuana has also been shown to decrease plasma prolactin levels in women (106). Gold The use of gold therapy during breast-feeding is controversial. Variable milk levels have been reported and the extent of absorption by the infant is unknown. In one report, the mean milk level of aurothioglucose was 41 µg/L in a nursing woman receiving weekly injections of 50 mg. The M/S ratio was 0.01. The drug was not present in her infant’s serum or urine (limit of detection 5 ⫻ 10⫺7 mg/L). (107) In another report, sodium aurothio-

218

Taddio and Ito

malate given weekly at a dose of 25 mg produced maximum milk levels of 40 µg/L, and the infant’s urine contained 0.4 µg/L. The infant experienced transient facial edema, which could not be explained (108). The highest milk level reported in a woman receiving 10 mg of sodium aurothiomalate monthly was 93 µg/L. The M/S ratio varied from 0.02– 0.03. The infant plasma level was 51 µg/L; no adverse reactions were observed (109). Nursing infants may have continued exposure to gold long after a dose is given to the mother due to gold’s long half-life (approximately 1 week in serum) (107). Gold has also been shown to be excreted in increasing concentrations over time in milk (107,109). Infants who are nursing during maternal gold therapy should be monitored very closely for adverse effects. Iodine Drugs containing iodine (e.g., potassium iodide, povidone-iodine) are not recommended during breast-feeding. Iodine is transported into breast-milk and can result in iodine-induced goiter and hypothyroidism in the infant (110). Extensive topical administration and vaginal use of povidone-iodine by breast-feeding mothers has led to transient congenital hypothyroidism (111,112) and grossly elevated serum iodine levels in their infants (113). Oral Contraceptives Much controversy has surrounded the use of oral contraceptives during lactation. The topic has been extensively reviewed by Koetsawang (114), who concluded that combination estrogen and progestogen oral contraceptives, even low-dose, forms, can decrease milk yield. Progestogen-only contraceptives are associated with less changes. The World Health Organization (115) recently studied the effects of different types of contraception on milk and in nursing infants. They found that combined oral contraceptives caused slight changes in milk composition as well as a decrease in milk yield. Hormonal contraceptives, however, were not associated with any significant difference in infant weight, fat fold, or rate of discontinuation of breast-feeding by the mothers compared to a control group. Oral contraceptives are not recommended for use in early lactation or in nutritionally deficient mothers because the potential adverse effects on milk supply may be clinically significant in these women. Overall, progestogen-only contraceptives may be preferable. Initiation of oral contraceptives during nursing should be delayed. However, there is no consensus as to how long (116). Nursing prolongs amenorrhea in postpartum women; therefore oral contraceptives need not be started right away. Oral contraceptives should probably not begin until breast-feeding is fully established (approximately 6 weeks) or even later.

CONCLUSIONS Most drugs taken by breast-feeding women are excreted into breast milk and a determination of the risk to the infant must be made. For most drugs, risk assessment is made by considering the dose delivered to the infant from nursing. Other factors in risk assessment include considering the toxicity profile of the drug. The amount of drug consumed by the infant during nursing is usually less than 5% of the maternal dose (milligrams per kilogram). This small amount is tolerable without

Drugs and Breast-Feeding

219

toxicity by most infants. Common sense dictates that even though many drugs may be taken safely during nursing, mothers should not be exposed to them unnecessarily. If maternal drug therapy is required, the specific agents used should be chosen with the consideration of minimizing any risks to the infant. In this way, interruption of breastfeeding is rarely justified. The following principles should be applied: the drug used should be appropriate for treating the mother’s condition, it should be unlikely to cause adverse effects in the infant, and finally, the infant should be monitored for potential adverse effects. Clinical Case Answer The woman should be told that fluoxetine and its active metabolite are excreted into milk and may be detected in infant urine. The baby should be observed for adverse effects.

REFERENCES 1. Forman MR. Review of research on the factors associated with choice and duration of infant feeding in less-developed countries. Pediatrics 1984; 74(suppl):667–694. 2. Simopoulos AP, Grave GD. Factors associated with the choice and duration of infant-feeding practice. Pediatrics 1984; 74(suppl):603–614. 3. Matheson I. Drugs taken by mothers in the puerperium. BMJ 1985; 290:1588–1589. 4. Passmore CM, McElnay JC, D’Arcy PF. Drugs taken by mothers in the puerperium: inpatient survey in Northern Ireland. BMJ 1984; 289:1593–1596. 5. American Academy of Pediatrics Working Group on Breastfeeding. Breastfeeding and the use of human milk. Pediatrics 1997; 100(6):1035–1539. 6. Berlin CM. The excretion of drugs and chemicals in human milk. In: Yaffe SJ, ed. Pediatric Pharmacology: Therapeutic Principles in Practice. New York: Grune & Stratton, 1980, pp 137–147. 7. Kirksey A, Groziak SM. Maternal drug use: evaluation of risks to breast-fed infants. World Rev Nutr Diet 1984; 43:60–79. 8. Catz CS, Giacoia GP. Drugs and breast milk. Pediatr Clin North Am 1972; 19:151–166. 9. Noel GL, Suh HK, Frantz AG. Prolactin release during nursing and breast stimulation in postpartum and nonpostpartum subjects. J Clin Endocrinol Metab 1974; 38:413–423. 10. Vorherr H. The Breast: Morphology, Physiology and Lactation. New York: Academic Press, 1974, p 73. 11. Jenness R. The composition of human milk. Semin Perinatol 1979; 3:225–239. 12. Casey CE, Hambidge KM. Nutritional aspects of human lactation. In: Neville MC, Neifert MR, eds. Lactation Physiology, Nutrition, and Breast-Feeding. New York: Plenum Press, 1983, pp 199–248. 13. Jenness R. The composition of human milk. Semin Perinatol 1979; 3:225–239. 14. Hartmann PE, Prosser CG. Physiological basis of longitudinal changes in human milk yield and composition. Fed Proc 1984; 43:2448–2453. 15. Lonnerdal B. Effects of maternal dietary intake on human milk composition. J Nutr 1986; 116:499–513. 16. Pickles VR. Blood-flow estimations as indices of mammary activity. J Obstet Gynaecol Br Emp 1953; 60:301–311. 17. Neville MC, Keller R, Seacat J, et al. Studies in human lactation: milk volumes in lactating women during the onset of lactation and full lactation. Am J Clin Nutr 1988; 48:1375–1386. 18. Dewey KG, Heinig MJ, Nommsen LA, et al. Maternal versus infant factors related to breast milk intake and residual milk volume: the Darling study. Pediatrics 1991; 87:829–837.

220

Taddio and Ito

19. Morriss FH, Brewer ED, Spedale SB, et al. Relationship of human milk pH during course of lactation to concentrations of citrate and fatty acids. Pediatrics 1986; 78:458–464. 20. Prentice A, Prentice AM, Whitehead RG. Breast-milk fat concentrations of rural African women: 1. Short-term variations within individuals. Br J Nutr 1981; 45:483–494. 21. Hytten FE. Clinical and chemical studies in human lactation. BMJ 1954; 1:175–182. 22. Lammi-Keefe CJ, Ferris AM, Jensen RG. Changes in human milk at 0600, 1000, 1400, 1800, and 2200h. J Pediatr Gastroenterol Nutr 1990; 11:83–88. 23. Changes with time in the lipids of human milk. Nutr Rev 1984; 42:12–13. 24. Anderson GH. The effect of prematurity on milk composition and its physiological basis. Fed Proc 1984; 43:2438–2442. 25. Besunder JB, Reed MD, Blumer JL. Principles of drug biodisposition in the neonate. a critical evaluation of the pharmacokinetic-pharmacodynamic interface (part 1). Clin Pharmacokinet 1988; 14:189–216. 26. Dutton GJ. Developmental aspects of drug conjugation, with special reference to glucuronidation. Annu Rev Pharmacol Toxicol 1978; 18:17–35. 27. Atkinson HC, Begg EJ, Darlow BA. Drugs in human milk: clinical pharmacokinetic considerations. Clin Pharmacokinet 1988; 14:217–240. 28. Schanker LS. Passage of drugs across body membranes. Pharmacol Rev 1962; 14:501– 530. 29. Wilson JT. Determinants and consequences of drug excretion in breast milk. Drug Metab Rev 1983; 14:619–652. 30. Wilson JT, Brown RD, Hinson JL, et al. Pharmacokinetic pitfalls in the estimation of the breast milk/plasma ratio for drugs. Annu Rev Pharmacol Toxicol 1985; 25:667–689. 31. Benet LZ. General treatment of linear mammillary models with elimination from any compartment as used in pharmacokinetics. J Pharm Sci 1972; 61:536–541. 32. Miller GE, Banerjee NC, Stowe CM, Diffusion of certain weak organic acids and bases across the bovine mammary gland membrane after systemic administration. J Pharmacol Exp Ther 1967; 157:245–253. 33. Rasmussen F. Mammary excretion of sulphonamides. Acta Pharmacol (Kbh) 1958; 15:139– 148. 34. Rasmussen F. Mammary excretion of benzylpenicillin, erythromycin and penethamate hydroiodide. Acta Pharmacol (Kbh) 1959; 16:194–200. 35. Sisodia CS, Stowe CM. The mechanism of drug secretion into bovine milk. Ann NY Acad Sci 1964; 111:650–661. 36. Wilson JT, Brown RD, Cherek DR, et al. Drug excretion in human breast milk: principles, pharmacokinetics and projected consequences. Clin Pharmacokinet 1980; 5:1–66. 37. Fleishaker JC, Desai N, McNamara PJ. Factors affecting the milk-to-plasma drug concentration ratio in lactating women: physical interactions with protein and fat. J Pharm Sci 1987; 76:189–193. 38. Atkinson HC, Begg EJ. Prediction of drug concentrations in human skim milk from plasma protein binding and acid-base characteristics. Br J Clin Pharmacol 1988; 25:495–503. 39. Atkinson HC, Begg EJ. Prediction of drug distribution into human milk from physiochemical characteristics. Clin Pharmacokinet 1990; 18:151–167. 40. Ito S, Koren G. A novel index for expressing exposure of the infant to drugs in breast milk. Br J Clin Pharmacol 1994; 38:99. 41. Bennett PN, ed. Drugs and Human Lactation, 2nd ed. Amsterdam: Elsevier, 1996. 42. Briggs GG, Freeman RK, Yaffe SJ. Drugs in Pregnancy and Lactation, 4th ed. Baltimore: Williams & Wilkins, 1994. 43. American Academy of Pediatrics, Committee on Drugs. The transfer of drugs and other chemicals into human milk. Pediatrics 1994; 93:137–150. 44. Matheson I, Lunde PKM, Notarianni L. Infant rash caused by paracetamol in breast milk? Pediatrics 1985; 76:651–652.

Drugs and Breast-Feeding

221

45. Schmimmel MS, Eidelman AJ, Wilschanski MA, et al. Toxic effects of atenolol consumed during breast feeding. J Pediatr 1989; 114:476–478. 46. Cunningham AS, Jelliffe DB, Jelliffe EFP. Breast-feeding and health in the 1980s: a global epidemiologic review. J Pediatr 1991; 118:659–666. 47. Karjalainen J, Martin JM, Knip M, et al. A bovine albumin peptide as a possible trigger of insulin-dependent diabetes mellitus. N Engl J Med 1992; 327:302–307. 48. Koletzko S, Sherman P, Corey M, et al. Role of infant feeding practices in development of Crohn’s disease in childhood. BMJ 1989; 298:1617–1618. 49. Naumburg EG, Meny RG. Breast milk opioids and neonatal apnea. Am J Dis Child 1988; 142:11–12. 50. Rhone-Poulenc. Flagyl. In: Krogh CME, ed. Compendium of Pharmaceuticals and Specialties. 24th ed. Ottawa: Canadian Pharmaceutical Association, 1989, pp 409–410. 51. Beard CM, Noller KL, O’Fallon WM, et al. Lack of evidence for cancer due to use of metronidazole. N Engl J Med 1979; 301:519–522. 52. Stoukides CA. The galactopharmacopedia: quinolone antibiotics and breastfeeding. J Hum Lact 1991; 7:143–144. 53. Schluter G. Toxicology of ciprofloxacin. In: Neu HC, Weuta H, eds. First International Ciprofloxacin Workshop. Leverkusen, Germany: Excerpta Medica, 1985, pp 61–67. 54. Chysky V, Kapila K, Hullmann R, et al. Safety of ciprofloxacin in children: worldwide clinical experience based on compassionate use—emphasis on joint evaluation. Infection 1991; 19:289–296. 55. Schaad UB, Stoupis C, Wedgwood J, et al. Clinical, radiologic and magnetic resonance monitoring for skeletal toxicity in pediatric patients with cystic fibrosis receiving a three-month course of ciprofloxacin. Pediatr Infect Dis J 1991; 10:723–729. 56. Needs CJ, Brooks PM. Antirheumatic medication during lactation. Br J Rheumatol 1985; 24:291–297. 57. Riant P, Urien S, Albengres E, et al. High plasma protein binding as a parameter in the selection of beta blockers for lactating women. Biochem Pharmacol 1986; 35:4579–4581. 58. Boutroy MJ. Fetal and neonatal effects of the beta-adrenoceptor blocking agents. Dev Pharmacol Ther 1987; 10:224–231. 59. Kok THHG, Taitz LS, Bennett MJ, et al. Drowsiness due to clemastine transmitted in breast milk. Lancet 1982; 1:914–915. 60. Duckman S, Hubbard JF. The role of fluids in relieving breast engorgement without the use of hormones. Am J Obstet Gynecol 1950; 60:200–204. 61. Kanto JH. Use of benzodiazepines during pregnancy, labour and lactation, with particular reference to pharmacokinetic considerations. Drugs 1982; 23:354–380. 62. Mallol J, Obregon MJ, Morreale de Escobar G. Analytical artifacts in radioimmunoassay of L-thyroxin in human milk. Clin Chem 1982; 28:1277–1282. 63. Moller B, Bjorkhem I, Falk O, et al. Identification of thyroxine in human breast milk by gas chromatography–mass spectrometry. J Clin Endocrinol Metab 1983; 56:30–34. 64. Jansson L, Ivarsson S, Larsson I, et al. Tri-iodothyronine and thyroxine in human milk. Acta Paediatr Scand 1983; 73:703–705. 65. Mizuta H, Amino N, Ichihara K, et al. Thyroid hormones in human milk and their influence on thyroid function of breast-fed babies. Pediatr Res 1983; 17:468–471. 66. Momotani N, Yamashita R, Yoshimoto M, et al. Recovery from foetal hypothyroidism: evidence for the safety of breast-feeding while taking propylthiouracil. Clin Endocrinol 1989; 31:591–595. 67. Cooper DS. Antithyroid drugs: to breast-feed or not to breast-feed. Am J Obstet Gynecol 1987; 157:234–235. 68. Cooper DS. Antithyroid drugs. N Engl J Med 1984; 311:1353–1362. 69. McGuire TM, Mitchell IB, Wright AH, et al. Update on excretion of drugs in breast milk— part 2. Aust J Hosp Pharm 1988; 18:150–164.

222

Taddio and Ito

70. Lewis JH, Weingold AB, and the Committee on FDA-related matters, American College of Gastroenterology: the use of gastrointestinal drugs during pregnancy and lactation. Am J Gastroenterol 1985; 80:912–923. 71. Camanni F, Genazzani AR, Massara F, et al. Prolactin-releasing effect of domperidone in normoprolactinemic and hyperprolactinemic subjects. Neuroendocrinology 1980; 30:2–6. 72. Kauppila A, Anunti P, Kivinen S, et al. Metoclopramide and breast feeding: efficacy and anterior pituitary responses of the mother and the child. Eur J Obstet Gynecol Reprod Biol 1985; 19:19–22. 73. Ehrenkranz RA, Ackerman BA. Metoclopramide effect on faltering milk production by mothers of premature infants. Pediatrics 1986; 78:614–620. 74. Sousa PLR. Metoclopramide and breast-feeding. BMJ 1975; 1:513. 75. Fantus B, Dyniewicz JM. Phenolphthalein administration to nursing women. Am J Dig Dis Nutr 1936; 3:184–185. 76. Baldwin WF. Clinical study of senna administration to nursing mothers: assessment of effects on infant bowel habits. Can Med Assoc J 1963; 89:566–568. 77. Werthmann MW, Krees SV. Quantitative excretion of senokot in human breast milk. Med Ann DC 1973; 42:4–5. 78. White GJ, White M. Breastfeeding and drugs in human milk. Vet Hum Toxicol 1984; 26: (suppl 1). 79. O’Reilly RA. Anticoagulant, antithrombotic, and thrombolytic drugs. In: Gilman AG, Goodman LS, Rall TW, et al, eds. Goodman and Gilman’s the Pharmacological Basis of Therapeutics, 7th ed. Toronto: Collier MacMillan Canada, 1985, pp 1338–1359. 80. Coulam CB, Moyer TP, Jiang N-S, et al. Breast-feeding after renal transplantation. Transplant Proc 1982; 13:605–609. 81. Grekas DM, Vasiliou SS, Lazarides AN. Immunosuppressive therapy and breast-feeding after renal transplantation. Nephron 1984; 37:68. 82. Johns DG, Rutherford LD, Leighton PC, et al. Secretion of methotrexate into human milk. Am J Obstet Gynecol 1972; 112:978–980. 83. Brooks PM, Needs CJ. The use of antirheumatic medication during pregnancy and in the puerperium. Rheum Dis Clin North Am 1989; 15:789–806. 84. Peters F, Del Pozo E, Conti A, et al. Inhibition of lactation by a long-acting bromocriptine. Obstet Gynecol 1986; 67:82–85. 85. Scapin F, Buonaccorsi S, Tronconi G, et al. Metergoline versus bromocriptine in the prevention of puerperal lactation: a double-blind clinical trial. Eur J Clin Pharmacol 1982; 22:181– 183. 86. Varga L, Lutterbeck PM, Pryor JS, et al. Suppression of puerperal lactation with an ergot alkaloid: a double-blind study. BMJ 1972; 2:743–744. 87. Illingworth RS. Abnormal substances excreted in human milk. Practitioner 1953; 171:533– 538. 88. Kesaniemi YA. Ethanol and acetaldehyde in the milk and peripheral blood of lactating women after ethanol administration. J Obstet Gynaecol Br Commonw 1974; 81:84–86. 89. Lawton ME. Alcohol in breast milk. Aust N Z J Obstet Gynaecol 1985; 25:71–73. 90. Wyckerheld Bisdom CJ. Alcohol and nicotine poisoning in nurslings. JAMA 1937; 109:178. 91. Cobo E. Effect of different doses of ethanol on the milk-ejecting reflex in lactating women. Am J Obstet Gynecol 1973; 115:817–821. 92. Little RE, Anderson KW, Ervin CH, et al. Maternal alcohol use during breast-feeding and infant mental and motor development at one year. N Engl J Med 1989; 321:425–430. 93. Carlson HE, Wasser HL, Reidelberger RD. Beer-induced prolactin secretion: a clinical and laboratory study of the role of salsolinol. J Clin Endocrinol Metab 1985; 60:673–677. 94. Ferguson BB, Wilson DJ, Schaffner W. Determination of nicotine concentrations in human milk. Am J Dis Child 1976; 130:837–839.

Drugs and Breast-Feeding

223

95. Labrecque M, Marcoux S, Weber J-P, et al. Feeding and urine cotinine values in babies whose mothers smoke. Pediatrics 1989; 83:93–97. 96. Schwartz-Bickenbach D, Schulte-Hobein B, Abt S, et al. Smoking and passive smoking during pregnancy and early infancy: effects on birth weight, lactation period, and cotinine concentrations in mother’s milk and infant’s urine. Toxicol Lett 1987; 35:73–81. 97. Schulte-Hobein B, Schwartz-Bickenbach D, Abt S, et al. Cigarette smoke exposure and development of infants throughout the first year of life: influence of passive smoking and nursing on cotinine levels in breast milk and infants’ urine. Acta Paediatr 1992; 81:550–557. 98. Vio F, Salazar G, Infante C. Smoking during pregnancy and lactation and its effects on breast-milk volume. Am J Clin Nutr 1991; 54:1011–1016. 99. Said G, Patois E, Lellouch J. Infantile colic and parental smoking. BMJ 1984; 289:660. 100. Greenberg RA, Haley NJ, Etzel RA, et al. Measuring the exposure of infants to tobacco smoke: nicotine and cotinine in urine and saliva. N Engl J Med 1984; 310:1075–1078. 101. Luck W, Nau H. Nicotine and cotinine concentrations in serum and urine of infants exposed via passive smoking or milk from smoking mothers. J Pediatr 1985; 107:816–820. 102. Chasnoff IJ, Lewis DE, Squires L. Cocaine intoxication in a breast-fed infant. Pediatrics 1987; 80:836–838. 103. Kaufman KR, Petrucha RA, Pitts FN, et al. PCP in amniotic fluid and breast milk: case report. J Clin Psychiatry 1983; 44:269–270. 104. Perez-Reyes M, Wall ME. Presence of ∆9-tetrahydrocannabinol in human milk. N Engl J Med 1982; 307:819–820. 105. Astley SJ, Little RE. Maternal marijuana use during lactation and infant development at one year. Neurotoxicol Teratol 1990; 12:161–168. 106. Mendelson JH, Mello NK, Ellingboe J. Acute effects of marihuana smoking on prolactin levels in human females. J Pharmacol Exp Ther 1985; 232:220–222. 107. Rooney TW, Lorber A, Veng-Pedersen P, et al. Gold pharmacokinetics in breast milk and serum of a lactating woman. J Rheumatol 1987; 14:1120–1122. 108. Bell RAF, Dale IM. Gold secretion in maternal milk. Arthritis Rheum 1976; 19:1374. 109. Bennett PN, Humphries SJ, Osborne JP, et al. Use of sodium aurothiomalate during lactation. Br J Clin Pharmacol 1990; 29:777–779. 110. Braverman LE. Iodine induced thyroid disease. Acta Med Austr 1990; 17:29–33. 111. Delange F, Chanoine JP, Abrassart C, et al. Topical iodine, breastfeeding, and neonatal hypothyroidism. Arch Dis Child 1988; 63:106–107. 112. Danziger Y, Pertzelan A, Mimouni M. Transient congenital hypothyroidism after topical iodine in pregnancy and lactation. Arch Dis Child 1987; 62:295–296. 113. Postellon DC, Aronow R. Iodine in mother’s milk. JAMA 1982; 247:463. 114. Koetsawang S. The effects of contraceptive methods on the quality and quantity of breast milk. Int J Gynaecol Obstet 1987; 25(suppl):115–127. 115. World Health Organization (WHO) task force on oral contraceptives, special programme of research, development, and research training in human reproduction. Effects of hormonal contraceptives on breast milk composition and infant growth. Stud Fam Plann 1988; 19:361– 369. 116. Laukaran VH. The effects of contraceptive use on the initiation and duration of lactation. Int J Gynaecol Obstet 1987; 25(Suppl):129–142. 117. Berlin CM, Yaffe SJ, Ragni, M. Disposition of acetaminophen in milk, saliva, and plasma of lactating women. Pediatr Pharmacol (New York) 1980; 1:135–141. 118. Notarianni LJ, Oldham HG, Bennett PN. Passage of paracetamol into breastmilk and its subsequent metabolism by the neonate. Br J Clin Pharmacol 1987; 24:63–67. 119. Bitzen P-O, Gustafsson B, Jostell KG, et al. Excretion of paracetamol in human breast milk. Eur J Clin Pharmacol 1981; 20:123–125. 120. Findlay JWA, DeAngelis RL, Kearney MF et al. Analgesic drugs in breast milk and plasma. Clin Pharmacol Ther 1981; 29:625–633.

224

Taddio and Ito

121. Quinn PG, Kuhnert BR, Kaine CJ, et al. Measurement of meperidine and normeperidine in human breast milk by selected ion monitoring. Biomed Environ Mass Spectrom 1986; 13: 133–135. 122. Wittels B, Scott DT, Sinatra RS. Exogenous opioids in human breast milk and acute neonatal neurobehavior: a preliminary study. Anesthesiology 1990; 73:864–869. 123. Feilberg VL, Rosenborg D, Christensen CB et al. Excretion of morphine in human breast milk. Acta Anaesthesiol Scand 1989; 33:426–428. 124. Robieux I, Koren G, Vandenbergh H, et al. Morphine excretion in breast milk and resultant exposure of a nursing infant. Clin Toxicol 1990; 28:365–370. 125. Marx CM, Pucino F, Carlson JD, et al. Oxycodone excretion in human milk in the puerpium. DICP 1986; 20:474. 126. Lau RJ, Emery MG, Galinsky RE. Unexpected accumulation of acyclovir in breast milk with estimation of infant exposure. Obstet Gynecol 1987; 69:468–471. 127. Meyer LJ, de Miranda P, Sheth N, et al. Acyclovir in human breast milk. Am J Obstet Gynecol 1988; 158:586–588. 128. Taddio A, Klein J, Koren G. Acyclovir excretion in human breast milk. Ann Pharmacother 1994; 28:585–587. 129. Kafetzis DA, Siafas CA, Georgakopoulos PA, et al. Passage of cephalosporins and amoxicillin into the breast milk. Acta Paediatr Scand 1981; 70:285–288. 130. Campbell AC, McElnay JC, Passmore CM. The excretion of ampicillin in breast milk and its effect on the suckling infant. Br J Clin Pharmacol 1991; 31:230P. 131. Matsuda S. Transfer of antibiotics into maternal milk. Biol Res Pregnancy Perinatol 1984; 5:57–60. 132. Matheson I, Samseth M, Sande HA. Ampicillin in breast milk during puerperal infections. Eur J Clin Pharmacol 1988; 34:657–659. 133. Havelka J, Hejzlar M, Popov V, et al. Excretion of chloramphenicol in human milk. Chemotherapy 1968; 13:204–211. 134. Plomp TA, Thiery M, Maes RAA. The passage of thiamphenicol and chloramphenicol into human milk after single and repeated oral administration. Vet Hum Toxicol 1983; 25:167– 172. 135. Cover DL, Mueller BA. Ciprofloxacin penetration into human breast milk: a case report. DICP 1990; 24:703–704. 136. Gardner DK, Gabbe SG, Harter C. Simultaneous concentrations of ciprofloxacin in breast milk and in serum in mother and breast-fed infant. Clin Pharm 1992; 11:352–354. 137. Giamarellou H, Kolokythas E, Petrikkos G, et al. Pharmacokinetics of three newer quinolones in pregnant and lactating women. Am J Med 1989; 87:49S–51S. 138. Steen B, Rane A. Clindamycin passage into human milk. Br J Clin Pharmacol 1982; 13: 661–664. 139. Smith JA, Morgan JR, Rachlis AR, et al. Clindamycin in human breast milk. Can Med Assoc J 1975; 112:806. 140. Mann CF. Clindamycin and breast-feeding. Pediatrics 1980; 66:1030. 141. Schilling CG, Seay RE, Larson TA, Meier KR. Excretion of fluconazole in human breast milk. Pharmacotherapy 1993; 13:287. 142. Force RW. Fluconazole concentrations in breast milk. Pediatr Infect Dis J 1995; 14:235–236. 143. Berlin CM, Lee C. Isoniazid and acetylisoniazid disposition in human milk, saliva and plasma. Fed Proc 1979; 38:426. 144. Moretti ME, Ito S, Koren G. Disposition of maternal ketoconazole in breast milk. Am J Obstet Gynecol 1995; 173:1625–1626. 145. Passmore CM, McElnay JC, Rainey EA, et al. Metronidazole excretion in human milk and its effect on the suckling neonate. Br J Clin Pharmacol 1988; 26:45–51. 146. Heisterberg L, Branebjerg PE. Blood and milk concentrations of metronidazole in mothers and infants. J Perinat Med 1983; 11:114–120.

Drugs and Breast-Feeding

225

147. Erickson SH, Oppenheim GL, Smith GH. Metronidazole in breast milk. Obstet Gynecol 1981; 57:48–50. 148. Traeger A, Peiker G. Excretion of nalidixic acid via mother’s milk. Arch Toxicol 1980; 4(suppl):388–390. 149. Belton EM, Jones RV. Haemolytic anaemia due to nalidixic acid. Lancet 1965; 2:691. 150. Varsano I, Fischl J, Shochet SB. The excretion of orally ingested nitrofurantoin in human milk. J Pediatr 1973; 82:886–887. 151. Hosbach RE, Foster RB. Absence of nitrofurantoin from human milk. JAMA 1967; 202: 145. 152. Pons G, Rey E, Richard M-O, et al. Nitrofurantoin excretion in human milk. Dev Pharmacol Ther 1990; 14:148–152. 153. Greene HJ, Burkhart B, Hobby GL, et al. Excretion of penicillin in human milk following parturition. Am J Obstet Gynecol 1946; 51:732–733. 154. Matheson I, Samseth M, Loberg R, et al. Milk transfer of phenoxymethylpenicillin during puerperal mastitis. Br J Clin Pharmacol 1988; 25:33–40. 155. Miller RD, Salter AJ. The passage of trimethoprim/sulphamethoxazole into breast milk and its significance. In: Daikos CK, ed. Progress in Chemotherapy. Vol 1. Antibacterial Chemotherapy. Athens: Hellenic Society for Chemotherapy, 1974, pp 687–691. 156. Kauffman RE, O’Brien C, Gilford P. Sulfisoxazole secretion into human milk. J Pediatr 1980; 97:839–841. 157. Posner AC, Prigot A, Konicoff NG. Further observations on the use of tetracycline hydrochloride in prophylaxis and treatment of obstetric infections. In: Antibiotics Annual 1954– 1955. New York: Medical Encyclopedia, 1954–55, pp 594–598. 158. Reyes MP, Ostrea EM, Cabinian AE, et al. Vancomycin during pregnancy: does it cause hearing loss or nephrotoxicity in the infant? Am J Obstet Gynecol 1989; 161:977–981. 159. Pynnonen S, Sillanpaa M. Carbamazepine and mother’s milk. Lancet 1975; 2:563. 160. Kuhnz W, Jager-Roman E, Rating D, et al. Carbamazepine and carbamazepine-10,11-epoxide during pregnancy and postnatal period in epileptic mothers and their nursed infants: pharmacokinetics and clinical effects. Pediatr Pharmacol (New York) 1983; 3:199–208. 161. Froescher W, Eichelbaum M, Niesen M, et al. Carbamazepine levels in breast milk. Ther Drug Monit 1984; 6:266–271. 162. Niebyl JR, Blake DA, Freeman JM, et al. Carbamazepine levels in pregnancy and lactation. Obstet Gynecol 1979; 53:139–140. 163. Pynnonen S, Kanto J, Sillanpaa M, et al. Carbamazepine: placental transport, tissue concentrations in foetus and newborn, and level in milk. Acta Pharmacol Toxicol (Kbh) 1977; 41: 244–253. 164. Kaneko S, Sato T, Suzuki K. The levels of anticonvulsants in breast milk. Br J Clin Pharmacol 1979; 7:624–627. 165. Kaneko S, Suzuki K, Sato T, et al. The problems of antiepileptic medication in the neonatal period: is breast-feeding advisable? In: Janz D, Bossi L, Dam M, et al, eds. Epilepsy, Pregnancy, and the Child. New York: Raven Press, 1982, pp 343–348. 166. Frey B, Schubiger G, Musy JP. Transient cholestatic hepatitis in a neonate associated with carbamazepine exposure during pregnancy and breast-feeding. Eur J Pediatr 1990; 150:136–138. 167. Koup JR, Rose JQ, Cohen ME. Ethosuximide pharmacokinetics in a pregnant patient and her newborn, Epilepsia 1978; 19:535–539. 168. Rane A, Tunell R. Ethosuximide in human milk and in plasma of a mother and her nursed infant. Br J Clin Pharmacol 1981; 12:855–858. 169. Kuhnz W, Koch S, Jakob S, et al. Ethosuximide in epileptic women during pregnancy and lactation period: placental transfer, serum concentrations in nursed infants and clinical status. Br J Clin Pharmacol 1984; 18:671–677. 170. Steen B, Rane A, Lonnerholm G, et al. Phenytoin excretion in human breast milk and plasma levels in nursed infants. Ther Drug Monit 1982; 4:331–334.

226

Taddio and Ito

171. Rane A, Garle M, Borga O, et al. Plasma disappearance of transplacentally transferred diphenylhydantoin in the newborn studied by mass fragmentography. Clin Pharmacol Ther 1974; 15:39–45. 172. Mirkin BL. Diphenylhydantoin: placental transport, fetal localization, neonatal metabolism, and possible teratogenic effects. J Pediatr 1971; 78:329–337. 173. Kuhnz W, Koch S, Helge H, et al. Primidone and phenobarbital during lactation period in epileptic women: total and free drug serum levels in the nursed infants and their effects on neonatal behavior. Dev Pharmacol Ther 1988; 11:147–154. 174. Nau H, Rating D, Hauser I, et al. Placental transfer and pharmacokinetics of primidone and its metabolites phenobarbital, PEMA and hydroxyphenobarbital in neonates and infants of epileptic mothers. Eur J Clin Pharmacol 1980; 18:31–42. 175. Alexander FW. Sodium valproate and pregnancy. Arch Dis Child 1979; 54:240. 176. von Unruh GE, Froescher W, Hoffmann F, et al. Valproic acid in breast milk: how much is really there? Ther Drug Monit 1984; 6:272–276. 177. Dickinson RG, Harland RC, Lynn RK, et al. Transmission of valproic acid (depakene) across the placenta: half-life of the drug in mother and baby. J Pediatr 1979; 94:832–835. 178. Nau H, Rating D, Koch S, et al. Valproic acid and its metabolites: placental transfer, neonatal pharmacokinetics, transfer via mother’s milk and clinical status in neonates of epileptic mothers. J Pharmacol Exp Ther 1981; 219:768–777. 179. Stahl MMS, Neiderud J, Vinge E. Thrombocytopenic purpura and anemia in a breast-fed infant whose mother was treated with valproic acid. J Pediatr 1997; 130:1001–1003. 180. Brixen-Rasmussen L, Halgrener J, Jorgensen A. Amitriptyline and nortriptyline excretion in human breast milk. Psychopharmacology 1982; 76:94–95. 181. Bader TF, Newman K. Amitriptyline in human breast milk and the nursing infant’s serum. Am J Psychiatry 1980; 137:855–856. 182. Schimmel MS, Katz EZ, Shaag Y, et al. Toxic neonatal effects following maternal clomipramine therapy. J Toxicol Clin Toxicol 1991; 29:479–484. 183. Stancer HC, Reed KL. Desipramine and 2-hydroxydesipramine in human breast milk and the nursing infant’s serum. Am J Psychiatry 1986; 143:1597–1600. 184. Kemp J, Ilett KF, Booth J, et al. Excretion of doxepin and n-desmethyldoxepin in human milk. Br J Clin Pharmacol 1985; 20:497–499. 185. Matheson I, Pande H, Alertsen AR. Respiratory depression caused by n-desmethyldoxepin in breastmilk. Lancet 1985; 2:1124. 186. Taddio A, Ito S, Koren G. Excretion of fluoxetine and its metabolite, norfluoxetine, in human breast milk. J Clin Pharmacol 1996; 36:42–47. 187. Isenberg KE. Excretion of fluoxetine in human breast milk. J Clin Psychiatry 1990; 51: 169. 188. Burch KJ, Wells BG. Fluoxetine/norfluoxetine concentrations in human milk. Pediatrics 1992; 89:676–677. 189. Wright S, Dawling S, Ashford JJ. Excretion of fluvoxamine in breast milk. Br J Clin Pharmacol 1991; 31:209. 190. Sovner R, Orsulak PJ. Excretion of imipramine and desipramine in human breast milk. Am J Psychiatry 1979; 136(4A):451–452. 191. Ware MR, DeVane CL. Imipramine treatment of panic disorder during pregnancy. J Clin Psychiatry 1990; 51:482–484. 192. Matheson I, Skjaeraasen J. Milk concentrations of flupenthixol, nortriptyline and zuclopenthixol and between-breast differences in two patients. Eur J Clin Pharmacol 1988; 35:217– 220. 193. Altshuler LL, Burt VK, McMullen M, Hendrick V. Breastfeeding and sertraline: a 24-hour analysis. J Clin Psychiatry 1995; 56:243–245. 194. Verbeeck RK, Ross SG, McKenna EA. Excretion of trazodone in breast milk. Br J Clin Pharmacol 1986; 22:367–370.

Drugs and Breast-Feeding

227

195. Gaillot J, Heusse D, Hougton GW, et al. Pharmacokinetics and metabolism of zopiclone. Pharmacology 1983; 27(suppl):76–91. 196. Matheson I, Sande HA, Gaillot J. The excretion of zopiclone into breast milk. Br J Clin Pharmacol 1990; 30:267–271. 197. Hilbert J, Radwanski E, Affrime MB, et al. Excretion of loratadine in human breast milk. J Clin Pharmacol 1988; 28:234–239. 198. Lucas BD, Purdy CY, Scarim SK, et al. Terfenadine breast milk excretion and pharmacokinetics in lactating women. Pharmacotherapy 1992; 12:506. 199. Lucas BD, Purdy CY, Scarim SK, et al. Terfenadine pharmacokinetics in breast milk in lactating women. Clin Pharmacol Ther 1995; 57:398–402. 200. Findlay JWA, Butz RF, Sailstad JM, et al. Pseudoephedrine and triprolidine in plasma and breast milk of nursing mothers. Br J Clin Pharmacol 1984; 18:901–906. 201. Boutroy MJ, Bianchetti G, Dubruc C, et al. To nurse when receiving acebutolol: is it dangerous for the neonate? Eur J Clin Pharmacol 1986; 30:737–739. 202. White WB, Andreoli JW, Wong SH, et al. Atenolol in human plasma and breast milk. Obstet Gynecol 1984; 63:42S–44S. 203. Liedholm H, Melander A, Bitzen P-O, et al. Accumulation of atenolol and metoprolol in human breast milk. Eur J Clin Pharmacol 1981; 20:229–231. 204. Kulas J, Lunell N-O, Rosing U, et al. Atenolol and metoprolol: a comparison of their excretion into human breast milk. Acta Obstet Gynecol Scand Suppl 1984; 118:65–69. 205. Thorley KJ. Pharmacokinetics of atenolol in pregnancy and lactation. Drugs 1983; 25 (suppl 2):215–218. 206. Thorley KJ, McAinsh J. Levels of the beta-blockers atenolol and propranolol in the breast milk of women treated for hypertension in pregnancy. Biopharm Drug Dispos 1983; 4:299– 301. 207. Devlin RG, Fleiss PM. Captopril in human blood and breast milk. J Clin Pharmacol 1981; 21:110–113. 208. Hartikainen-Sorri A-L, Heikkinen JE, Koivisto M. Pharmacokinetics of cloniding during pregnancy and nursing. Obstet Gynecol 1987; 69:598–600. 209. Okada M, Inoue H, Nakamura Y, et al. Excretion of diltiazem in human milk. N Engl J Med 1985; 312:992–993. 210. Huttunen K, Gronhagen-Riska C, Fyhrquist F. Enalapril treatment of a nursing mother with slightly impaired renal function. Clin Nephrol 1989; 31:278. 211. Redman CWG, Kelly JG, Cooper WD. The excretion of enalapril and enalaprilat in human breast milk. Eur J Clin Pharmacol 1990; 38:99. 212. Liedholm H, Wahlin-Boll E, Hanson A, et al. Transplacental passage and breast milk concentrations of hydralazine. Eur J Clin Pharmacol 1982; 21:417–419. 213. Michael CA. Use of labetalol in the treatment of severe hypertension during pregnancy. Br J Clin Pharmacol 1979; 8:211S–215S. 214. Lunell NO, Kulas J, Rane A. Transfer of labetalol into amniotic fluid and breast milk in lactating women. Eur J Clin Pharmacol 1985; 28:597–599. 215. Hauser GJ, Almog S, Tirosh M, et al. Effect of α-methyldopa excreted in human milk on the breast-fed infant. Helv Paediatr Acta 1985; 40:83–86. 216. Jones HMR, Cummings AJ. A study of the transfer of α-methyldopa to the human foetus and newborn infant. Br J Clin Pharmacol 1978; 6:432–434. 217. White WB, Andreoli JW, Cohn RD. Alpha-methyldopa disposition in mothers with hypertension and in their breast-fed infants. Clin Pharmacol Ther 1985; 37:387–390. 218. Hoskins JA, Holliday SB. Determination of α-methyldopa and methyldopate in human breast milk and plasma by ion-exchange chromatography using electrochemical detection. J Chromatogr 1982; 230:162–167. 219. Sandstrom B, Regardh C-G. Metoprolol excretion into breast milk. Br J Clin Pharmacol 1980; 9:518–519.

228

Taddio and Ito

220. Lindeberg S, Sandstrom B, Lundborg P, et al. Disposition of the adrenergic blocker metoprolol in the late-pregnant woman, the amniotic fluid, the cord blood and the neonate. Acta Obstet Gynecol Scand Suppl 1984; 118:61–64. 221. Valdivieso A, Valdes G, Spiro TE, et al. Minoxidil in breast milk. Ann Intern Med 1985; 102:135. 222. Devlin RG, Duchin KL, Fleiss PM. Nadolol in human serum and breast milk. Br J Clin Pharmacol 1981; 12:393–396. 223. Taddio A, Oskamp M, Ito S, et al. Is nifedipine use during labour and breast-feeding safe for the neonate? Pediatr Res 1996; 39:248A. 224. Ehrenkranz RA, Ackerman BA, Hulse JD. Nifedipine transfer into human milk. J Pediatr 1989; 114:478–480. 225. Penny WJ, Lewis MJ. Nifedipine is excreted in human milk. Eur J Clin Pharmacol 1989; 36:427–428. 226. White WB, Yeh SC, Krol GJ. Nitrendipine in human plasma and breast milk. Eur J Clin Pharmacol 1989; 36:531–534. 227. Santiago TM, Lopez LM. Nitrendipine: a new dihydropyridine calcium-channel antagonist for the treatment of hypertension. DICP 1990; 24:167–175. 228. Sioufi A, Hillion D, Lumbroso P, et al. Oxprenolol placental transfer, plasma concentrations in newborns and passage into breast milk. Br J Clin Pharmacol 1984; 18:453–456. 229. Fidler J, Smith V, DeSwiet M. Excretion of oxprenolol and timolol in breast milk. Br J Obstet Gynaecol 1983; 90:961–965. 230. Smith MT, Livingstone I, Hooper WD, et al. Propranolol, propranolol glucuronide, and naphthoxylactic acid in breast milk and plasma. Ther Drug Monit 1983; 5:87–93. 231. Bauer JH, Pape B, Zajicek J, et al. Propranolol in human plasma and breast milk. Am J Cardiol 1979; 43:860–862. 232. Taylor EA, Turner P. Anti-hypertensive therapy with propranolol during pregnancy and lactation. Postgrad Med J 1981; 57:427–430. 233. O’Hare MF, Murnaghan GA, Russell CJ, et al. Sotalol as a hypotensive agent in pregnancy. Br J Obstet Gynaecol 1980; 87:814–820. 234. Hackett LP, Wojnar-Horton RE, Dusci LJ, et al. Excretion of sotalol in breast milk. Br J Clin Pharmacol 1990; 29:277. 235. Wagner X, Jouglard J, Moulin M, et al. Coadministration of flecainide acetate and sotalol during pregnancy: lack of teratogenic effects, passage across the placenta, and excretion in human breast milk. Am Heart J 1990; 119:700–702. 236. Anderson HJ. Excretion of verapamil in human milk. Eur J Clin Pharmacol 1983; 25:279–280. 237. Miller MR, Withers R, Bhamra R, et al. Verapamil and breast-feeding. Eur J Clin Pharmacol 1986; 30:125–126. 238. Inoue H, Unno N, Ou M-C, et al. Level of verapamil in human milk. Eur J Clin Pharmacol 1984; 26:657–658. 239. Anderson P, Bondesson U, Mattiasson I, et al. Verapamil and norverapamil in plasma and breast milk during breast feeding. Eur J Clin Pharmacol 1987; 31:625–627. 240. Bailey DN, Weibert RT, Naylor AJ, et al. A study of salicylate and caffeine excretion in the breast milk of two nursing mothers. J Anal Toxicol 1982; 6:64–68. 241. Unsworth J, d’Assis-Fonseca A, Beswick DT, et al. Serum salicylate levels in a breast fed infant. Ann Rheum Dis 1987; 46:638–639. 242. Clark JH, Wilson WG. A 16-day-old breast-fed infant with metabolic acidosis caused by salicylate. Clin Pediatr 1981; 20:53–54. 243. Fowler PD. Voltarol: diclofenac sodium. Clin Rheum Dis 1979; 5:427–464. 244. Smith IJ, Hinson JL, Johnson VA, et al. Flurbiprofen in post-partum women: plasma and breast milk disposition. J Clin Pharmacol 1989; 29:174–184. 245. Cox SR, Forbes KK. Excretion of flurbiprofen into breast milk. Pharmacotherapy 1987; 7: 211–215.

Drugs and Breast-Feeding

229

246. Weibert RT, Townsend RJ, Kaiser DG, et al. Lack of ibuprofen secretion into human milk. Clin Pharm 1982; 1:457–458. 247. Townsend RJ, Benedetti TJ, Erickson SH, et al. Excretion of ibuprofen into breast milk. Am J Obstet Gynecol 1984; 149:184–186. 248. Lebedevs TH, Wojnar-Horton RE, Yapp P, et al. Excretion of indomethacin in breast milk. Br J Clin Pharmacol 1991; 32:751–754. 249. Eeg-Olofsson O, Malmros I, Elwin C-E, et al. Convulsions in a breast-fed infant after maternal indomethacin. Lancet 1978; 2:215. 250. Wischnik A, Manth SM, Lloyd J, et al. The excretion of ketorolac tromethamine into breast milk after multiple oral dosing. Eur J Clin Pharmacol 1989; 36:521–524. 251. Buchanan RA, Eaton CJ, Koeff ST, et al. The breast milk excretion of mefenamic acid. Curr Ther Res 1968; 10:592–596. 252. Jamali F, Stevens DRS. Naproxen excretion in milk and its uptake by the infant. DICP 1983; 17:910–911. 253. Ostensen M. Piroxicam in human breast milk. Eur J Clin Pharmacol 1983; 25:829–830. 254. Ostensen M, Matheson I, Laufen H. Piroxicam in breast milk after long-term treatment. Eur J Clin Pharmacol 1988; 35:567–569. 255. Wiles DH, Orr MW, Kolakowska T. Chlorpromazine levels in plasma and milk of nursing mothers. Br J Clin Pharmacol 1978; 5:272–273. 256. Stewart RB, Karas B, Springer PK. Haloperidol excretion in human milk. Am J Psychiatry 1980; 137:849–850. 257. Whalley LJ, Blain PG, Prime JK. Haloperidol secreted in breast milk. BMJ 1981; 282: 1746–1747. 258. Ohkubo T, Shimoyama R, Sugawara K. Measurement of haloperidol in human breast milk by high-performance liquid chromatography. J Pharmaceut Sci 1992; 81:947–949. 259. Olesen OV, Bartels U, Poulsen JH. Perphenazine in breast milk and serum. Am J Psychiatry 1990; 147:1378–1379. 260. Tunnessen WW, Hertz CG. Toxic effects of lithium in newborn infants: a commentary. J Pediatr 1972; 81:804–807. 261. Schou M, Amdisen A. Lithium and pregnancy—III: Lithium ingestion by children breastfed by women on lithium treatment. BMJ 1973; 2:138. 262. Sykes PA, Quarrie J, Alexander FW. Lithium carbonate and breast-feeding. BMJ 1976; 4: 1299. 263. Weinstein MR, Goldfield M. Lithium carbonate treatment during pregnancy: report of a case. Dis Nerv Syst 1969; 30:828–832. 264. Rylance GW, Woods CG, Donnelly MC, et al. Carbimazole and breastfeeding. Lancet 1987; 1:928. 265. Tegler L, Lindstrom B. Antithyroid drugs in milk. Lancet 1980; 2:591. 266. Johansen K, Anderson AN, Kampmann JP, et al. Excretion of methimazole in human milk. Eur J Clin Pharmacol 1982; 23:339–341. 267. Kampmann JP, Johansen K, Hansen JM, et al. Propylthiouracil in human milk. revision of a dogma. Lancet 1980; 1:736–737. 268. Low LCK. Lang J, Alexander WD. Excretion of carbimazole and propylthiouracil in breast milk. Lancet 1979; 2:1011. 269. Anderson PO, McGuire GG. Neonatal alprazolam withdrawal-possible effects of breast feeding. DICP 1989; 23:614. 270. Oo CY, Kuhn RJ, Desai N, et al. Pharmacokinetics in lactating women: prediction of alprazolam transfer into milk. Br J Clin Pharmacol 1995; 40:231–236. 271. Soderman P, Matheson I. Clonazepam in breast milk. Eur J Pediatr 1988; 147:212–213. 272. Fisher JB, Edgren BE, Mammel MC, et al. Neonatal apnea associated with maternal clonazepam therapy: a case report. Obstet Gynecol 1985; 66:34S–35S. 273. Dusci LJ, Good SM, Hall RW, et al. Excretion of diazepam and its metabolites in human

230

274. 275. 276. 277. 278. 279. 280. 281. 282. 283. 284. 285. 286. 287. 288. 289. 290. 291. 292. 293. 294. 295. 296. 297. 298. 299.

Taddio and Ito milk during withdrawal from combination high dose diazepam and oxazepam. Br J Clin Pharmacol 1990; 29:123–126. Wesson DR, Camber S, Harkey M, et al. Diazepam and desmethyldiazepam in breastmilk. J Psychoactive Drugs 1985; 17:55–56. Brandt R. Passage of diazepam and desmethyldiazepam into breast milk. Arzneimittelforschung 1976; 26:454–7. Patrick MJ, Tilstone WJ, Reavey P. Diazepam and breast-feeding. Lancet 1972; 1:542–543. Erkkola R, Kanto, J. Diazepam and breast-feeding. Lancet 1972; 1:1235–1236. Whitelaw AGL, Cummings AJ, McFadyen IR. Effect of maternal lorazepam on the neonate. BMJ 1981; 282:1106–1108. Matheson I, Lunde PKM, Bredesen JE. Midazolam and nitrazepam in the maternity ward: milk concentrations and clinical effects. Br J Clin Pharmacol 1990; 30:787–793. Wretlind M. Excretion of oxazepam in breast milk. Eur J Clin Pharmacol 1987; 33:209– 210. Lebedevs TH, Wojnar-Horton RE, Yapp P, et al. Excretion of temazepam in breast milk. Br J Clin Pharmacol 1992; 33:204–206. McKenna WJ, Harris L, Rowland E, et al. Amiodarone therapy during pregnancy. Am J Cardiol 1983; 51:1231–1233. Pitcher D, Leather HM, Storey GCA, et al. Amiodarone in pregnancy. Lancet 1983; 1:597– 598. Strunge P, Frandsen J, Andreasen F. Amiodarone during pregnancy. Eur Heart J 1988; 9: 106–109. Plomp TA, Vulsma T, de Vijlder JJM. Use of amiodarone during pregnancy. Eur J Obstet Gynecol Reprod Biol 1992; 43:201–207. Gutgesell M, Overholt E, Boyle R. Oral bretylium tosylate use during pregnancy and subsequent breastfeeding: a case report. Am J Perinatol 1990; 7:144–145. Loughnan PM. Digoxin excretion in human breast milk. J Pediatr 1978; 92:1019–1020. Chan V, Tse TF, Wong V. Transfer of digoxin across the placenta and into breast milk. Br J Obstet Gynaecol 1978; 85:605–609. Finley JP, Waxman MB, Wong PY, et al. Digoxin excretion in human milk. J Pediatr 1979; 94:339–340. Reinhardt D, Richter O, Genz T, et al. Kinetics of the translactal passage of digoxin from breast feeding mothers to their infants. Eur J Pediatr 1982; 138:49–52. Barnett DB, Hudson SA, McBurney A. Disopyramide and its n-monodesalkyl metabolite in breast milk. Br J Clin Pharmacol 1982; 14:310–312. MacIntosh O, Buchanan N. Excretion of disopyramide in human breast milk. Br J Clin Pharmacol 1985; 19:856–857. Hoppu K, Neuvonen PJ, Korte T. Disopyramide and breast feeding. Br J Clin Pharmacol 1986; 21:553. McQuinn RL, Pisani A, Wafa S, et al. Flecainide excretion in human breast milk. Clin Pharmacol Ther 1990; 48:262–267. Timmis AD, Jackson G, Holt DW. Mexiletine for control of ventricular dysrhythmias in pregnancy. Lancet 1980; 2:647–648. Lewis AM, Johnston A, Patel L, et al. Mexiletine in human blood and breast milk. Postgrad Med J 1981; 57:546–547. Pittard WB, Glazier H. Procainamide excretion in human milk. J Pediatr 1983; 102:631– 633. Libardoni M, Piovan D, Busato E, et al. Transfer of propafenone and 5-OH-propafenone to foetal plasma and maternal milk. Br J Clin Pharmacol 1991; 32:527–528. Hill LM, Malkasian GD. The use of quinidine sulfate throughout pregnancy. Obstet Gynecol 1979; 54:366–368.

Drugs and Breast-Feeding

231

300. Soderman P, Hartvig P, Fagerlund C. Acetazolamide excretion into human breast milk. Br J Clin Pharmacol 1984; 17:599–600. 301. Werthmann MW, Krees SV. Excretion of chlorothiazide in human breast milk. J Pediatr 1972; 81:781–783. 302. Mulley BA, Parr GD, Pau WK, et al. Placental transfer of chlorthalidone and its elimination in maternal milk. Eur J Clin Pharmacol 1978; 13:129–131. 303. Cominos DC, Van der Walt A, Van Rooyen AJL. Suppression of postpartum lactation with furosemide. S Afr Med J 1976; 50:251–252. 304. Miller ME, Cohn RD, Burghart PH. Hydrochlorothiazide disposition in a mother and her breast-fed infant. J Pediatr 1982; 101:789–791. 305. Phelps DL, Karim A. Spironolactone: relationship between concentrations of dethioacetylated metabolite in human serum and milk. J Pharm Sci 1977; 66:1203. 306. Somogyi A, Gugler R. Cimetidine excretion into breast milk. Br J Clin Pharmacol 1979; 7: 627–629. 307. Oo CY, Kuhn RJ, Desai N, McNamara PJ. Active transport of cimetidine into human milk. Clin Pharmacol Ther 1995; 58:548–555. 308. Hofmeyr GJ, Sonnendecker EWW. Secretion of the gastrokinetic agent cisapride in human milk. Eur J Clin Pharmacol 1986; 30:735–736. 309. Hofmeyr GJ, van Iddekinge B. Domperidone and lactation. Lancet 1983; 1:647. 310. Hofmeyr GJ, van Iddekinge B, Blott JA. Domperidone: secretion in breast milk and effect on puerperal prolactin levels. Br J Obstet Gynaecol 1985; 92:141–144. 311. Courtney TP, Shaw RW, Cedar E, et al. Excretion of famotidine in breast milk. Br J Clin Pharmacol 1988; 26:639P. 312. Nikodem VC, Hofmeyr GJ. Secretion of the antidiarrhoeal agent loperamide oxide in breast milk. Eur J Clin Pharmacol 1992; 42:695–696. 313. Kauppila A, Arvela P, Koivisto M, et al. Metoclopramide and breast-feeding: transfer into milk and the newborn. Eur J Clin Pharmacol 1983; 25:819–823. 314. Lewis PJ, Devenish C, Kahn C. Controlled trial of metoclopramide in the initiation of breast feeding. Br J Clin Pharmacol 1980; 9:217–219. 315. Obermeyer BD, Bergstrom RF, Callaghan JT, et al. Secretion of nizatidine into human breast milk after single and multiple doses. Clin Pharmacol Ther 1990; 47:724–730. 316. Riley AJ, Crowley P, Harrison C. Transfer of ranitidine to biological fluids: milk and semen. In: Misiewicz JJ, Wormsley KG, eds. The Clinical Use of Ranitidine. Oxford, England: Medicine Publishing Foundation, 1981, pp 78–81. 317. Kearns GL, McConnell RF, Trang JM, et al. Appearance of ranitidine in breast milk following multiple dosing. Clin Pharm 1985; 4:322–324. 318. Nelis GF. Diarrhea due to 5-aminosalicylic acid in breast milk. Lancet 1989; 1:383. 319. Jenss H, Weber P, Hartmann F. 5-aminosalicylic acid and its metabolite in breast milk during lactation. Am J Gastroenterol 1990; 85:331. 320. Klotz U, Harings-Kaim A. Negligible excretion of 5-aminosalicylic acid in breast milk. Lancet 1993; 342:618–619. 321. Edstein MD, Veenendaal JR, Newman K, et al. Excretion of chloroquine, dapsone and pyrimethamine in human milk. Br J Clin Pharmacol 1986; 22:733–735. 322. Ogunbona FA, Onyeji CO, Bolaji OO, et al. Excretion of chloroquine and desethylchloroquine in human milk. Br J Clin Pharmacol 1987; 23:473–476. 323. Ette EI, Essien EE, Ogonor JI, et al. Chloroquine in human milk. J Clin Pharmacol 1987; 27:499–502. 324. Nation RL, Hackett LP, Dusci LJ, et al. Excretion of hydroxychloroquine in human milk. Br J Clin Pharmacol 1984; 17:368–369. 325. Ostensen M, Brown ND, Chiang PK, et al. Hydroxychloroquine in human breast milk. Eur J Clin Pharmacol 1985; 28:357.

232

Taddio and Ito

326. Erkkola R, Kanto J, Allonen H, et al. Excretion of methylergometrine (methylergonovine) into the human breast milk. Int J Clin Pharmacol Biopharm 1978; 16:579–580. 327. Ost L, Wettrell G, Bjorkhem I, et al. Prednisolone excretion in human milk. J Pediatr 1985; 106:1008–1011. 328. Katz FH, Duncan BR. Entry of prednisone into human milk. N Engl J Med 1975; 293:1154. 329. McKenzie SA, Selley, JA, Agnew JE. Secretion of prednisolone into breast milk. Arch Dis Child 1975; 50:894–896. 330. Jarnerot G, Into-Malmberg M-B. Sulphasalazine treatment during breast feeding. Scand J Gastroenterol 1979; 14:869–871. 331. Berlin CM, Yaffe SJ. Disposition of salicylazosulfapyridine (azulfidine) and metabolites in human breast milk. Dev Pharmacol Ther 1980; 1:31–39. 332. Esbjorner E, Jarnerot G, Wranne L. Sulphasalazine and sulphapyridine serum levels in children to mothers treated with sulphasalazine during pregnancy and lactation. Acta Paediatr Scand 1987; 76:137–142. 333. Branski D, Kerem E, Gross-Kieselstein E, et al. Bloody diarrhea-a possible complication of sulfasalazine transferred through human breast milk. J Pediatr Gastroenterol Nutr 1986; 5: 316–317. 334. Wojnar-Horton RE, Hackett LP, Yapp P, et al. Distribution and excretion of sumatriptan in human milk. Br J Clin Pharmacol 1996; 41:217–221. 335. Stec GP, Greenberger P, Ruo TI, et al. Kinetics of theophylline transfer to breast milk. Clin Pharmacol Ther 1980; 28:404–408. 336. Yurchak AM, Jusko WJ. Theophylline secretion into breast milk. Pediatrics 1976; 57:518– 520. 337. Orme M L’E, Lewis PJ, de Swiet M, et al. May mothers given warfarin breast-feed their infants? BMJ 1977; 1:1564–1565. 338. McKenna R, Cole ER, Vasan U. Is warfarin sodium contraindicated in the lactating mother? J Pedatr 1983; 103:325–327. 339. Krogh CME, ed. Compendium of Pharmaceuticals and Specialties. 31 st ed. Ottawa: Canadian Pharmaceutical Association, 1996. 340. McElvoy GK, Litvak K, Welsh OH, et al, eds. American Hospital Formulary Service Drug Information. Bethesda, MD: American Society of Hospital Pharmacists, 1992.

14 Poisoning in Pregnancy Milton Tenenbein University of Manitoba, Winnipeg, Manitoba, Canada

Clinical Case You attend a 16-year-old pregnant (12 weeks gestation) woman who had tried to commit suicide by taking 60 caplets of Tylenol (325 mg each). The patient, who weighs 50 kg, did not tell anyone about her action for 24 hours, but then became frightened by her continued vomiting. Her boyfriend did not suspect anything because ‘‘she had morning sickness anyway.’’ How should she be treated?

INTRODUCTION Exposure to drugs and chemicals can have an adverse effect on the outcome of a pregnancy. We generally consider low-level exposures (doses insufficient to harm the expectant mother), which we evaluate by the criteria of reproductive wastage and production of dysmorphic offspring. Very little attention has been paid to the issue of acute poisoning during pregnancy. In such situations there is potential risk for the mother as well as the fetus. The fetal and maternal risks are not necessarily equal. Extent of toxin exposure may be different on both sides of the placenta. Although most agents that are absorbed across the gastrointestinal epithelium freely traverse the placenta, some, such as iron, have specialized transplacental absorptive mechanisms (1,2). In a massive overdose situation, this mechanism may become a rate-limiting step, thus resulting in a relatively smaller fetal exposure. Conversely, some agents, such as salicylates, are present in higher concentrations in the fetus (3,4). Fetal metabolic pathways are often immature. While this may seem to put the fetus at greater risk, it can also be protective when, for example, toxicity is due to an intracellularly generated metabolite, not the ingestant itself. Acetaminophen is an example of such an agent. Because the mother is at risk, she often requires treatment. Does the presence of a conceptus demand modified therapy? Do interventions to manage poisoning—such as syrup of ipecac, hemodialysis, or specific antidotes—present a risk to the fetus? In some

233

234

Tenenbein

cases, a unique decision in the management of the acutely poisoned patient may be required: should a potentially viable fetus be delivered on an emergency basis to prevent damage or death? This chapter reviews the management of the poisoned expectant mother. Adverse effects to the neonate from drugs administered in therapeutic doses to the mother are discussed elsewhere in this book, as are teratogenesis and the fetal and neonatal effects of substance abuse. First, the epidemiology of acute poisonings during pregnancy is reviewed. Then we discuss the general management of the overdosed expectant mother and the management of various specific poisonings.

EPIDEMIOLOGY OF POISONING DURING PREGNANCY Most poisonings during pregnancy are suicidal gestures. However, in one large series of hospitalized poisoned pregnant patients, 14% of the cases were accidental (5). In this series of 162 patients, there were two maternal and four fetal deaths. This cohort, which was gleaned from a specialized unit for the treatment of intoxications, consisted of relatively serious cases. The occurrence of only four fetal deaths, a 2.5% rate, would seem to indicate a relative resistance to an acute toxic insult. If indeed this is so, it may help to explain the limited number of published cases of overdoses during pregnancy, since most patients do well. In a subsequent series by the same authors (6), 2 out of 559 women died. Of 213 fetuses in the first month of pregnancy, 111 ended in early loss, 3 in clinical miscarriage, 12 survived to delivery, 73 pregnancies were terminated, and one woman died. Pregnant women may threaten suicide to strengthen their case for a therapeutic abortion (7). Conversely, pregnancy provides a unique motivation for drug overdose (i.e., an attempt to induce an abortion). In Czeizel’s series, 8% overdosed for the express purpose of inducing an abortion, and 23% subsequently had a therapeutic abortion (5). Some agents are commonly considered as abortifacients. If overdose of quinine during pregnancy is encountered, one should strongly consider the possibility that an attempt was made to induce abortion (8). Historically, lead was often ingested to induce abortions (9), and from time to time various herbal remedies have been utilized for this purpose (10). Whitlock and Edwards, who reviewed pregnancy and attempted suicide (11), felt that the incidence of suicidal gestures among pregnant women was at least equal to that among nonpregnant women. These authors reported that 7% of all women making suicidal gestures are likely to be pregnant. Although pregnant women are less likely to kill themselves, suicide does account for 1% of all deaths during pregnancy, and pregnancy is an associated factor in 5% of all female suicides (11). In a later study by the same group, they reported that pregnant women trying to commit a suicide through an overdose tended to be of lower socioeconomic class than controls and were more likely to be unmarried (12). Rayburn et al. reviewed drug overdose during pregnancy in an ambulatory population (13). Their cohort was gleaned from the case records of a poison control center. In a 4-year period, 0.07% of all telephone consultations involved overdoses by pregnant women. Although the ingestants were similar to those of the general population, one notable difference was an increased frequency of ingestion of vitamins and iron, which constituted the second most common group of ingestants. This should not be surprising, since pregnant women are routinely placed on these supplements. However, it is worrisome because the management of iron overdose is complex enough without the presence of a fetus.

Poisoning in Pregnancy

235

Czeizel studied drug overdose and pregnancy from another perspective (14). He wondered whether the prior occurrence of an overdose adversely affects subsequent reproductive function. Happily he found that a severe overdose does not lead to decreased fertility, increased pregnancy wastage, or subsequent dysmorphology (14).

APPROACH TO THE POISONED PREGNANT PATIENT At the time of first contact with the poisoned woman, the fact that she is pregnant may not be disclosed by her or may not even known to her. Jones and colleagues describe an overdosed patient in whom ruptured ectopic pregnancy was missed for significant time because of the patient’s normal menstrual history. These authors very logically recommend that a pregnancy test should be performed in all women of reproductive age who present with poisoning or drug overdose (15). The general management of the pregnant patient who has taken an acute overdose should not differ from that of the nonpregnant individual. The proper approach is well described elsewhere (16) and includes acute stabilization (airway, breathing, circulation), history, physical examination, supportive care, nonspecific antipoison therapy (prevention of absorption, enhancement of elimination), and specific antipoison therapy (administration of an antidote). In addition to such routine management, fetal well-being must be monitored. Recently, the approach to preventing the absorption of ingested poisons has undergone reappraisal and modification. The time-honored interventions of ipecac-induced emesis and orogastric lavage have been questioned and administration of activated charcoal has become the primary gastrointestinal decontamination procedure (17–19). In addition to this change in the approach to the prevention of absorption of poisons, an additional procedure, whole-bowel irrigation with polyethylene glycol-electrolyte lavage solution, has been introduced as a new intervention (20). Its use in a pregnant patient has been reported (21). From the perspective of the management of the overdosed pregnant patient, the trend away from syrup of ipecac circumvents potential concerns regarding the teratogenic potential and safety of this drug during pregnancy. Unlike ipecac, activated charcoal is not absorbed; thus adverse effects on the fetus would not be expected. The risk to the fetus from the administration of most antidotes is unknown. The scant available information is discussed below in connection with the poison in question. These risks may manifest as teratogenesis or as acute fetal toxicity if the antidote has agonistic properties. An example of the latter situation is the administration of atropine for organophosphate pesticide poisoning. Since the transplacental delivery of the poison and the antidote may not be similar, although the mother’s atropine dose may be optimal, the fetus is at risk to receive a relative underdose or overdose of this antidote. The latter situation has been documented (22). Despite these potential fetal risks, the needs of the mother are paramount. If the indication for an antidote exists, the presence of a gravid uterus must not be considered to contraindicate its administration. A tragic case of a maternal death associated with the withholding of an antidote because of fears of teratogenicity has been published (23,24). In a prospective study, Czeizel reported that, after excluding heavy drinking resulting in fetal alcohol syndrome, the overall rate of congenital anomalies (9%) was not higher than among controls (25). Although there is very limited literature on most forms of therapy for poisoning during pregnancy, some experiences with hemodialysis have been published. There are

236

Tenenbein

at least three reports of the use of acute hemodialysis during pregnancy to treat acute drug overdose (26–28) as well as reviews of pregnancy in hemodialysis patients (29–31). Acute hemodialysis does not seem to have an adverse effect on the fetus. Although women on chronic hemodialysis have increased reproductive wastage, premature labors, and growthretarded newborns, these are more likely a consequence of their underlying renal disease.

ACETAMINOPHEN Published Experience After nutritional supplements, acetaminophen is the commonest drug taken by pregnant women (32). It has been documented as the drug that is most frequently taken in overdose during pregnancy (13), and the published experience of overdoses during pregnancy is greatest for this agent (Table 1) (33–43). Since acetaminophen crosses the placenta (44), the fetus is at potential risk. The first case was published in 1978 but was poorly documented. The overdose occurred during the first half of pregnancy and was not treated with specific antidote; the mother suffered severe hepatotoxicity: serum aspartate transaminase (AST) of 9550 IU/ L. The fetus survived, only to be therapeutically aborted 2 weeks later. Since there was no description of a pathological examination of the abortus, it is not known whether there was any fetal toxicity. Between 1982 and 1986, eight case reports were found. Byer et al. (34) described a 26-year-old, 36-week-pregnant woman who ingested 32.5 g of acetaminophen. Some 41/2 hours after the ingestion, her serum concentration was just barely into the potential toxicity section of the nomogram, than expected (16–24 hours after overdose), the hepatic microscopy supported acetaminophen toxicity. Thus, toxicity must be considered to be the probable cause for the fetal death. Two cases were described in 1986. Robertson et al. (40) described a 21-year-old 16-week-pregnant woman who ingested 36 g of acetaminophen. Her serum concentration was well into the toxic range of the nomogram. Because oral N-acetylcysteine was not tolerated, this antidote was administered intravenously. There was no maternal hepatotoxicity, and a normal newborn was delivered at term. The second patient, described by Ludmir et al. (41), was 16 weeks pregnant and had ingested 64 g of acetaminophen. Her serum concentrations were also well into the toxic range, and N-acetylcysteine therapy was begun at 20 hours after the overdose, at a time when it was unlikely to alter the course. This patient developed severe hepatotoxicity, from which she fully recovered. A premature labor and delivery resulted in a normal 32-week newborn that subsequently thrived. The largest experience of acetaminophen overdose during pregnancy was published in 1989 (42). In this multicenter study, complete data were available in 60 of 113 cases, and one of these is of particular interest: a stillborn fetus at 33 weeks gestation, with death having occurred 2 days after maternal overdose. The mother’s serum acetaminophen concentration was toxic by nomogram and she received a full oral course of N-acetylcysteine beginning 12 hours after acetaminophen ingestion. Significant hepatotoxicity occurred in the mother, who survived. The fetal serum acetaminophen concentration was grossly elevated, and at autopsy the fetus was found to have massive hepatonecrosis. This case is very similar to that of Haibach et al. (39), strengthening the hypothesis that the fetus is at greatest risk if the acetaminophen overdose occurs during the third trimester.

Poisoning in Pregnancy

Table 1

Acetaminophen Overdose During Pregnancy

Case

Gestation (weeks)

Potential maternal toxicity

1

20

Yes

No

Yes

na

2 3 4 5 6 7 8 9 10

36 29 38 18 36 28 16 16 33

Yes Yes Yes Yes No Yes Yes Yes Yes

Yes No Yes Yes, but late Yes No Yes Yes, but late Yes

No Yes No Yes No Yes No Yes Yes

6 weeks 16 h 17 h 23 weeks 7h na 24 weeks 16 weeks na

Full course of antidote

Maternal toxicity

Timing of delivery after overdose

Fetal outcome

Ref.

Survival; subsequent therapeutic abortion Normal Survival; no hepatotoxicity Survival; no hepatotoxicity Normal Survival; no hepatotoxicity Stillborn Normal Normal Stillborn

29 30 31 32 33 34 35 36 37 38

Note: na, not available.

237

238

Tenenbein

Riggs et al. (42) also attributed a first-trimester fetal loss to an acetaminophen overdose. However, the mother was also very seriously affected and she died. It was not determined whether the fetal loss was secondary to the mother’s critical condition or primary to fetal hepatotoxicity. Another large series describing 48 cases was published in 1990 (43). Details of individual cases are sketchy, making commentary difficult. Of interest, though, was the observation of no birth defects associated with first-trimester N-acetylcysteine therapy. Of the 10 cases summarized in Table 1, case 6 is of little interest because of negligible potential for toxicity. Cases, 1, 3, 5, 7, 9, and 10 all exhibited maternal hepatotoxicity. The two intrauterine deaths, cases 7 and 10, were due to third-trimester overdoses. The other four fetuses did well. Of these, only one was in the third trimester (case 3), and this baby was delivered 6 hours after the overdose. Thus, it would seem that maternal hepatotoxicity in the third trimester is a marker for potential fetal demise. However, the clinical usefulness of this association (if it exists at all) is quite limited, since both fetuses who died did so prior to or during the early development of maternal toxicity. Pharmacology and Toxicology Estimation of the fetal risk from a maternal acetaminophen overdose requires an understanding of the metabolism of this drug by the mature organism and the differences that may exist in the fetus. We discuss only briefly the metabolism of this drug, since it is well reviewed elsewhere (45). In therapeutic amounts, acetaminophen is largely excreted as urinary sulfates or glucuronides. A small amount is oxidized by the cytochrome P-450 mixed-function oxidase system. This produces a highly reactive metabolite that binds to hepatocellular macromolecules, producing hepatoxicity. This effect can be prevented by complexation with hepatic glutathione. In overdose situations, sulfation and glucuronidation become saturated, thus presenting an increased load to the cytochrome P-450 pathway. Hepatotoxicity ensues after glutathione has been depleted. Thus it is an intracellularly generated metabolite, not the parent acetaminophen, that produces toxicity. Administration of N-acetylcysteine is protective because it acts as a glutathione precursor (46). Therefore, for hepatotoxicity to occur, the fetal hepatocyte must have an active cytochrome P-450 system. Absent or decreased capacity for sulfation, glucuronidation, and glutathione generation would increase the risk of fetal liver damage. In general, cytochrome P-450, sulfation, and glutathione are present in human fetal livers but glucuronidation is not (47,48). Not surprisingly, the extent of the activities of these processes is poorly documented. It varies among xenobiotics and with gestational age. However, acetaminophen is one of the few drugs that has undergone human fetal metabolic studies (45). Hepatocytes were harvested from fetuses at 18–23 weeks gestation. Cytochrome P-450 activity, glutathione generation, and sulfation were demonstrated along with an absence of glucuronidation. Mean cytochrome P-450 activity was only 10% of adult values, with a linear increase occurring over the gestational period under study. Thus, degree of fetal risk from maternal acetaminophen ingestion would seem to correlate with gestational age. This suggestion is supported by the observation that the two fetal deaths (39,42) were third-trimester gestations. Treatment of Acetaminophen Overdose During Pregnancy The management of the pregnant woman with an acetaminophen overdose should not differ from that of the nonpregnant individual—that is, stabilization, supportive care, ap-

Poisoning in Pregnancy

239

propriate gastrointestinal decontamination, and specific therapy (N-acetylcysteine administration), ideally based on serum acetaminophen concentration determination. A recent study documents that N-acetylcysteine crosses the human placenta and achieves fetal concentrations well within those associated with antidote effects (9.4 ⫾ 1.3 µg/mL). There were no adverse effects among these babies (49). The safety of this antidote for the fetus has not been established, but preliminary information is encouraging (42). In any event, N-acetylcysteine should not be withheld if the mother is at risk, but it has been shown to have negligible transplacental passage in sheep (51). Since oral administration in humans produces significantly lower plasma concentrations (51), the intravenous route in the mother seems to be the logical choice for protection of her fetus. Also, the fetal delivery of N-acetylcysteine may be compromised after maternal oral administration because of first-pass hepatic uptake, particularly if the maternal liver is in a relatively glutathionedepleted state owing to the acetaminophen overdose. Furthermore, the oral route is often poorly tolerated because of vomiting (52). Nevertheless, the findings of Selden et al. (51) point to a poor fetal prognosis in significant third-trimester acetaminophen overdose. The limited data available support immediate delivery of a mature fetus for direct extrauterine N-acetylcysteine therapy if the maternal serum acetaminophen concentration is well into the toxic range by nomogram. If the risk for delivery is too great because of fetal immaturity, direct intrauterine cannulation of the umbilical vasculature for N-acetylcysteine administration could be considered. However, neither of these options is likely to be practical because of the length of time required for their organization and implementation.

SALICYLATES Published Experience Five reports of in utero salicylate intoxication in pregnant women were found. Three were due to an acute ingestion (53–55), whereas the other two were the result of subacute or chronic toxicity (51,52). There were two deaths, both in the acute group (48,50). Jackson (48), in 1948, described a woman in her eighth month of pregnancy who delivered a 2.7kg stillborn fetus 17 hours after an overdose of 200 g of acetylsalicylic acid. Although at autopsy tentorial tears and cerebral hemorrhage were found, there was ‘‘a high concentration of salicylate’’ in the cord blood, indicating at least that the fetus was alive prior to the overdose. Nevertheless, the contribution of salicylate toxicity to the fetal demise is unclear. Also, given the mother’s benign course, it is unlikely that she actually ingested 200 g of acetylsalicylic acid. In 1961 Earle (54) described a 3.5-kg neonate whose mother had ingested 15–18 g of acetylsalicylic acid 27 hours prior to delivery. At 20 hours of age, the baby was acidemic, with a salicylate concentration of 350 mg/L. One hour earlier, the mother’s concentrations had been 220 mg/L; it had been 380 mg/L 20 hours prior to delivery. An exchange transfusion resulted in removal of an estimated 135 mg of salicylate. Since this represents only 40 mg/kg and is equivalent to two-fifths of a tablet, it is difficult to ascribe benefit to this intervention. Mother and baby did well. Rejent and Baik described a woman in her eighth month of pregnancy who ingested 36.5 g of acetylsalicylic acid (55). She was manifesting salicylism, but the fetus was not felt to be in distress. However, 20 hours later, the baby was dead. The postmortem fetal serum salicylate concentration was 243 mg/L, with a brain concentration of 200 µg/g.

240

Tenenbein

Since the latter amount is similar to fatal brain concentrations in mice (56), it is likely that the fetal death was due to intrauterine salicylate toxicity. Lynd et al. (57) described a case of chronic intrauterine salicylism in a 2.6-kg male neonate who was irritable and feeding poorly at 36 hours of age. It was learned that his mother had self-medicated with undetermined amounts of salicylates throughout her ninth month. The baby was hyperpneic and hypertonic, and he appeared malnourished. His serum salicylate level was 310 mg/L at 36 hours. A concentration of 383 mg/L was found upon analysis of the cord blood. The baby was treated with induced diuresis and his condition improved. His irritability, hypertonia, poor feeding, and intrauterine malnutrition may have been due to the salicylism, but other causes such as maternal drug abuse were not conclusively ruled out. Ahlfors et al. (58) described a 3-kg 37-week female neonate delivered by cesarean section for fetal distress. Her mother had consumed 3 g of acetylsalicylic acid daily for a flulike illness for several days prior to delivery. Shortly after birth, tachypnea, a compensated metabolic acidosis, and a cord blood concentration of 473 mg/L were documented. An exchange transfusion at 21 hours had a negligible effect on the serum salicylate concentration, but the neonate gradually recovered. An interesting observation was that the salicylate displaced bilirubin from its albumin bindings sites, thus necessitating interventions to lower bilirubin concentrations so as to prevent kernicterus. Buck and colleagues (59) described a neonate in metabolic acidosis, tachypnea, and hypoglycemia born to a mother who had taken aspirin throughout pregnancy. The authors highlight the similarities between symptoms of neonatal sepsis and those of a toxic reaction to salicylates. The toxic effects of salicylates on the mother are well described elsewhere (60). Because lungs do not function in the fetus, stimulation of respiration is of no concern. However, the ability of salicylates to uncouple oxidative phosphorylation makes them a general cellular poison, with the brain in particular being a target organ because of its inability to compensate with anaerobic energy production. Hill’s work with mice supports central nervous system toxicity as the mechanism of death in salicylism (56). Unlike the case with acetaminophen, in salicylate poisoning the parent compound rather than the metabolite produces the toxicity, thus placing the fetus at risk. Several factors would seem to support the hypothesis that there may be a greater risk to the fetus than to the mother. Salicylate traverses the placenta and is found in higher concentrations in the fetus (3,4). The fetus has a lower arterial pH than the adult, making the blood-tointracellular pH gradient less. Since the drug is a weak acid, this lower gradient favors a relatively greater proportion of the fetal salicylate load entering the central nervous system. In addition, the fetus has less capacity to buffer the acidemic stress imposed by the salicylate and, relative to the mother, a reduced capacity to metabolize and excrete this toxin. Indeed, in at least three cases, paired newborn-maternal sera demonstrated salicylate persistence in the neonate (54,55,57). Treatment of Salicylate Overdose During Pregnancy The treatment of salicylate poisoning is supportive care (60). This includes patient stabilization; appropriate gastrointestinal decontamination; administration of fluid, electrolyte, and glucose; and—when indicated—an extracorporeal removal intervention. In pregnancy, it is hoped that the positive effects of these maneuvers will be reflected transplacentally. However, several of the previously described fetal factors (higher serum concentra-

Poisoning in Pregnancy

241

tions, larger proportion of salicylate in the brain, lower buffering capacity, and decreased salicylate metabolism) would somewhat negate such benefits. Therefore, when the fetus is potentially viable ex utero, consideration should be given to prompt delivery. This provides the opportunity for direct provision of care to the newborn.

IRON Published Experience Prenatal vitamins and iron have been documented as the second most common overdosed drug group during pregnancy (13). A series of 49 cases from the United Kingdom has been published (61). However, the vast majority were either nontoxic or negligibly toxic overdoses, and the retrospective nature of the review makes analysis difficult. We have managed five cases, and there are six other case reports in the literature (23,24,62–66). The most notable observation is that the fetus seems to fare better than the mother. The literature reports one overdose in the first trimester (62), two in the second (23,24,63), and three during the third (58–60). One case was nontoxic by serum iron concentration and deferoxamine challenge criteria (63). The only fetal death occurred as a spontaneous abortion in a 17-year-old girl who was seriously ill and subsequently died of the overdose (23,24). The other five babies did well, although two maternal deaths (65,66) occurred shortly after the delivery of neonates who showed no signs of iron toxicity. Three mothers were treated with deferoxamine (one in the second trimester and two in the third trimester), and these babies were normal. Two other cases occurred prior to deferoxamine availability and were treated with other chelators no longer used in iron poisoning (62,65). Both these babies did well. Three pregnancies were delivered within 36 hours of overdose, and the cord blood or early postpartum serum iron concentrations were in the normal range (64–66). The fetus that died and its placenta were examined histologically for evidence of iron toxicity and for the presence of increased iron. Neither was found (23,24). This is also true for the only other placenta that was examined (64). The pathophysiology of acute iron poisoning is well reviewed elsewhere (67,68). This can be a difficult poisoning to treat, and several aspects of its therapy are controversial (67). Major features of iron poisoning include gastrointestinal hemorrhage, shock, acidosis, hepatic failure, and coagulopathy. Death is usually the result of cardiovascular collapse or hepatic failure. Although the introduction of the specific iron chelator deferoxamine improved the prognosis, optimal use of this antidote has not been established. The well-being of a fetus presents an additional challenge to the management of an already complex problem. The passage of iron across cellular membranes and its transport throughout the body involve complex processes. Iron, in physiological conditions, exists in its oxidized (ferric) state. Since ferric iron is insoluble, the plasma protein transferrin is required for its transport. Iron traverses membranes by receptor-mediated endocytosis, which is an active, rapid, unidirectional process able to function against a concentration gradient. Transferrin is bound by its specific receptor and is then internalized by endocytosis. The iron is cleaved from the transferrin and retained, whereas the transferrin is returned as apotransferrin. This has been documented as the iron transport system across the human placenta (1,2). Therefore, only transferrin-bound iron would be eligible for transplacental passage. This is further supported by studies in pregnant ewes (Fig. 1). Despite massive induced hyperferremia, only a negligible amount of iron was passed to the fetus (69). Thus, unlike the case with other poisons discussed in this chapter, the placenta acts as a barrier to iron.

242

Tenenbein

Figure 1 Maternal and fetal serum iron concentrations in control- and deferoxamine-treated animals. Plotted values are mean ⫾ standard error. Ewes in both groups received intravenous iron as 2 mg/kg maternal body weight over the first 60 minutes. Ewes in the deferoxamine group then received intravenous deferoxamine mesylate as 50 mg/kg maternal body weight over 15 minutes. Open circles ⫽ control ewes; solid circles ⫽ control fetuses; open squares ⫽ deferoxamine ewes; solid squares ⫽ deferoxamine fetuses. (Reproduced by permission from Steven C. Curry, M.D.).

This would explain why the fetus seems to do better than the mother. This ‘‘placental block’’ is supported by lack of neonatal hyperferremia (64–66), lack of histological iron toxicity (23,24,64). and two maternal deaths despite survival of the offspring (65,66). Thus, it would seem that the risk to the fetus is not from the iron itself but is secondary to the induced pathophysiological derangements in the mother. The use of the specific iron chelator deferoxamine during pregnancy is of concern because the product monograph cites it as a proved animal teratogen. These studies have not be published but have been referred to as personal communications in two case reports (63,70). They were briefly described as the prolonged administration of very high doses to pregnant nonhyperferremic mice and rabbits during early gestation. This resulted in skeletal anomalies and decreased ossification in the offspring. Because deferoxamine is negligibly absorbed across the gastrointestinal tract and is a charged and relatively large molecule, it would not be expected to cross the placenta. Therefore, the adverse effects observed in the rodents would most likely be due to chelation of essential nutrients required for skeletal maturation. This would occur over periods of time more prolonged than would typify the treatment of an iron overdose. In addition, chelation of other nutrients would be less likely in a hyperferremic state.

Poisoning in Pregnancy

243

These speculations are supported by three case reports of deferoxamine therapy during pregnancy (70–72). In two of these, deferoxamine was administered for the first 16 and 19 weeks of pregnancy in thalassemic women (70,71). In the third case, a woman received two intramuscular injections during her 10th week. In no case was there evidence of teratogenesis. In the three published cases of iron overdose in pregnant women treated with deferoxamine, all the babies did well (63,64,66). However, these were second- and third-trimester pregnancies. Treatment of Iron Overdose During Pregnancy Management principles should follow those of the nonpregnant patient (67,68). The teratogenic risk of deferoxamine is probably overstated. Nevertheless, as in all situations, maternal well-being takes precedence over fetal concerns. The death of an iron-poisoned mother associated with the withholding of deferoxamine because of concern over teratogenesis was especially tragic (23,24). ORGANOPHOSPHATE PESTICIDES Published Experience At least five cases of organophosphate pesticide poisoning during pregnancy have been described (22,73–75), but two are of limited interest. One of the latter provides only the results of postmortem analysis of various maternal and fetal tissues from a 19-year-old girl who, in her fifth month of pregnancy, committed suicide by ingesting mecarbam (73). The authors demonstrated that this pesticide crosses the placenta and, at least in this maternal–fetal pair, was found in higher concentrations in the conceptus. The other is a 24year-old woman in her third month of pregnancy whose malathion overdose was successfully treated with atropine, obidoxime, and assisted ventilation (74). She underwent a therapeutic abortion 2 months later, and there was no description of the products of conception. Thus, all that can be concluded is that appropriate treatment of organophosphorus poisoning during early pregnancy can result in full recovery of the mother and maintenance of her pregnancy. Two others, a 22-year-old woman in her 36th week who overdosed with methamidophos and a 25-year-old in her 16th week poisoned with fenthion, were both appropriately managed with atropine, pralidoxime, and respiratory support (75). No abnormalities of pregnancy were described, and both women delivered normal babies at term. Thus, it is possible to successfully manage organophosphate poisoning in the second and third trimesters. Perhaps the most interesting is the case report of Weis et al. (22). Although this 21year-old female in her 34th–35th weeks of pregnancy never admitted to ingesting a pesticide, the rapid onset of a severe cholinergic syndrome, the absence of plasma and erythrocyte cholinesterase activity, and the dramatic response to large doses of atropine preclude any other diagnosis. However, the stress of spontaneous onset of labor early on, along with the atropine therapy, resulted in a fetal heart rate of 200 beats/min. An emergency cesarean section was done, resulting in the delivery of a small, floppy, depressed neonate. As in the mother, assisted ventilation and atropine infusions were required. Curiously, the mother seemed to have been more severely affected, since she had a longer requirement for assisted ventilation and atropine, her cholinesterases took longer to return to normal, and her therapeutic atropine dose was a toxic dose for the fetus in utero.

244

Tenenbein

The pathophysiology of acute organophosphate pesticide is well described elsewhere (76). Of interest is the occurrence of two delayed paralytic syndromes. The better known is a peripheral neuropathy involving the distal extremities (77). Onset is a few weeks after the acute phase and is separated from it by a period of recovery. Respiratory failure is not a feature, and at worse its occurrence could complicate subsequent labor and delivery. The second paralytic syndrome was described in 1987 as the ‘‘intermediate syndrome’’ (78). It consists of proximal muscle weakness, multiple cranial nerve palsies, and respiratory failure beginning within 24–96 hours after ingestion. There may not be an apparent period of recovery between onset and the acute phase. The occurrence of this syndrome along with its required therapeutic interventions represent a more serious risk to pregnancy. Interestingly, this intermediate syndrome occurred in three of the cases described above (74,75). Not surprisingly, little is known regarding the human transplacental passage of the many organophosphate pesticides. However, if maternal toxicity follows ingestion, transgastrointestinal epithelial passage has occurred, making fetal entry a result to be expected. This was confirmed for mecarbam (73). The amount passed on to the fetus relative to the maternal body burden would likely differ from compound to compound. Although mecarbam was found in higher concentration in the fetus (73), the more severe maternal course in the case of Weis et al. (20) suggests a relatively smaller fetal burden in that situation. Of additional concern is the possibility of innate differences in the sensitivity to organophosphates by maternal and fetal cholinesterase systems. Decreased activities of neonatal plasma and red cell cholinesterase ranging from 50–70% of adult values have been consistently documented (79–82). Therefore, increased fetal sensitivity to cholinesterase-inhibiting pesticides would be expected. These issues of relative placental passage and potentially differing fetal and maternal sensitivities are important because the chief antidote, atropine, has potent agonistic properties. Therefore, there is the potential for a therapeutic maternal atropine dose being either subtherapeutic or toxic for the fetus. Treatment of Organophosphate Insecticide Poisoning During Pregnancy The basic management in a pregnant woman should not differ from that for the nonpregnant patient. It includes appropriate life support and gastrointestinal decontamination, meticulous respiratory care, and the administration of atropine and pralidoxime (76). The fetus should be closely monitored. If the maternal condition is satisfactory but distress is documented in a potentially viable fetus, consideration should be given to immediate delivery to permit the initiation of therapy ex utero.

DIGITALIS Published Experience With the advent of specific antidotal therapy for digitalis poisoning (83), it is important to discuss poisoning in pregnancy with this drug. However, only one case could be located (84): a 26-year-old woman who during her seventh month of pregnancy ingested 8.9 mg of digitoxin, demonstrating clinical and electrocardiographic evidence of digitalis toxicity. Although the patient was quite ill, she did well with supportive care. On presentation, fetal heart tones were irregular, with a rate of 68 beats per minute and regular. Spontaneous

Poisoning in Pregnancy

245

labor on the fourth day produced a female weighing 2160 g, in poor condition, who died 3 days postpartum. An electrocardiogram at 15 minutes of age supported a digitalis effect on the neonate. Autopsy findings were consistent with prolonged intrauterine anoxia. Because this report goes back to 1960, documentation does not meet current standards. Notably absent are serum drug concentrations and sophisticated fetal monitoring data. Nevertheless, there is a strong case for intrauterine digitalis toxicity. Acute digitalis poisoning and its treatment with Fab antibody fragments is well described elsewhere (85). Therefore our discussion is limited to issues pertinent to this problem in pregnancy (the transplacental passage of these cardiac glycosides and of the Fab antidigoxin antibody fragments). Radioactive tracer studies have demonstrated transplacental passage of both digoxin (86) and digitoxin (87) in humans during the first half of pregnancy. This result is further supported by the practice of treating fetal tacharrhythmias with maternal digitalization. Nagashima et al. report three such cases (88). They demonstrated arrhythmia conversion and similar cord and maternal serum digoxin concentrations. They also cite several other similar case reports. Therefore, the fetus is at risk for cardiovascular dysfunction due to maternal cardiac glycoside overdose. Indeed, Sherman and Locke documented fetal bradycardia (84). Data on the transplacental passage of Fab antibody fragments are lacking. If this antidote does not cross the placenta, its administration to the mother would be of no direct benefit to the fetus. However, transplacental passage is likely, since these fragments are freely excreted in the urine. Nevertheless, a brief case report exploring the use of this agent in eclampsia suggests otherwise (89). This conclusion is questionable, however, since the fetal presence of the fragments was only indirectly inferred from measurements of an endogenous digitalis-like factor, and the maternal and newborn samples were not simultaneous. Treatment of Digitalis Poisoning During Pregnancy The management of the pregnant woman with an acute digitalis overdose should be based on the same principles advised for nonpregnant individuals (85)—namely, initial stabilization, appropriate gastrointestinal decontamination, measurement of serum electrolyte and digoxin concentrations, cardiac monitoring, and supportive care. In addition, fetal cardiac rhythm should be monitored [with fetal echocardiography (88) if available]. If there is maternal indication (hyperkalemia or life-threatening rhythm disturbances) for the administration of antidigoxin antibody fragment, this should be done even though its safety and efficacy for the fetus is unknown. Hopefully, coexistent fetal cardiac arrhythmias would be simultaneously corrected. If not, a second maternal dose should be considered. If still unsuccessful, a viable fetus could be delivered and the antidote administered directly. In the case of a previable fetus, intrauterine cannulation of the fetal vasculature for direct antidote administration and for possible fetal blood sampling could be considered. Another possibility, albeit seemingly unlikely, would be the occurrence of a fetal indication for antidotal therapy in a healthy mother. If this were to occur, the same approach would seem to be appropriate. CAMPHOR Published Experience Camphor ingestions have become a relative rarity, likely owing to concerted efforts to discourage the use of this substance. However, since at least four poisonings during pregnancy have been described (83–86), the topic is reviewed for historical interest.

246

Tenenbein

Weiss and Catalano (90) described a 26-year-old woman at term who ingested 60 mL of camphorated oil instead of castor oil. She had two seizures but delivered a normal baby, who had the obvious odor of camphor. Camphor was qualitatively demonstrated in the newborn’s blood. Riggs et al. (91) described a 40-year-old woman at term who also ingested 60 mL of camphorated oil instead of castor oil She too had a seizure. thirty-six hours later, she delivered a daughter in very poor condition, who died in 30 minutes. Since the course was complicated by preeclampsia, abruptio placentae with hemorrhage, and breech presentation, the contribution of camphor to the fetal demise is uncertain. Camphor was qualitatively demonstrated in the maternal blood, cord blood, and amniotic fluid and in the infant’s brain, liver, and kidneys. Blackmon and Curry (92) described a hospitalized 32-year-old woman in her third month of pregnancy with a threatened abortion who was given 45 mL of camphorated oil in error. Although the patient had four seizures, she carried to term, delivering a normal baby. Jacobziner and Raybin (93) mentioned in passing a 17-year-old pregnant girl who drank 60 mL of camphorated oil to induce an abortion. Although she recovered and her conceptus remained vital, further details were not given. Pharmacology, Toxicology, and Treatment Camphor is both a gastrointestinal irritant and a central nervous system stimulant. Acute symptoms include nausea, vomiting, abdominal pain, tremors, convulsions, and coma. In two of the cases above, placental passage was demonstrated (90,91). The extent of this phenomenon was not quantified. No adverse effects on the conceptus were detected in three of the four cases, and the contribution of camphor to the one postpartum newborn death is uncertain. There is no specific therapy for camphor poisoning.

LEAD Published Experience Aspects to be considered in the discussion of lead exposure during pregnancy include spontaneous abortion, teratogenesis, symptoms and signs of toxicity in the mother or infant, and neurodevelopmental sequelae in the offspring. In keeping with the mandate of this chapter, we discuss only maternal and fetal toxicity. Five such case reports were located (94–99). In two of these, only the mothers demonstrated clinical toxicity (94– 96), whereas in three others toxicity developed only in the offspring during early infancy (97–99). Two other cases of significantly increased maternal and fetal lead burdens are reviewed, even though the mother-child pairs had no evidence of clinical toxicity (100,101). In 1964 Angle and McIntyre described a woman in her eighth month of pregnancy who complained of fatigue and abdominal pain (94). She had been exposed to the fumes of burning battery casings for the preceding 2 months, and her blood lead concentration was 240 µg/dL. The patient was treated with parenteral calcium disodium edentate (EDTA) for 1 week. Four weeks later she delivered a normal male. Cord blood lead concentration was reported as less than 60 µg/dL. However, the polarographic methodology used in this assay is questionable because of its low sensitivity. The other example of maternal toxicity is a 17-year-old girl who presented in her eighth month of pregnancy with abdominal pain, paresthesias in her feet, calf pain, gingival lead lines, and moderate anemia. She had been eating plaster chips from her apartment walls during her pregnancy. Blood lead and amniotic fluid lead concentrations were 86 and 90 µg/dL, respectively.

Poisoning in Pregnancy

247

The patient was treated with a 3-day course of EDTA. At delivery 8 days later, her blood lead had fallen to a near normal of 26 µg/dL, while the cord and amniotic fluid remained elevated at 79 and 86 µg/dL, respectively. There was no evidence of clinical toxicity in the neonate; however, skeletal radiographs demonstrated lead lines in the long bones. The baby was treated with a course of EDTA at 2 weeks and at 5 months of age, and her developmental evaluation was normal at 18 months. Three cases of fetal lead exposure presenting as toxicity in early infancy were located (97–99). The mothers were asymptomatic. Palmisano et al. (97) described a 2.5-monthold female infant with failure to thrive, hypertonic legs, spontaneous tremors, and hyperactive deep tendon reflexes; she had been born at term with a weight of 1900 g. The mother frequently drank illicit whiskey, which in this region was known to be contaminated by lead. No other sources of lead were found. Although there were no blood lead concentrations, increased lead burdens were documented in both mother and child by urinary excretion criteria. The infant received a 3-day course of EDTA. This baby’s intrauterine grown retardation, failure to thrive, and neurological abnormalities cannot solely be attributed to her fetal experience with lead because of the concomitant intrauterine alcohol exposure. However, the following two cases offer more convincing evidence that intrauterine lead acquisition produces toxicity during early infancy. Sensirivantana et al. (98) described a 2-month-old girl whose symptoms were seizures, anemia, basophilic stippling of red blood cells, skeletal lead lines, and brown nails. There had been a significant exposure to lead fumes during the 7th month of pregnancy because the mother had burned several electric motor rotors in a metal-salvaging operation. Pregnancy, labor, delivery, and neonatal course were otherwise unremarkable. Lead concentration in the maternal hair was markedly elevated, as was the baby’s blood concentration (113 µg/dL). Extremely high values were also found in the baby’s hair and nails (brown nails). The elevated hair (maternal and child) and nail values as well as the presence of skeletal lead lines support an intrauterine lead acquisition. The infant was treated with dimercaprol, EDTA, and d-penicillamine. The second case, described by Ghafour et al. (99), was a 36-hour-old neonate with seizures, opisthotonic posturing, and frequent spasms. Her blood lead concentrations were subsequently found to be elevated at 66 and 81 µg/dL at 12 and 17 days of age, with a maternal value of 76. The mother used lead-glazed cooking utensils and a lead-based eye cosmetic, both known to cause lead poisoning. She had a history of two miscarriages and two other children with neonatal seizures. This baby was treated with both dimercaprol and EDTA. In the other two reports of fetal lead exposure, both the mother and the baby were asymptomatic. Singh et al. (99) described a 20-year-old woman who sustained an acute lead exposure in the third trimester as a result of removing paint in her home. A blood lead concentration of 61 µg/dL was found, which declined to 39 at delivery, with cord blood level of 50. Ryu et al. (101) described a normal female neonate whose mother had worked for the preceding 3 years in a battery manufacturing plant. The mother’s blood lead concentration varied from a high of 57 µg/dL to 33 at delivery. Elevated concentrations were also documented in the cord blood and at 3 and 6 days of age. Pharmacology, Toxicology, and Treatment Lead serves no useful purpose in the body. It produces chronic rather than acute toxicity, and most individuals with increased body burdens are asymptomatic. Lead is a well-known reproductive toxin, producing spontaneous abortions, decreased fertility, and possible tera-

248

Tenenbein

togenesis. Exposure to lead during infancy can result in adverse neurodevelopmental sequelae, and it is speculated that the developing fetal nervous system may have an increased sensitivity. Maternal cord blood screening studies (102–104) as well as several of the case reports above demonstrate that lead crosses the placenta freely. However, it is unlikely that EDTA, the most commonly used lead chelator, enters the fetus. There has been only one well-documented case of EDTA therapy during pregnancy (95,96). Although the mother’s blood lead level declined from 86 to 26 µg/dL, the cord blood concentration remained elevated at 79, suggesting nonpassage of this chelator. Furthermore, human studies have demonstrated negligible penetration through the gastrointestinal epithelium, red blood cell membranes, and blood-brain barrier (105), making transplacental passage unlikely. There are no similar data for dimercaprol, but because it penetrates the central nervous system, fetal penetration would seem more likely. However, its use in adults is seldom indicated. Since dimercaprol must be given intramuscularly and is associated with several adverse effects, it is a poor candidate for maternal administration for fetal benefit. Dimercaptosuccinic acid (106), a new lead chelator that is structurally similar to dimercaprol and is given orally, seems to be the best choice. However, there is no reported experience with this drug during pregnancy. Even with a chelator exhibiting good fetal penetration, transplacental chelation would be difficult. Although chelators efficiently clear the blood, this compartment is refilled from tissue stores during the first few days after administration. Thus, prolonged courses would be required; this practice could deny the fetus essential trace elements. Thus, maternal indications would seem to be the only valid reason for the administration of lead chelators during pregnancy.

MISCELLANEOUS POISONINGS DURING PREGNANCY The following toxicants are discussed together because of limited experience with women who have ingested them during pregnancy. Arsenic Historically, organic arsenicals were used as antisyphilitics and antiparasitics. Iatrogenic toxicities were not uncommon; Kantor and Levin described one such case during pregnancy and cited others (107). Since other therapeutic agents have long since replaced these compounds, they are not discussed further. Two cases of acute inorganic arsenic poisonings during pregnancy were found (108,109). A 17-year-old girl in her seventh month of pregnancy ingested a rat poison containing arsenic trioxide (108). She received one dose of dimercaprol 24 hours after the ingestion; 3 days later, the patient spontaneously delivered an 1100-g female who died of hyaline membrane disease. High concentrations of arsenic were found in the infant’s liver, kidneys, and brain. The mother survived, although she required hemodialysis for renal failure in the postpartum period. A 30-year-old woman in her 28th week of pregnancy was poisoned with arsenic trioxide (109). She survived despite multiple organ failure and adult respiratory distress syndrome. Intrauterine death occurred during the fourth or fifth day after ingestion. Toxic concentrations of arsenic were found in the fetal liver, kidneys, stomach, and spleen.

Poisoning in Pregnancy

249

Both the foregoing cases show that inorganic arsenic freely crosses the placenta and that mother and fetus alike are at risk. Ideally, supportive and specific therapy should be effective for both patients. The management of arsenic poisoning is well reviewed elsewhere (110). The experience of Lugo et al. (108) regarding the safety of dimercaprol during pregnancy is not helpful, since only one dose was given during the third trimester and the baby died of hyaline membrane disease. Kantor and Levin (107) administered a course of dimercaprol to a woman in her 26th week of pregnancy (iatrogenic organic arsenic toxicity) and concluded that ‘‘it probably had no deleterious effect on the pregnancy.’’ However, dimercaptosuccinic acid, an orally effective analog of dimercaprol, is now available for the treatment of lead poisoning (106). It has been widely used elsewhere for arsenic poisoning and seems to be a better choice (111). However, there is no reported experience with this drug during pregnancy. Paraquat Overdose of the herbicide paraquat is associated with a very high mortality. Hemoperfusion, its touted intervention, is at best of questionable efficacy. The few described cases of paraquat poisoning during pregnancy reflect this dismal prognosis (112–114). Because of decreased substrate for the generation of free oxygen radicals, there is at least theoretical advantage for the fetus in the relatively hypoxic intrauterine milieu. And since hemoperfusion is of questionable value and technically difficult in the neonate, the argument for nonintervention of a potentially viable pregnancy is tenable unless fetal distress is present. Nutmeg The common spice nutmeg has anticholinergic properties and—being an halucinogen— is abused from time to time. A 29-year-old woman in her 30th week of pregnancy ingested a large amount of nutmeg secondary to a cooking error (115). She experienced mild to moderate anticholinergic syndrome. Fetal tachycardia (170 beats per minute) was documented, suggesting transplacental passage. The mother did well with supportive care and delivered a normal baby at term. Cyanide In recent animal studies, Curry and colleagues documented that coadministration of sodium thiosulfate with sodium nitroprosside to pregnant ewes prevented increases in fetal red cell cyanide concentrations. Their results suggest that coadministration of thiosulfate at doses currently in use in nonpregnant patients will prevent fetal as well as maternal cyanide toxicity (116). Quinine When quinine overdose occurs during pregnancy, the desire to induce an abortion should be suspected. Dannenberg et al. (8) described 4 such cases and reviewed 66 others. The maternal death rate was 16% and the abortion rate without maternal death was only 4%. In addition, at least 59% of the offspring had congenital anomalies. Therefore, quinine is of dubious efficacy as an abortifacient, has significant associations with maternal morbidity and mortality, and may be teratogenic.

250

Tenenbein

In overdose, quinine produces tinnitus, deafness, nausea, vomiting, vasodilation, visual impairment, and cardiac arrhythmias (117,118). Permanent visual disturbances and blindness are potential sequelae. Fatalities are usually due to cardiac toxicity. Treatment is nonspecific (general supportive measures and gastrointestinal decontamination as needed). From a theoretical perspective, acid diuresis would hasten the renal elimination of quinine and is often touted in the management of this poisoning. However, this intervention as well as dialysis and hemoperfusion have been shown to be ineffective (117). Since these are far from benign interventions, they should be avoided. Naphthalene Naphthalene poisoning during pregnancy has been reported twice (119,120). In both instances, the mothers were chronic mothball ingestors who presented at term with a hemolytic anemia. Hemolytic anemia and jaundice were found in both babies in the immediate postpartum period. Management of this problem should include the discontinuation of the ingestion of naphthalene and transfusion therapy of both mother and newborn as indicated. The latter should be monitored for hyperbilirubinemia and managed accordingly. It is conceivable that fetal anemia may be identified prior to term. In such instances, intrauterine transfusion should be considered. Hantson et al. (121) have described the first case of methanol poison in late pregnancy. The woman, who ingested 250–500 mL methanol at 38 weeks of gestation, developed initial concentrations of 230 mg/dL of methanol and 33.6 mg/dL of formate. Following normal gynecological examination and fetal monitoring, the poisoning was treated with ethanol, bicarbonate, and hemodialysis. At delivery 6 days later, both the mother and baby were in good health. Ciguatera Ciguatera is a fish-borne food poisoning characterized by gastrointestinal and neurological symptoms occurring within a few hours of ingestion (122). Two cases of ciguatera poisoning during pregnancy have been reported (123,124). One patient was in the early part of her second trimester (123). She had typical symptoms soon after ingestion, including increased fetal movements that persisted for a few hours. Her baby was normal at term and for the 10 months that he was followed thereafter. There were no neurological or muscular abnormalities. The other case involved a woman at term (124). She also developed typical symptoms soon after ingestion of the bad fish, and she experienced ‘‘bizarre,’’ ‘‘tumultuous’’ fetal movements for 24 hours. An elective cesarean section 2 days later produced a term male with left-sided facial palsy and possible myotonia of the hands. The liquor was meconium-stained, and a mild meconium aspiration syndrome was managed without assisted ventilation. In both instances assays of specimens from the fish confirmed ciguatera poisoning. The differences in the two newborns are probably a consequence of the timing of the fetal exposures. The treatment of this poisoning is symptomatic and supportive (123). Amanita Phalloides Poisoning during pregnancy with Amanita phalloides, a very toxic wild mushroom, has been briefly documented in a 21-year-old woman in her eighth month of pregnancy (125).

Poisoning in Pregnancy

251

Early in her course, both blood and amniotic fluid were sampled for the toxin. The former was positive and the latter negative. Two months later, a healthy baby without biochemical evidence of hepatotoxicity was born. Belliardo et al. (125) suggest that their negative amniotic fluid assay and the neonate’s lack of hepatotoxicity support nonpassage of the toxin across the placenta. More data are needed to support this claim. Poisonous Snake Bites Dunnihoo et al. (126) reviewed the literature on poisoning by members of the Crotalidae family (rattlesnakes, cottonmouths, and copperheads) during pregnancy. The venom of these reptiles has procoagulant activity. The authors described 30 cases in limited detail. The maternal mortality rate was 10%, and fetal wastage (intrauterine deaths plus spontaneous abortions) was 43%. There was no mention of antivenin therapy. Significant transplacental passage of this antidote would seem unlikely because of its high molecular weight. This, along with an apparent increase in risk for the fetus, prompts consideration of emergency delivery of a mature fetus in cases of severe maternal envenomation. However, this decision must be balanced against the risk this procedure presents to a mother with compromised coagulation. Clinical Case Answer This young woman has ingested acetaminophen in a toxic dose, where fatalities are not uncommon. She postponed treatment, thus substantially further increasing her risk for severe, or even fatal liver damage. Recent research suggests that N-acetylcysteine (NAC) therapy may have some efficacy even at this late stage. Her general status and especially liver function should be closely followed. In some institutions she may receive NAC even at this stage. Although the drug has been implicated in interrupting collagen formation in some babies, there is no clear-cut established causation. Available literature, although scarce, suggests that the fetus has an increased risk of liver damage and stillbirth. REFERENCES 1. Aisen P, Brown EB. The iron binding function of transferrin in iron metabolism. Semin Hematol 1977; 14:31–53. 2. Huebers HA, Finch CA. Transferrin. Physiologic behavior and clinical implications. Blood 1984; 64:763–767. 3. Garrettson LK, Procknal JA, Levy G. Fetal acquisition and neonatal elimination of a large amount of salicylate. Clin Pharmacol Ther 1975; 17:98–103. 4. Levy G, Procknal JA, Garrettson LK. Distribution of salicylate between neonatal and maternal serum at diffusion equilibrium. Clin Pharmacol Ther 1975; 18:210–214. 5. Czeizel A, Szentesi I, Szekeres J, et al. Pregnancy outcome and health conditions of offspring of self-poisoned women. Acta Pediatr Hung 1984; 25:209–236. 6. Czeizel AE, Mosonyi A. Monitoring of early fetal development in women exposed to large doses of chemicals. Environ Mol Mutagen 1997; 30:240–244. 7. Sim M. Abortion and the psychiatrist. BMJ 1963; 2:145–148. 8. Dannenberg AL, Dorfman SF, Johnson J. Use of quinine for self-induced abortion. South Med J 1983; 76:846–849. 9. Hall A. The increasing use of lead as an abortifacient. BMJ 1905; 1:584–587.

252

Tenenbein

10. Gold J, Cates W Jr. Herbal abortifacients. JAMA 1980; 243:1365–1366. 11. Whitlock FA, Edwards JE. Pregnancy and attempted suicide. Comp Psychiatry 1968; 9:1– 12. 12. Lendvay A, Czeizel AE. A behavioral teratologic study on offspring of self poisoned pregnant women. Acta Paediatr Hung 1992; 32:347–349. 13. Rayburn W, Aronow R, Delancy B, Hogan MJ. Drug overdose during pregnancy: an overview from a metropolitan poison control center. Obstet Gynecol 1984; 64:611–614. 14. Czeizel A, Szentesi I, Molnar G. Lack of effect of self-poisoning on subsequent reproductive outcome. Mutat Res 1984; 127:175–182. 15. Jones JS, Dickerson K, Carlson S. Unrecognized pregnancy in the overdose or poisoned patient. Am J Emerg Med 19997; 15:538–541. 16. Kulig K. Initial management of ingestions of toxic substances. N Engl J Med 1992; 326: 1677–1681. 17. Kulig K, Bar-Or D, Cantrill SV, et al. Management of acutely poisoned patients without gastric emptying. Ann Emerg Med 1985; 14:562–567. 18. Albertson TE, Derlet RW, Foulke GE, et al. Superiority of activated charcoal alone compared with ipecac and activated charcoal in the treatment of acute toxic ingestions. Ann Emerg Med 1989; 18:56–59. 19. Merigian KS, Woodard M, Hedges J Jr, et al. Prospective evaluation of gastric emptying in the self-poisoning patient. Am J Emerg Med 1990; 8:479–483. 20. Tenenbein M. Whole bowel irrigation as a gastrointestinal decontamination procedure after acute poisoning. Med Toxicol 1988; 3:77–84. 21. Van Ameyde KJ, Tenenbein M. Whole bowel irrigation during pregnancy. Am J Obstet Gynecol 1989; 160:646–647. 22. Weis OF, Muller FO, Lyell H, et al. Materno-fetal cholinesterase inhibitor poisoning. Anesth Analg 1983; 62:233–235. 23. Strom RL, Schiller P, Seeds AF, ten Bensel R. Fatal iron poisoning in a pregnant female. Minn Med 1976; 59:483–489. 24. Manoguerra AS. Iron poisoning: report of a fatal case in an adult. Am J Hosp Pharm 1976; 33:1088–1090. 25. Czeizel AE, Tomcsik M, Timor L. Teratologic evaluation of 178 infants born to mothers who attempted suicide by drugs during pregnancy. Obstet Gynecol 1997; 90:195–201. 26. Theil GB, Richter RW, Powell MR, Doolan PD. Acute Dilantin poisoning. Neurology 1961; 11:138–142. 27. Kurtz GG, Michael UF, Morosi HJ, Vaamonde CA. Hemodialysis during pregnancy: report of a case of glutethimide poisoning complicated by acute renal failure. Arch Intern Med 1966; 118:30–32. 28. Vaziri ND, Kumar KP, Mirahmadi K, Rosen SM. Hemodialysis in treatment of acute chloral hydrate poisoning. South Med J 1977; 70:377–378. 29. Trebbin WM. Hemodialysis in pregnancy. JAMA 1979; 241:1811–1812. 30. Wing AJ, Brunner FP, Brynger H, et al. Successful pregnancies in women treated by dialysis and kidney transplantation. Br J Obstet Gynecol 1980; 87:839–845. 31. Hou S. Pregnancy in women requiring dialysis for renal failure. Am J Kidney Dis 1987; 368–373. 32. Rayburn W, Wible-Kant J, Bledsoe P. Changing trends in drug use during pregnancy. J Reprod Med 1982; 27:569–575. 33. Silverman JJ, Carithers RL Jr. Acetaminophen overdose. Am J Psychiatry 1978; 135:114– 115. 34. Byer AJ, Trayler TR, Semmer JR. Acetaminophen overdose in the third trimester of pregnancy. JAMA 1982; 247:3114–3115. 35. Lederman S, Fysh WJ, Tredger M, Gamsu HR. Neonatal paracetamol poisoning: treatment by exchange transfusion. Arch Dis Child 1983; 58:631–633.

Poisoning in Pregnancy

253

36. Ruthnum P, Goel KM. ABC of poisoning: paracetamol. BMJ 1984; 289:1538–1539. 37. Stokes IM. Paracetamol overdose in the second trimester of pregnancy. Br J Obstet Gynaecol 1984; 91:286–288. 38. Roberts I, Robinson MJ, Mughal MZ, et al. Paracetamol metabolites in the neonate following maternal overdose. Br J Clin Pharmacol 1984; 18:201–206. 39. Haibach H, Akhter JE, Muscato MS, et al. Acetaminophen overdose with fetal demise. Am J Clin Pathol 1984; 82:240–242. 40. Robertson RG, Van Cleave BL, Collins JJ Jr. Acetaminophen overdose in the second trimester of pregnancy. J Fam Pract 1986; 23:267–268. 41. Ludmir J, Main DM, Landon MB, Gabbe SG. Maternal acetaminophen overdose at 15 weeks of gestation. Obstet Gynecol 1986; 67:750–751. 42. Riggs BS, Bronstein AC, Kulig K, et al. Acute acetaminophen overdose during pregnancy. Obstet Gynecol 1989; 74:247–253. 43. McElhatton PT, Sullivan GM, Volans GN, Fitzpatrick R. Paracetamol overdose during pregnancy: an analysis of the outcomes of cases referred to the Teratology Information Service of the National Poison Information Service. Hum Exp Toxicol 1990; 9:147–153. 44. Levy G, Garrettson LK, Soda DM. Evidence of placental transfer of acetaminophen. Pediatrics 1975; 55:895. 45. Jackson CH, MacDonald NC, Cornett JWD. Acetaminophen: a practical pharmacologic review. Can Med Assoc J 1984; 131:25–37. 46. Lauterberg BH, Corcoran GB, Mitchell JR. Mechanism of action of N-acetylcysteine in the protection against hepatotoxicity of acetaminophen in rats in vivo. J Clin Invest 1983; 71: 980–991. 47. Rane A, Tomson G. Prenatal and neonatal drug metabolism in man. Eur J Clin Pharmacol 1980; 18:9–15. 48. Perucca E. Drug metabolism in pregnancy, infancy and childhood. Pharmacol Ther 1987; 34:129–143. 49. Horowitz RS, Dart RC, Jarvie DR, et al. Placental transfer of N-acetylcysteine following human maternal acetaminophen toxicity. J Clin Toxicol 1997; 35:447–451. 50. Rollins DE, von Bahr C, Glaumann H, et al. Acetaminophen: potentially toxic metabolite formed by human fetal and adult liver microsomes and isolated fetal liver cells. Science 1979; 205:1414–1416. 51. Selden BS, Curry SC, Clark RF, et al. Transplacental transport of N-acetylcysteins in an ovine model. Ann Emerg Med 1991; 20:1069–1072. 52. Prescott LF. Treatment of severe acetaminophen poisoning with intravenous acetylcysteine. Arch Intern Med 1981; 141:386–389. 53. Jackson AV. Toxic effects of salicylate on the fetus and mother. J Pathol Bacteriol 1948; 60:587–593. 54. Earle R Jr. Congenital salicylate intoxication—report of a case. N Engl J Med 1961; 265: 1003–1004. 55. Rejent TA, Baik S. Fatal in utero salicylism. J Forens Sci 1985; 30:942–944. 56. Hill JB. Salicylate intoxication. N Engl J Med 1973; 288:1110–1113. 57. Lynd PA, Andreasen AC, Wyatt RJ. Intrauterine salicylate intoxication in a newborn. Clin Pediatr (Phil) 1976; 15:912–913. 58. Ahlfors CE, Shwer ML, Ford KW. Bilirubin-albumin binding in neonatal salicylate intoxication. Dev Pharmacol Ther 1982; 4:47–60. 59. Buck ML, Grebe TA, Bond GR. Toxic reaction to salicylate in a newborn infant: similarities to neonatal sepsis. J Pediatr 1993; 122:955–958. 60. Temple AR. Acute and chronic effects of aspirin toxicity and their treatment. Arch Intern Med 1981; 141:364–369. 61. McElhatton PR, Roberts JC, Sullivan FM. The consequences of iron overdose and its treatment with desferrioxamine in pregnancy. Hum Exp Toxicol 1991; 10:251–259.

254

Tenenbein

62. Dugdale AE, Powel LW. Acute iron poisoning: its effects and treatment. Med J Aust 1965; 2:990–992. 63. Blanc P, Hryhorczuk D, Danel I. Deferoxamine treatment of acute iron intoxication in pregnancy. Obstet Gynecol 1984; 64:12S–14S. 64. Rayburn WF, Donn SM, Wulf ME. Iron overdose during pregnancy: successful therapy with deferoxamine. Am J Obstet Gynecol 1983; 147:717–718. 65. Richards R, Brooks SEH. Ferrous sulphate poisoning in pregnancy with afibrinogenaemia as a complication. West Indian Med J 1966; 15:134–140. 66. Olenmark M, Biber B, Dottori O, Rybo G. Fatal iron intoxication in late pregnancy. J Toxicol Clin Toxicol 1987; 25:347–359. 67. Banner W Jr, Tong TG. Iron poisoning. Pediatr Clin North Am 1986; 33:393–409. 68. Proudfoot AT, Simpson D, Dyson EH. Management of acute iron poisoning. Med Toxicol 1986; 1:83–100. 69. Curry SC, Bond GR, Raschke R, et al. An ovine model of maternal iron poisoning in pregnancy. Ann Emerg Med 1990; 19:632–638. 70. Thomas RM, Skalicka AE. Successful pregnancy in transfusion-dependent thalassemia. Arch Dis Child 1980; 55:572–574. 71. Martin K. Successful pregnancy in β-thalassemia major. Aust Paediatr J 1983; 19:182–183. 72. Christiaens GCML, Rijksen G, Marx J, et al. Desferrioxamine in pregnancy. Arch Gynecol 1985; 237(suppl):80. 73. Papadopoulou-Tsoukali H, Njau S. Mother-fetus postmortem toxicologic analysis in a fatal overdose with mecarbam. Forens Sci Int 1987; 35:249–152. 74. Gadoth N, Fisher A. Late onset of neuromuscular block in organophosphorus poisoning. Ann Intern Med 1978; 88:654–655. 75. Karalliedde L, Senanayake N, Ariaratam A. Acute organophosphorus insecticide poisoning during pregnancy. Hum Toxicol 1988; 7:363–364. 76. Tafuri J, Roberts J. Organophosphate poisoning. Ann Emerg Med 1987; 16:193–202. 77. Sananayake N, Johnson MK. Acute polyneuropathy after poisoning by a new organophosphate insecticide. N Engl J Med 1982; 306:155–157. 78. Sananayazke N, Karalliedde L. Neurotoxic effects of organophosphorus insecticides: an intermediate syndrome. N Engl J Med 1987; 316:761–763. 79. Jones PEH, McCance RA. Enzyme activities in the blood of infants and adults. Biochem J 1949; 45:464–467. 80. Zsigmond EK, Downs JR. Plasma cholinesterase activity in newborns and infants. Can Anaesth Soc J 1971; 18:278–285. 81. Karlsen RL, Sterri S, Lyngaas S, Fonnum F. Reference values for erythrocyte acetylcholinesterase and plasma cholinesterase activities in children, implications for organophosphate intoxication. Scand J Clin Lab Invest 1981; 41:301–302. 82. Sanz P, Rodriguez-Vicente MC, Diaz D, et al. Red cell and total blood acetylcholinesterase and pseudocholinesterase in humans: observed variances. J Clin Toxicol 1991; 29: 81–90. 83. Smith TW, Butler VP Jr, Haber F, et al. Treatment of life-threatening digitalis intoxication with digoxin-specific Fab antibody fragments: experience in 26 cases. N Engl J Med 1982; 307:1357–1362. 84. Sherman JL Jr, Locke RV. Transplacental neonate digitalis intoxication. Am J Cardiol 1960; 6:834–837. 85. Ellenhorn MJ, Barceloux DG. Digitalis. In: Medical Toxicology: Diagnosis and Treatment of Human Poisoning. New York: Elsevier, 1988, pp 200–207. 86. Saarikoski S. Placental transfer and fetal uptake of H-digoxin in humans. Br J Obstet Gynaecol 1976; 83:879–884. 87. Okita GT, Plotz EJ, Dans ME. Placental transfer of radioactive digitoxin in pregnant women and its fetal distribution. Circ Res 1956; 4:376–380.

Poisoning in Pregnancy

255

88. Nagashima M, Asai T, Suzuki C, et al. Intrauterine supraventricular tacharrhythmias and transplacental digitalization. Arch Dis Child 1986; 61:996–1000. 89. Goodlin RC. Antidigoxin antibodies in eclampsia. N Engl J Med 1988; 318:518–519. 90. Weiss J, Catalano P. Camphorated oil intoxication during pregnancy. Pediatrics 1973; 52: 713–714. 91. Riggs J, Hamilton R, Hamel S, McCabe J. Camphorated oil intoxication in pregnancy. Obstet Gynecol 1965; 25:255–258. 92. Blackmon WP, Curry HB. Camphor poisoning: report of case occurring during pregnancy. J Fla Med Assoc 1957; 43:999–1000. 93. Jacobziner H, Raybin HW. Camphor poisoning. Arch Pediatr 1962; 79:28–30. 94. Angle CR, McIntire MS. Lead poisoning during pregnancy. Am J Dis Child 1964; 108:436–439. 95. Pearl M, Boxt LM. Radiographic findings in congenital lead poisoning. Radiology 1980; 136:83–84. 96. Timpo AE, Amin JS, Casalino MB, Yuceoglu AM. Congenital lead intoxication. J Pediatr 1979; 94:765–767. 97. Palmisano PA, Sneed RC, Cassady G. Untaxed whiskey and fetal lead exposure. J Pediatr 1969; 75:869–872. 98. Sensirivantana R, Supachadhiwong O, Phancharoen S, Mitrakul C. Neonatal lead poisoning. Clin Pediatr (Phil) 1983; 22:582–584. 99. Ghafour SY, Khuffash FA, Ibrahim HS, Reavey PC. Congenital lead intoxication with seizures due to prenatal exposure. Clin Pediatr (Phil) 1984; 23:282–283. 100. Singh N, Donovan CM, Hanshaw JB. Neonatal lead intoxication in a prenatally exposed infant. J Pediatr 1978; 93:1019–1021. 101. Ryu JE, Ziegler EE, Fomon SJ. Maternal lead exposure and blood lead concentration in infancy. J Pediatr 1978; 93:476–478. 102. Gershanik JJ, Brooks GG, Little JA. Blood lead values in pregnant women and their offspring. Am J Obstet Gynecol 1974; 119:508–511. 103. Zetterlund B, Winberg J, Lundgren G, Johansson G. Lead in umbilical cord blood correlated with the blood lead of the mother in areas of low, medium or high atmospheric pollution. Acta Paediatr Scand 1977; 66:169–175. 104. Angell NF, Lavery JP. The relationship of blood lead levels to obstetric outcome. Am J Obstet Gynecol 1982; 142:40–45. 105. Foreman H, Trujillo TT. The metabolism of C-14 labeled ethylenediaminetetra-acetic acid in human beings. J Lab Clin Med 1954; 43:566–571. 106. Graziano JH, Siris ES, LoIancono N, et al. 2,3-Dimercaptosuccinic acid as an antidote for lead intoxication. Clin Pharmacol Ther 1985; 37:431–438. 107. Kantor HI, Levin PM. Arsenical encephalopathy in pregnancy with recovery. Am J Obstet Gynecol 1948; 56:370–374. 108. Lugo G, Cassady G, Palmisano P. Acute maternal arsenic intoxication with neonatal death. Am J Dis Child 1969; 117:328–330. 109. Bolliger CT, van Zijl P, Louw JA. Multiple organ failure with the adult respiratory distress syndrome in homicidal arsenic poisoning. Respiration 1992; 59:57–61. 110. Ellenhorn MJ, Barceloux DG. Arsenic. In: Medical Toxicology: Diagnosis and Treatment of Human Poisonings. New York: Elsevier, 1988, pp 1012–1016. 111. Aposhian HV. DMSA and DMPS—water-soluble antidotes for heavy metal poisoning. Annu Rev Pharmacol Toxicol 1983; 23:193–215. 112. Talbot AR, Fu CC. Paraquat intoxication during pregnancy: a report of 9 cases. Vet Hum Toxicol 1988; 30:12–17. 113. Fennelly JJ, Gallagher JT, Carroll RJ. Paraquat poisoning in a pregnant woman. BMJ 1968; 3:722–725. 114. Musson FA, Porter CA. Effect of ingestion of paraquat on a 20-week gestation fetus. Postgrad Med J 1982; 58:731–732.

256

Tenenbein

115. Lavy G. Nutmeg intoxication in pregnancy. J Reprod Med 1987; 32:63–64. 116. Curry SC, Carlton MW, Raschke RA. Prevention of fetal and maternal cyanide toxicity from nitroprusside with confusion of sodium thiosulfate in gravid ewe. Anesth Analg 1997; 84: 1121–1126. 117. Batemman DN, Blin PG, Woodhouse KW, et al. Pharmacokinetics and clinical toxicity of quinine overdose. Lack of efficacy of techniques intended to enhance elimination. Q J Med 1985; 54:125–131. 118. Dyson EH, Proudfoot AT, Prescott LF, Heyworth R. Death and blindness due to overdose of quinine. BMJ 1985; 291:31–33. 119. Anziulewicz JA, Dick HJ, Chiarulli EE. Transplacental naphthalene poisoning. Am J Obstet Gynecol 1959; 78:519–521. 120. Zinkham WH, Childs B. A defect of gluthathione metabolism in erythrocytes from patients with a naphthalene-induced hemolytic anemia. Pediatrics 1958; 22:461–471. 121. Hantson P, Lambermont JY, Mahieu P. Methanol poisoning during late pregnancy. J Clin Toxicol 1997; 35:187–191. 122. Gillespie NC, Lewis RJ, Pearn JH, et al. Ciguatera in Australia: occurrence, clinical features, pathophysiology and management. Med J Aust 1986; 145:584–590. 123. Senecal PE, Osterloh JD. Normal fetal outcome after maternal ciguateric toxin exposure in the second trimester. J Toxicol Clin Toxicol 1991; 29:473–478. 124. Pearn J, Harvey P, DeAmbrosis W, et al: Ciguatera and pregnancy. Med J Aust 1982; 1: 57–58. 125. Belliardo F, Massano G, Accomo S. Amatoxins do not cross the placental barrier. Lancet 1983; 1:1381. 126. Dunnihoo DR, Rush BM, Wise RB, et al. Snake bite poisoning in pregnant: a review of the literature. J Reprod Med 1992; 37:653–658.

15 Carbon Monoxide Poisoning During Pregnancy Benoit Bailey Hoˆpital Ste-Justine, Montreal, Quebec, Canada

INTRODUCTION Year after year, carbon monoxide accounts for many poisonings and fatalities worldwide. Its unique properties make it a potent silent killer. Sources of carbon monoxide include incomplete combustion of hydrocarbons and metabolism of methylene chloride. Carbon monoxide is not only dangerous to the adult but, as it easily crosses the placenta, may also endanger the fetus. Exposure to carbon monoxide during pregnancy can be caused by attempted suicide, accidents due to defective gas appliances or other sources, or in the workplace, such as firefighters or when workers are exposed to methylene chloride. Once carbon monoxide poisoning is recognized during pregnancy, not necessarily an easy task, physicians will be challenged by deciding if the pregnant patient should be treated, and if so, if she should received 100% oxygen by mask only or also hyperbaric oxygen. Upon disposition of such patients, the outcome of the pregnancy will also become a question for the mother, her family, and the physician. This chapter aims at helping to counsel women exposed to carbon monoxide during pregnancy.

PATHOPHYSIOLOGY AND CLINICAL PRESENTATIONS Carbon monoxide is an odorless, colorless, tasteless, nonirritating gas that impairs oxygen delivery and utilization by different mechanisms, leading to cellular hypoxia. Carbon monoxide binds to hemoglobin 250 times more avidly than oxygen, causing a decrease in the amount of hemoglobin available to transport oxygen. The carboxyhemoglobins produced also impair the release of oxygen by increasing oxygen binding to hemoglobin: this leads to a shift in the oxyhemoglobin dissociation curve to the left (1). It also binds to cytochrome oxidase, thus impairing cellular respiration (1). Carbon monoxide poisoning during pregnancy is challenging. The gas readily crosses the placenta and tends to accumulate in the fetal blood because of its greater affinity for fetal hemoglobin than for adult hemoglobin; this causes fetal levels that are 10–15% higher than in the maternal blood (2,3). Also, because the fetal oxyhemoglobin 257

258

Bailey

curve lies further to the left than that of the adult, the fetus is even more affected by the shift to the left caused by carbon monoxide (2,3). Furthermore, the half-life of carboxyhemoglobin is increased in the fetus in comparison to adult: 7–9 hours versus 4–6 hours, respectively (2,3). Therefore, the fetus is exposed to a higher carbon monoxide level than the pregnant woman and for a longer period of time. In addition to the direct fetal effect of carbon monoxide, the mother’s carbon monoxide poisoning will decrease the oxygen content of the uterine artery, thus decreasing fetal oxygen delivery. This decrease oxygen content will also contribute to fetal hypoxia. Symptoms of carbon monoxide poisoning are often nonspecific and may be confused with those of other illnesses; a high degree of suspicion is needed. Patients with acute poisoning may present with nausea, vomiting, dizziness, and headache in the mild case. As the poisoning becomes more severe, various neurological and cardiac symptoms appear, leading to coma, myocardial depression, and death. Chronic carbon monoxide poisoning will be associated with nausea, vomiting, and headache followed by insidious intellectual deterioration and, in some patients, myocardial ischemia (1). Presentations in the pregnant woman will not differ. However, because of the facts previously discussed, carbon monoxide poisoning will represent a unique challenge.

ACUTE POISONING DURING PREGNANCY Most of the information on acute carbon monoxide poisoning during pregnancy comes from case reports or retrospective cases series (4–44). It is not surprising that most of these cases are the more serious poisonings in the mothers, with adverse outcomes in the infants. There is very limited information on mild poisoning during pregnancy, the most prevalent type of poisoning usually encountered according to the Toxic Exposure Surveillance System of the American Association of Poison Control Centers (45). Table 1 shows reported cases of carbon monoxide poisoning with clinical grading of central nervous system severity according to the New York Hyperbaric Group (46) and maternal and fetal outcomes that were evaluated with the original paper (4–25,43,44); some older and foreign-language literature could not be evaluated (26–42). Table 2 describes the clinical grading of the central nervous system (CNS) severity of the New York Hyperbaric Group (46). Grade of maternal carbon monoxide toxicity in relation to fetal outcome is summarized in Table 3. From the cases of carbon monoxide poisoning during pregnancy that could be retrieved in the literature (Table 1), it appears that published cases are evenly distributed between mild and severe cases and that the pregnancy outcome was worst when the mother suffered severe toxicity. Normal pregnancy outcome following either grade 4 or 5 carbon monoxide poisoning was published only once (44). However, because of the bias involved in reporting cases, this does not necessary reflect the reality. Abnormalities Induced by Carbon Monoxide Poisoning During Pregnancy Numerous animal models have demonstrated the adverse effects on the fetus of carbon monoxide poisoning during pregnancy: these include malformations, neurological dysfunction, decreased birth weight, and increased fetal death. Most of the abnormalities induced by carbon monoxide poisoning reported during human pregnancy are neurological

Carbon Monoxide Poisoning During Pregnancy

259

dysfunctions or fetal deaths. Several neurological dysfunctions have been described in newborns that suffered carbon monoxide poisoning during pregnancy: hypotonia, hyporeflexia, difficulty swallowing and sucking, and seizure (13). Dysfunction described in infants and children have included mental and motor disabilities, seizures, microcephaly, hydrocephaly, tetraplegia, and involuntary movements (47,48). Furthermore, it is important to note that whenever postmortem examinations were done, there was evidence of lesions predominantly in the basal ganglia and the deeper layers of the gray matter (4). These lesions are similar to those caused by other hypoxic-ischemic injuries in children or animals (47), giving plausibility to the association of carbon monoxide poisoning during pregnancy and the observed neurological dysfunctions. Acute carbon monoxide poisoning has also been associated with various malformations of the upper and lower extremities (23,35,38,49), mongoloid-type malformations (21), and multiple abnormalities including low-set ears, oral cavity anomalies, and hypoplastic organ development (8,23) when exposure occurred in the first trimester. Definitive association between the carbon monoxide poisoning and the malformations is not possible because of the absence of a controlled study. However, because of the numbers of CNS dysfunctions described and the biological plausibility of the lesions, carbon monoxide can be considered a human teratogen—the term teratogen being use here in the broad sense and referring not only to morphologic malformations. However, the extent of the problem is difficult to appreciate from case series. Prospective Evaluation of Newborns Exposed to Carbon Monoxide Poisoning During Pregnancy There has been only one study aimed at evaluating prospectively the outcome of acute exposure to carbon monoxide during pregnancy (50). The Motherisk study collected and followed cases of carbon monoxide poisoning during pregnancy from different centers over a period of more than 3 years (50). When cases were identified and referred to Motherisk, participating physicians were sent a questionnaire to complete. Pregnant women were contacted and asked similar questions. The questionnaire explored the type and duration of carbon monoxide exposure, nature and severity of symptoms, time elapsed before treatment (if any), carboxyhemoglobin levels, and measurement of carbon monoxide in the air (50). The severity of the carbon monoxide exposure was graded according to the clinical grading system of the New York Hyperbaric Group of CNS severity poisoning (Table 2) (46). A second telephone contact was made to the mother a few weeks after the expected date of delivery to obtain data on the final course of pregnancy and birth. Thereafter, infants were followed up annually by telephone with a developmental questionnaire derived from the Denver Developmental Screening Test. Detailed, structured questions were asked of the mother by a developmental pediatrician (50). In all, 40 cases were collected: there were 3 twin births, one therapeutic abortion at 16 weeks gestation, and 4 pending births. Therefore, analysis was based on a total of 38 children. Exposures occurred in the first trimester in 12 cases and the second and third trimesters in 14 cases each. Table 4 summarizes cases for which a carboxyhemoglobin level was available. The outcome of pregnancy related to the severity of carbon monoxide poisoning was reported in 36 cases. All 31 women with grade 1 or 2 poisoning had a normal outcome. Of the 29 grade 1 cases, 19 had no treatment and 10 received high-flow 100% oxygen by mask. There were 2 grade 2 patients: one had no treatment while the second one had both high-flow 100% oxygen by mask and hyperbaric oxygen.

260

Table 1

Case Report or Series of Carbon Monoxide Poisonings During Pregnancy and Maternal and Fetal Outcome According to Severity Mother

Timing w w w m m w w w m m w m m m

5.4 24.5 35 — — 2.8 23 9.6 — — 7 6.5 — —

Grade

Treatment

1 1 1 5 5 5 1 1 5 5 4 5 5 5

O2 O2 HBO — O2 O2 O2 O2 — — O2 — — —

Outcome D/C Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive

Outcome Short term Alive Alive Alive Alive Alive Died Alive Alive Alive Stillborn Stillborn Alive Alive Alive

Long term

Reference

Developmental delay (4 y) Normal 6 m Normal (2 m) Dead 7 d (T-21 ⫹ CF) Encephalopathy (5 m) 30 w gestation/multiple anomalies N/A N/A Encephalopathy (8 y) — — Developmental delay (13 m) Encephalopathy (4 y) Encephalopathy (7 y)

(10) (6) (43) (21) (23) (8) (8) (8) (22) (25) (14) (13) (21) (21)

Bailey

6 8 8 2 3 13 16 20 6 6 28 7 7 7

COHb

Infant

39 23.7 22.4 — — — — 5.3 47.2 52 5 — 6.3 — 32 — — — —

1 1 1⫹ 5 Dead 5 Dead 1 3 5 1⫹ 1⫹ 2 5 5 1⫹ 5 2 ?

O2 O2 HBO — — — — HBO HBO O2 O2 c/s HBO/C/S — O2 — O2 — —

Alive Alive Alive Alive Dead Alive Dead Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive

Alive Stillborn Alive Alive Stillborn Dead 13 h Stillborn Alive Alive Alive Stillborn Dead 24 d Alive Alive Stillborn Stillborn Stillborn Stillborn Stillborn

N/A — N/A Encephalopathy (3 y) — — — Normal 1 m Normal 6 m Normal (4 m) — — N/A Dead 10 d — — — — —

(8) (5) (11) (21) (20) (17) (19) (12) (9) (44) (8) (4) (7) (21) (8) (15) (16) (18) (24)

Carbon Monoxide Poisoning During Pregnancy

30 w 32 w 32 w 8m 8m 9m 9m 37 w 37 w 37 w 38 w 38 w 38 w 38 w 38 w 39 w Near term Term ?

Abbreviations: COHb ⫽ maximum carboxyhemoglobin level; D/C ⫽ maternal outcome at discharge from hospital; h ⫽ hour; w ⫽ week; m ⫽ month; y ⫽ year; C/S ⫽ cesarian section; O 2 ⫽ 100% oxygen by mask; HBO ⫽ hyperbaric oxygen treatment; N/A ⫽ not available; T-21 ⫽ trisomy 21; CF ⫽ cystic fibrosis.

261

262

Bailey

Table 2 Modified Clinical Grading of Central Nervous System Severity in Patients with Carbon Monoxide Poisoning Grade 1 Grade 1⫹ Grade 2 Grade 3 Grade 4 Grade 5

Alert, oriented. Symptoms include headache, dizziness, and nausea. As grade 1 but with a relative that was unconscious. Alert but with alterations in mental state, confusion, abnormal behavior. Symptoms more pronounced than grade 1. Not alert, disoriented, loss of recent memory, muscle weakness or incoordination. If intubated, responds to simple commands; if not, attention engaged easily. Very disoriented, depressed sensorium. Responses to simple commands limited and frequently inappropriate. Comatose; responds only to pain or no reaction.

Source: From Ref. 46

Adverse outcome occurred only after grade 4 or 5 poisoning—that is, in the more severe cases. There were 3 cases with grade 4 severity. Of these 3 cases, 2 had normal outcomes at 1 year of age after exposure at 27 and 28 weeks gestation with carboxyhemoglobin levels of 39 and 21% respectively: both were treated with hyperbaric oxygen. The third woman with grade 4 was exposed at 23 weeks gestation with a carboxyhemoglobin of 25%: she received high-flow 100% oxygen by mask for 2 hours. The delivery at term was uneventful except that the infant had an apneic spell immediately after birth. At 8 months, she had developmental delay diagnosed as secondary to cerebral palsy; computed tomography was compatible with postanoxic encephalopathy. The 2 cases with grade 5 severity had adverse outcomes. The first was stillborn at 29 weeks gestation; the mother was treated with high-flow 100% oxygen by mask because of a carboxyhemoglobin of 26%. The second woman had a fetal death at term. Grades 4 and 5 were associated with significantly worse pregnancy outcome compared with grades 1 and 2 ( p ⬍ 0.0001 by Fisher’s exact test). The authors concluded that pregnancy outcome following grades 4 and 5 poisoning in the study closely resembled that described in case reports (50). Moreover, despite limited numbers of patients in grades 4 and 5 (five in total), they noticed that two pregnant women treated with hyperbaric oxygen had normal outcomes and that three treated with high-flow 100% oxygen had adverse fetal outcomes. They also concluded that mild acciTable 3 Maternal Carbon Monoxide Toxicity Grade in Relation to Fetal Outcome

Maternal grade 1 1⫹ 2 3 4 5 ? Total

Outcome

No. of cases

Normal

CNS damage

Dead or stillborn

8 4 2 1 1 16 1 33

6 1 1 1 0 1 0 10

1 0 0 0 0 6 0 7

1 3 1 0 1 9 1 16

Sources: Data from Refs. 4–18, 21–23, 25, 43, 44.

Carbon Monoxide Poisoning During Pregnancy

263

Table 4 Cases of Carbon Monoxide Poisoning During Pregnancy Prospectively Collected and Followed with Maternal and Fetal Outcome According to Severity Mother Timing

COhb

Grade

Treatment

D/C

Infant outcome

23 w 27 w 28 w 29 w 30 w — — — — — — —

25 39 21 26 14 40–50 18 14 6 2 2 1

4 4 4 4 2 5 1 1 1 1 1 1

O2 HBO HBO O2 O 2 ⫹ HBO O2 O2 — — — — —

Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive

Cerebral palsy Normal Normal Stillborn Normal TA Normal Normal Normal Normal Normal Normal

Abbreviations: COHb ⫽ maximum carboxyhemoglobin level; D/C ⫽ maternal outcome at discharge from hospital; TA ⫽ therapeutic abortion; w ⫽ week; O 2 ⫽ 100% oxygen by mask; HBO ⫽ hyperbaric oxygen treatment. Source: From Ref. 50.

dental poisoning likely would do no harm. Limitations of the study included the absence of control and the lack of proper neurobehavioral assessments in most cases that could have detected subtle CNS dysfunction. Also, because it is likely that the importance of fetal effect depends not only on maternal symptoms but also on duration of exposure, the amount of carbon monoxide to which the fetus is exposed and probably the stage of fetal development (9), the whole picture was not described even in that prospective study. Despite these limitations and the limited number of cases and considering the rarity of carbon monoxide poisoning during pregnancy, it is unlikely that more patients can be collected easily or in a better way. We can conclude from the Motherisk study and retrospective case reports that a normal infant outcome is more probable in mothers who did not have severe symptoms of carbon monoxide poisoning independent of the treatment used. Also, in the more severe cases, the Motherisk study suggests that treatment with hyperbaric oxygen may influence outcome. This suggestion, despite being believed by most clinicians, is not based on enough data to support it conclusively.

EFFECT OF HYPERBARIC OXYGEN ON PREGNANCY Hyperbaric oxygen is used routinely in the treatment of severe carbon monoxide poisoning in nonpregnant patients. It works based on the fact that it increases the arterial oxygen content by increasing the oxygen dissolved in the plasma and by increasing the removal of carbon monoxide from the blood and the intracellular cytochrome (9). The half-life of maternal carboxyhemoglobin is 320 minutes with an inspired oxygen fraction of 21%. Administration of 100% oxygen by mask decreases the maternal carboxyhemoglobin halflife to 80 minutes. The use of hyperbaric oxygen, with 100% oxygen at 3 atmospheres

264

Bailey

absolute (ATA), further decrease the maternal half-life to 23 minutes (48). The half-life of fetal carboxyhemoglobin is said to be 7–9 hours (2,3). If the mother receives 100% oxygen by mask, the half-life decreases to 3–4 hours (2,3). The fetal carboxyhemoglobin half-life during hyperbaric oxygen is not known. Safety of Hyperbaric Oxygen During Pregnancy The use of hyperbaric oxygen during pregnancy was initially controversial because of the possible adverse effects of hyperoxia on the fetus, including teratogenicity, retinopathy of prematurity, and cardiovascular effects (9). This controversy was further increased by conflicting evidence from animal studies (9). It appears that animal studies with adverse effects all used higher pressures or longer durations of hyperbaric oxygen than are normally used to treat nonpregnant patients; studies with standard treatment did not show any adverse effects of hyperbaric oxygen in animals (51). Some pregnant women poisoned by carbon monoxide during pregnancy and treated by hyperbaric oxygen have been described (7,9,11,12,43). Also, Elkharrat et al. published a prospective study that was part of a study aimed at evaluating the effect of normobaric (high-flow 100% oxygen) and hyperbaric oxygen in acute carbon monoxide poisoning (52,53). They reported on 44 women who knew they were pregnant at the time of the carbon monoxide poisoning and were treated because of their pregnancy with hyperbaric oxygen irrespective of the severity of the poisoning, age of the pregnancy, or carboxyhemoglobin levels (52). Hyperbaric oxygen was carried out for 2 hours, including 1 hour at 2 ATA. High-flow 100% oxygen was also given until hyperbaric oxygen was administered and then for up to 4 hours. Patients were seen 1 month after the poisoning and written reports were obtain from either the obstetrician or the pediatrician at the time of delivery. Of the 44 patients reported, 10 had suffered a loss of consciousness and 2 had coma (grade 4 or 5), while the remaining 32 patients had only gastrointestinal symptoms, headaches, and/or dizziness (grade 1). Thirty-five women were seen at 1 month after the poisoning: 24 were asymptomatic while 11 had moderate symptoms. The pregnancy outcome was available in 38 women: 32 had normal infants at birth. One woman delivered prematurely and one was induced before term. Two women had spontaneous abortions and one opted for therapeutic abortion for personal reasons. Finally, one woman delivered an infant with Down’s syndrome confirmed by karyotype; this child died at 6 days of age because of major cardiopulmonary defects (52). No further follow-up was attempted to evaluate the outcomes of these pregnancies. The authors concluded that hyperbaric oxygen appeared to be safe in pregnant women acutely intoxicated with carbon monoxide (52). In particular, it did not increase the risk of spontaneous abortion in general or in particular when performed early during the pregnancy. These results are further supported by studies from the former Soviet Union, where more than 700 pregnant women have been treated with hyperbaric oxygen for different etiologies (9). However, these studies were not available for review. From the Elkharat study (52) and the former Soviet Union experience, we can conclude that hyperbaric oxygen appears to be safe when use during pregnancy. Because of the limited studies available on the subject, a subtle adverse effect cannot be ruled out. Indications of Hyperbaric Oxygen for Carbon Monoxide Poisoning During Pregnancy If there are some data to suggest that hyperbaric oxygen is probably safe during pregnancy, there is no study that has evaluated the indication. Based on a review of the literature in

Carbon Monoxide Poisoning During Pregnancy

265

1989, Van Hoesen et al, recommended the following for the treatment of pregnant patients poisoned by carbon monoxide (9). Hyperbaric oxygen should be administered if maternal carboxyhemoglobin is above 20% at any time during exposure, if the patient has suffered or demonstrates any neurological signs regardless of the carboxyhemoglobin level, or if signs of fetal distress are present in the form of fetal tachycardia, beat-to-beat variability, or late decelerations (9). When hyperbaric oxygen is not available, 100% oxygen should be administered by mask five times longer than for a nonpregnant patient. They also recommended that if the patient continues to demonstrate neurological signs or fetal distress is still present 12 hours after the initial treatment, additional hyperbaric oxygen treatment might be indicated. These recommendations are still in use despite the absence of formal validation. Any formal validation is unlikely to happen at this time in view of the wide acceptance that hyperbaric oxygen has in the treatment in pregnant women poisoned with carbon monoxide. Based on the results of the Motherisk study (50), these recommendations appear reasonable. But it is possible that many pregnant women have received hyperbaric oxygen without any benefit to their fetuses.

CHRONIC POISONING DURING PREGNANCY Chronic low-level carbon monoxide exposure and its effect on the fetus have been studied because of the importance and effect of maternal smoking on the pregnancy outcome and not because of poisoning from carbon monoxide. The infants born to women who smoke during pregnancy are known to have a higher risk of prematurity and lower-birth-weight as well as a greater risk of perinatal mortality (54). However, the precise role of carbon monoxide is not clear, since cigarettes contain other chemicals as well, including nicotine. There is no study in humans of the effect of low-level chronic exposure to a source of carbon monoxide other than cigarette smoking. It is therefore difficult to assess the effect of chronic carbon monoxide exposure on the fetus. Fortunately, situations where a defective gas appliance is responsible for carbon monoxide poisoning have become less common with the appearance of carbon monoxide detectors in the home (55). Because there is no information at this time, pregnant women with chronic carbon monoxide poisoning should be treated and counseled as if they had an acute poisoning. A ‘‘CHARGE’’ syndrome (ocular coloboma, heart defect, atresia or stenosis of choanae, retarded growth and development and CNS anomalies, genital hypoplasia, and ear anomalies and/or deafness) has been associated with chronic in utero exposure to carbon monoxide (56,57).

CONCLUSION There is no doubt that carbon monoxide is a teratogen throughout pregnancy. Multiple CNS abnormalities reported to date, despite reporting bias, and the biological plausibility of the observed dysfunction proves it. Its role as a teratogen during organogenesis—other than a CNS teratogen—is not clear. At this time, we do not know if carbon monoxide poisoning can cause malformations in humans. Adverse fetal and infant outcome appears to be linked to the severity of maternal symptoms in poisoning. Other factors may also be important but are not identified at the present time. Cases in which the mother exhibits only minor symptoms (GI symptoms, headache and dizziness) are likely to have good

266

Bailey

pregnancy outcome regardless of the use of 100% oxygen or hyperbaric oxygen. However, if 100% oxygen is used, it should be administered five times longer than in a nonpregnant patient. The role of hyperbaric oxygen in the treatment of severe carbon monoxide poisoning is clear. However, its effect on the outcome of pregnant patients with only minor symptoms is not known at present.

REFERENCES 1. Respiratory toxicology. In: Ellenhorn MJ, Schonwald S, Orgog G, Wasserberger J, eds. Ellenhorn’s Medical Toxicology, 2nd ed. Baltimore: Williams & Wilkins; 1997, pp. 1465–1474. 2. Longo LD. The biological effects of carbon monoxide on the pregnant women, fetus, and newborn infant. Am J Obstet Gynecol 1977; 129:69–103. 3. Hill EP, Hill JR, Power GG, Longo LD. Carbon monoxide exchanges between the human fetus and mother: a mathematical model. Am J Physiol 1977; 232:H311–H323. 4. Turpin JC, Escourolle E, Gray F, et al. Intoxication oxycarbone´e chez le foetus. Rev Neurol (Paris) 1978; 134:485–495. 5. Cramer CR. Fetal death due to accidental maternal carbon monoxide poisoning. Clin Toxicol 1982; 19:297–301. 6. Copel JA, Bowen F, Bolognese RJ. Carbon monoxide intoxication in early pregnancy. Obstet Gynecol 1992; 59:26S–28S. 7. Hollander DI, Nagey DA, Welch R, Pupkin M. Hyperbaric oxygen for the treatment of acute carbon monoxide poisoning in pregnancy. J Reprod Med 1987; 32:615–617. 8. Caravati EM, Adams CJ, Joyce SM, Schafer NC. Fetal toxicity associated with maternal carbon monoxide poisoning. Ann Emerg Med 1988; 17:714–717. 9. Van Hoesen KB, Camporesi EM, Moon RE, et al. Should hyperbaric oxygen be used to treat the pregnant patient for acute carbon monoxide poisoning? JAMA 1989; 261:1039–1043. 10. Woody RC, Brewster MA. Telencephalic dysgenesis associated with presumptive maternal carbon monoxide intoxication in the first trimester of pregnancy. Clin Toxicol 1990; 28:467– 475. 11. Brown DB, Mueller GL, Golich FC. Hyperbaric oxygen treatment for carbon monoxide poisoning in pregnancy: a case report. Aviat Space Environ Med 1992; 63:1011–1014. 12. Gabrielli A, Layon AL. Carbon monoxide intoxication during pregnancy: a case presentation and pathophysiologic discussion, with emphasis on molecular mechanisms. J Clin Anesth 1995; 7:82–87. 13. Beaudoing A, Gachon J, Butin LP, Bost M. Les conse´quences foe´tales de l’intoxication oxycarbone´e de la me`re. Pe´diatrie 1969; 24:539–553. 14. Farrow JR, Davis GJ, Roy TM, et al. Fetal death due to nonlethal maternal carbon monoxide poisoning. J Forens Sci 1990; 35:1448–1452. 15. Muller GL, Graham S. Intrauterine death of the fetus due to accidental carbon monoxide poisoning. N Engl J Med 1955; 252:1075–1078. 16. Tissier P. Asphyxie par le gaz d’e´clairage d’une femme au terme d’une grossesse. Obste´trique 1909; 14:911–914. 17. Schwedenberg TH. Leukoencephalopathy following carbon monoxide asphyxia. J Neuropathol Exp Neurol 1959; 18:597–608. 18. Burnod A, Maire G. Cause exceptionnelle de mort d’un foetus in ute´ro: intoxication de la me`re par oxyde de carbone. Bull Fe´d Soc Gyne´col Obstet 1956; 2:191–192. 19. Helpern M, Strassman G. Differenciation of fetal and adult human hemoglobin. Arch Pathol 1943; 35:776–782. 20. Balthazard V, Nicloux M. Intoxication mortelle oxycarbone´e chez une femme enceinte de huit mois. Arch Mens Obstet Gynecol 1913; 3:161–165.

Carbon Monoxide Poisoning During Pregnancy

267

21. Beau A, Neimann N, Pierson M. Du roˆle de l’intoxication oxycarbone´e gravidique dans le gene`se des ence´phalopathies ne´o-natales. Arch Fr Pe´diatr 1956; 13:130–143. 22. Desclaux P, Soulairac A, Morlon C. Intoxication oxycarbone´e au cours d’une gestation, arrieration mentale conse´cutive. Arch Fr Pe´diatr 1951; 8:316–318. 23. Zourbas J. Ence´phalopathie conge´nitale avec troubles du tonus neuro-musculaire vraisemblablement conse´cutive a` une intoxication par l’oxyde de carbone. Arch Fr Pe´diatr 1947; 4:513– 515. 24. Goldstein DP. Carbon monoxide poisoning in pregnancy. Am J Obstet Gynecol 1965; 92: 526–528. 25. Phillips P. Carbon monoxide poisoning during pregnancy. BMJ 1924; 1:14–15. 26. Weiler G, Ribe M, Kloppel A. Zur akuten und subakuten fotalen CO-Intoxikation. Z Rechtsmed 1983; 90:191–197. 27. Weiler G, Ribe M, Kloppel A. Zur Bewertung der kindlichen Schadigung bei akuter COVergiftung der graviden Frau. Geburtsh Frauenheilk 1984; 44:744–748. 28. Bankl H, Jellinger K. Zentralnervose Schaden nach fetaler Kohlenoxydvergiftung. Beitr Pathol Anat 1967; 135:350–76. 29. Colmant HJ, Wever H. Prenatalee Kohlenoxydvergiftung mit Organtod des Zentralnervensystems. Arch Psychiatr Zeitschr Ges Neurol 1963; 204:271–287. 30. Csermely M. Uber die pathogenese des cerebrum polycysticum. In: Jacob H, ed. IV Internationaler Kongress fur Neuropathologie, Munich. Stuttgart; Thieme, 1962; pp. 44–48. 31. Hallerdoven J. Uber eine Kohlenoxydvergiftung im fettalehen mit Entwicklungsstrungen der Hirnrunde. Allg Zchr Psychiatr 1949; 124:289–298. 32. Maresh R. Uber einem Fall von Kohlenoxydgahadigung der Kinder in der Gebarmutter. Wien Klin Wochschr 1929; 79:454–456. 33. Neuburger F. Fall einer intra-uterine Hirnschadigung nach einer Leuchtgastervergiftung der Mutter. Beitr Gericht Med 1932; 13:85–95. 34. Solcher H. Uber einem Fall von uberstandener fataler Kohlenoxydvergiftung. J Hirnforsch 1957; 3:49–55. 35. Bette H. Extremitaten Missbildungen nach Leuchtgasvergiftung der Mutter, kasuistike Beitrag zur Missbildungsforschung. Munch Med Wochenschr 1957; 99:1246. 36. Brander T. Microcephalus und Tetraplegie bei einem Kinde nach Kohlenmonoxydvergiftung der mutter wahrend der Schwangerschaft. Acta Padiatr 1940; 28:123–132. 37. Breslau F. Intoxication zweier schangeren mit holzleuchtgas: Tod und vorzeitige Geburt eines Kindes. Monatsschr Geburtsk Frauenkrankh 1859; 13:449–456. 38. Corneli F. Contributo sperimentale all’azione teratogenica dell’ossido di carbonio nei mammiferi. Ortop Traumatol 1955; 23:261. 39. Debre R. Le´sions ce´re´brales par anoxie ne´o-natale. Arch Fr Pe´diatr 1955; 12:673. 40. Freund MB. Ein fall von abstaiben der Frucht in siebenten Schwangerschaftsmonate in folge von nur massiger Intoxication der mutter durch Kohlenoxydgas. Monatsschr Geburtsch Frauenkrankh 1859; 14:31–33. 41. Marchal C, Manciaux M, Neimann M. Les foetopathies d’origine toxique. Rev Praticien 1967; 4:155. 42. Martland HS, Martland HSJ. Placental barrier in carbon monoxide barbiturate and radium poisoning. Am J Surg 1950; 80:270–279. 43. Larcan A, Landes P, Vert P. Intoxication oxycarbone´e au 2e mois de grossesse sans anomalie ne´onatale. Bull Fe´d Socs Gyne´c Obstet Lang Fr 1969; 22:338–339. 44. Margulies JL. Acute carbon monoxide poisoning during pregnancy. Am J Emerg Med 1986; 4:516–579. 45. Litovitz TL, Smilkstein M, Felberg L, et al. 1996 annual report of the American Association of Poison centers Toxic Exposure Surveillance System. Am J Emerg Med 1997; 15:447–500. 46. Peirce EC, Kaufman H, Bensky WH, et al. A registry for carbon monoxide poisoning in New York City. Clin Toxicol 1988; 26:419–441.

268

Bailey

47. Ginsberg MD, Myers RE. Fetal brain injury after maternal carbon monoxide intoxication. Neurology 1976; 26:15–23. 48. Peterson JE, Steward RD. Absorption and elimination of carbon monoxide by inactive young men. Arch Environ Health 1970; 21:165–171. 49. Ingalls TH, Philbrook FR. Monstrosities induced by hyperoxia. N Eng J Med 1958; 259:558– 560. 50. Koren G, Sharav T, Pastuszak A, et al. A multicenter, prospective study of fetal outcome following accidental carbon monoxide poisoning during pregnancy. Reprod Toxicol 1991; 5: 397–403. 51. Hardy KR, Thom SR. Pathophysiology and treatment of carbon monoxide poisoning. Clin Toxicol 1994; 32:613–629. 52. Elkharrat D, Raphael JC, Korach JM, et al. Acute carbon monoxide intoxication and hyperbaric oxygen in pregnancy. Intens Care Med 1991; 17:289–292. 53. Raphael JC, Elkharrat D, Jars-Guincestre MC, et al. Trial of normobaric and hyperbaric oxygen for acute carbon monoxide intoxication. Lancet 1989; 2:414–448. 54. De Haas JH. Parental smoking. Its effect on fetus and child health. Eur J Obstet Gynecol Reprod Biol 1975; 5:283–296. 55. Krenzelok EP, Roth R, Full R. Carbon monoxide: the silent killer with an audible solution. Am J Emerg Med 1996; 14:484–486. 56. Courtens W, Hennequin Y, Blum D, Vamos E. Charge association in a neonate exposed in utero to carbon monoxide. Birth Defects 1996; 30:407–412. 57. Hennequin Y, Blum D, Vamos E, et al. In utero carbon monoxide poisoning and multiple fetal abnormalities. Lancet 1993; 341:240.

16 Direct Drug Toxicity to the Fetus Orna Diav-Citrin The Hebrew University, Jerusalem, Israel

Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

Clinical Case One of your patients, a woman with chronic hypertension, is well controlled on captopril. In fact, after stormy years with β-adrenergic blockers, she is for the first time doing well. The patient finds out she is pregnant and would like to know whether she can stay on captopril. The fetal phase, from the end of the embryonic stage to term, is the period when growth and functional maturation of organs and systems already formed occur. Teratogen exposure in this period may adversely affect fetal growth (i.e., intrauterine growth restriction), the size of a specific organ, or the function of the organ rather than causing gross structural anomalies. The term fetal toxicity, rather than teratogenicity, is commonly used to describe such an effect. Fetal toxicity, unlike teratogenicity, is often predicted from the known pharmacological profile and toxicological effects of the agents involved. Fetal toxic effects may have a major impact on the morbidity and mortality of the neonate. In counseling pregnant women regarding drug exposure during pregnancy, it is important to address fetal toxic effects and estimate their risk. Table 1 reviews medicinal drugs that may cause direct fetal toxicity in humans and summarizes the evidence for the fetal toxic effect, with an attempt to estimate the rate of its occurrence. Whenever fetal toxic effects are based on case reports, they are used as a warning signal. In such cases, it is virtually impossible to give a true risk estimate because the denominator of the total number of exposed fetuses is unknown. Other chapters deal with drugs and substances of abuse and with neonatal withdrawal syndromes. The potential effect of psychoactive agents on the developing central nervous system, which belongs to the evolving field of behavioral teratology, is beyond the scope of this chapter. Clinical Case Answer Angiotensin-converting enzyme (ACE) inhibitors are potent antihypertensive drugs. Their use in the second and third trimesters of pregnancy has been associated with fetal toxicity, including intrauterine renal insufficiency. Neonatal hypotension, oliguria with renal failure, and hyperkalemia have been reported with the use of ACE inhibitors in late preg269

No.

270

Table 1

Direct Drug Toxicity to the Fetus Drug

1. Acebutolol

2. Acetaminophen

3. Acetazolamide

4. Acetohexamide

5. Albuterol

Fetal/neonatal toxic effects Lower mean birth weight than in neonates exposed, to pindolol; (higher than in those exposed to atenolol)

Neonatal apnea and hypotonia

Comments/recommendations Not known whether due to the degree of maternal hypertension, the potency of the drug, or a combination of these and other factors

Ref. 1

2

Neonates should be closely observed for signs of β-adrenergic blockade

3

C/R

Continuous maternal high daily dose

4

C/R 1/3

In an overdose situation Resolved after treatment

5 6

C/R

Oral hypoglycemics generally not recommended in pregnancy because of concerns regarding insufficient glycemic control Adverse reactions secondary to the cardiovascular and metabolic effects of the drug Effects more pronounced in diabetic patients Neonatal hypoglycemia should be prevented with adequate intake of glucose Direct adrenergic stimulation of the pituitary (?) Can be reversed with naloxone

7

C/R

8 9–14

15–19 20–21 22 23–24

Diav-Citrin and Koren

Transient hypoglycemia Lower blood pressure and heart rate than of similar infants exposed to methyldopa Polyhydramnios and fatal neonatal kidney disease in the newborn Fetal death with hepatic and renal toxicity Asymptomatic hypocalcemia, hypomagnesemia, and metabolic acidosis in the newborn Prolonged symptomatic hypoglycemia in the newborn Prolonged hypoglycemia and convulsions in the newborn Fetal tachycardia transient fetal atrial flutter Transient fetal hyperglycemia followed by hyperinsulinemia Higher growth hormone levels in cord blood

6. Alfentanil

Reported rate of occurrence

8. Ambenonium

Respiratory depression Transient sinusoidal fetal heart rate pattern Suppression of collagen-induced platelet aggregation Transient muscular weakness

9. Aminoglutethimide

Virilization

10. Amiodarone

11. Ammonium chloride 12. Amphetamine

13. Ampicillin

20% of newborns whose mothers were treated with choline-esterase inhibitors during pregnancy

May be caused by inhibition of adrenocortical function Not clinically significant

Prolonged QT interval Transient bradycardia Transient hypothyroidism Congenital hypothyroidism with goiter Transient asymptomatic hyperthyroidism Elevated serum iodine level IUGR

C/R C/R

Thyroid function tests recommended for newborns exposed to amiodarone in utero because of the high iodine content of the drug May be an effect of amiodarone, a result of the maternal disease or concurrent medications, or a combination of the above In large amounts near term Multifactorial: multidrug abuse, lifestyle, poor maternal health and antenatal care Due to the vasoconstrictive properties of the drug In an acute poisoning Maternal anaphylaxis

22 vs. 11% in not exposed

During the last 2 weeks of pregnancy, specific agents and doses not reported

Fetal acidosis IUGR Prematurity Cerebral injuries Intrauterine death Severe neonatal distress with neurologic sequelae Increased risk of retrolental fibroplasia in premature infants

No short-term neonatal adverse effects were observed Abnormal bleeding not reported Neonatal myasthenia may be caused by transplacental passage of anti-acetylcholine receptor immunoglobulin G antibodies, rather than a drug effect

25–32 33 34 35–36

37–38 39 40–41 44, 46, 48 45 41 42 43, 45

49–50 51 52 53 54 55 56

271

14. Antihistamines

17/40 (42.5%)

Direct Drug Toxicity to the Fetus

7. Alphaprodine

Table 1

Continued Drug

15. Aprotonin 16. Asparaginase

17. Aspirin

18. Atenolol

Decreased fibrinolytic activity Transient bone marrow hypoplasia in the newborn Chromosomal damage IUGR Increased perinatal mortality (stillbirths more than neonatal)

Reported rate of occurrence 2/7 C/R

Depressed platelet function In premature or low birth weight infants after full dose given near term Large doses

Immunosuppression

Chromosomal abberations

In combination with other antineoplastic agents Clinical significance not clear Some associated with antepartum hemorrhage, others may have been caused by closure of the ductus

Premature closure of the ductus arteriosus Decreased clotting ability Increased incidence of intracranial hemorrhage Toxic effects Congenital salicylate intoxication Depressed albumin-binding capacity IUGR Lower birth weight Persistent β-adrenergic blockade

IUGR

Comments/recommendations

Incidence of SGA: 20–40%

Ref. 57 58–63 64 65–67

68 69–76 77–80 81, 82 83, 84 85 86, 88

No increase in the incidence of 85 jaundice May be related to increased vascular resistance and is a function of length of drug exposure Newborns exposed near delivery should be 89, 94 closely observed during the first 24–48 hours for signs and symptoms of β-adrenergic blockade Dose reduction is recommended according 95, 97 to maternal leukocyte count at 32 weeks gestation to avoid neonatal leukopenia and thrombocytopenia Other potential contributors: underlying dis- 98–102 ease, (hypertension, vascular disease, and renal impairment), multiple medications Clinical significance unknown 103

Diav-Citrin and Koren

19. Azathioprine

Fetal/neonatal toxic effects

272

No.

Paralytic ileus

21. Betamethasone

Hypoglycemia Leukocytosis Transient constriction of the ductus arteriosus Transient decrease in glucocorticoid activity in the neonate Profound transient leukopenia with neutropenia and alopecia

22. Bleomycin

Two newborns

C/Rs Clinical significance not clear

C/R

Chromosomal aberrations 23. Bromides

24. Busulfan 25. Butorphanol

26. Caffeine

27. Calcitonin

IUGR Neonatal bromide intoxication (poor suck, weak cry, diminished Moro reflex, lethargy and hypotonia) IUGR Chromosomal aberrations Depressant effect Sinusoidal fetal heart rate pattern Low birth weight Tachyarrhythmias Premature atrial contraction Fine tremors Tachypnea

3 C/Rs

Rebounded and returned to normal in the first hours of life Other chemotherapeutic agents involved, by 12 weeks of age hair regrowth, at 1 year normal development except for moderate hearing loss In human marrow cells, significance to the fetus unknown Normal growth and development after several months

Clinical significance unknown 75% (38/51) vs. 13% (7/55) 25% vs. 1.7% 12.5% vs. 0 100% vs. 10.7% 25% vs. 3.5%

Clinical significance unknown

104 105 106, 107 108 109 110

111 112, 113 114–116

117, 118 119 120–122 123, 124

High consumption 125 Associated with maternal caffeine consump- 126 tion of more than 500 mg/day (n ⫽ 16) in comparison to offspring of women who used less than 250 mg/day (n ⫽ 56) of caffeine 127 Clinical significance unknown 128

273

Fetal behavioral and sleep pattern changes Marked increase of calcitonin concentrations in fetal serum at term

Exposure at term to chlorpromazine as well

Direct Drug Toxicity to the Fetus

20. Benztropine

No.

274

Table 1

Continued Drug

28. Camphor 29. Captopril

30. Carbamazepine

Fetal/neonatal toxic effects

31. Carbimazole 32. Chlorambucil

See methimazole Mutagenicity and carcinogenicity

33. Chloramphenicol

Low birth weight Cardiovascular collapse (‘‘gray-baby syndrome’’)

4 cases

C/R

Comments/recommendations

Ref.

Accidental ingestion

129–132

When used in the second and third trimesters of pregnancy. Speculated mechanism related to drug-induced oligohydramnios, causing a mechanical insult, combined with drug-induced fetal hypotension and decreased renal blood flow. In cases in which maternal disease requires captopril in late pregnancy, monitoring of amniotic fluid volume is advised during gestation as well as close observation of blood pressure and renal function in the neonate The levels were still within normal limits, questionable clinical significance Vitamin K prophylaxis advised to the mothers before the expected time of delivery Exposure during pregnancy and breastfeeding, resolved upon cessation of breast-feeding Converted in vivo to methimazole Not reported in newborns following in utero exposure

133

40% During the final stage of pregnancy

134 135–137

138

139–143 117 144

Diav-Citrin and Koren

Fetal toxicity and neonatal respiratory failure Oligohydramnios IUGR Renal dysplasia Intrauterine fetal death Fetal hypotension Neonatal anuria Renal failure Craniofacial deformations: hypocalvaria/ acalvaria Pulmonary hypoplasia Limb contractures Neonatal death Lower cord serum vitamin D levels compared to normal controls Neonatal vitamin K deficiency associated with early hemorrhagic disease of the newborn Transient cholestatic hepatitis

Reported rate of occurrence

35. Chloroquine

Marked depression of the infants (unresponsive, hypotonic, hypothermic and fed poorly) Cochleovestibular paresis

36. Chlorothiazide

Neonatal hypoglycemia

3 infants

Exposure within hours of delivery Hypotonicity persisted for up to a week

2 cases

2 cases 2 cases

Concern regarding cumulative ocular toxic- 146 ity of high doses used for rheumatological conditions if used for prolonged periods during pregnancy Newborns exposed to thiazide diuretics 147–150 near term should be observed for hypoglycemia as a result of maternal hyperglycemia Transfer of antiplatelet antibody demon148, 149, strated 151–155 150 157 156, 158 159

C/R

High doses, resolved within 3 weeks

160

2 cases

Doxepin coadministered in one Near term, may persist for months

104 161–165

4 cases

Near term, lasted for 4–6 days

166–168

Neonatal thrombocytopenia

37. Chlorpromazine

38. Chlorpropamide

Hemolytic anemia Hyponatremia Hypokalemia Fetal death attributed to maternal hemorrhagic pancreatitis Neonatal hypotonia, lethargy, depressed reflexes, and jaundice Paralytic lieus An extrapyramidal syndrome (tremors, hypertonia, spasticity, hyperactive reflexes) Prolonged neonatal symptomatic hypoglycemia secondary to hyperinsulinism Severe hypoglycemia Hyperbilirubinemia Polycythemia and hyperviscosity

145

Direct Drug Toxicity to the Fetus

34. Chlordiazepoxide

169 10/15 (67%) vs. 13/36 (36%) 4/15 (27%) vs. 1/36 (3.0%)

Generally, insulin provides better glycemic control and is the drug of choice for diabetes in pregnancy

275

Table 1

Drug

Fetal/neonatal toxic effects

Reported rate of occurrence

Fatal fetal subdural hematomas

C/R

40. Cimetidine 41. Cisplatin

Transient neonatal liver impairment Profound transient leukopenia with neutropenia and alopecia

C/R C/R

42. Clofazimine

Skin pigmentation at birth

43. Clonazepam

Neonatal apneic episodes with hypotonia and lethargy

44. Codeine 45. Coumarin derivatives

Neonatal respiratory depression CNS damage Hemorrhage IUGR Miscarriage Prematurity Stillbirth/neonatal death Pancytopenia Transient neonatal severe bone marrow hy- C/R poplasia Low birth weight 40% Chromosomal aberrations C/R

46. Cyclophosphamide

C/R

Comments/recommendations Possible result of vitamin K deficiency (caused by long-term use of high-dose cholestyramine), cholestasis, or both Exposure at term Other chemotherapeutic agents involved, by 12 weeks of age hair regrowth, at 1 year normal development except for moderate hearing loss (thought to be gentamicin- or cisplatin-related) In at least 3 infants, gradually resolved during a 1-year period Exposure throughout pregnancy and in breast-feeding Hypotonia resolved within 5 days, clinical apnea persisted for 10 days, for 10 weeks by follow-up pneumograms Use during labor Exposure in the second and/or third trimesters, probably caused by fetal or neonatal hemorrhage

Exposure to five other antineoplastics in the third trimester Combination chemotherapy including mercaptopurine and radiation Following administration of anticancer drugs during pregnancy Clinical significance unknown

Ref. 170

171 110

172, 173 174

175 176

177 58 117 64

Diav-Citrin and Koren

39. Cholestyramine

276

No.

Continued

C/R

Neonatal leukopenia Neonatal hypoglycemia and mild DIC IUGR

C/R C/R 8–45%

Chromosomal aberrations Pancytopenia

C/R

Low birth weight

40%

49. Dactinomycin

Low birth weight

40%

50. Danazol

Female pseudohermaphroditism

51. Dapsone

Neonatal hemolytic anemia Neonatal hyperbilirubinemia Anemia, hypoglycemia and electrolyte abnormalities Transient neutropenia at 2 months Severe transient bone marrow hypoplasia

⬃30–67% of female fetuses C/R C/R 1 case 2 cases C/R

Low birth weight

40%

53. Deferoxamine

Chromosomal aberrations Low neonatal iron levels

C/R

54. Dexamethasone

Neonatal leukocytosis

48. Cytarabine

52. Daunorubicin

May have been related to hydralazine concurrently administered to the mother Resolved spontaneously In the offspring of renal transplant patients, potentially multifactorial (maternal hypertension, renal function, immunosuppressive drugs, or a combination of the above) Clinical significance unknown Exposure in the third trimester, other neoplastics involved Following administration of anticancer drugs during pregnancy Following administration of anticancer drugs during pregnancy Exposure beyond 8 weeks gestation Spontaneously resolved within 10 days

Five other antineoplastic agents involved, thought to be secondary to mercaptopurine Following administration of anticancer drugs during pregnancy Clinical significance unknown In acute iron overdose at 34 weeks gestation White cell counts returned to normal in ⬃1 week

178 179 180 181

64, 182 177 117 117 183–192 193 194 195 196 58

117 64 197 198, 199

277

Neonatal thrombocytopenia

Direct Drug Toxicity to the Fetus

47. Cyclosporine

No.

278

Table 1

Continued Drug

55. Diatrizoate

57. Diazepam

58. Diazoxide

60. Digitalis 61. Dihydrocodeine bitartrate

Reported rate of occurrence

Clinical neonatal hypothyroidism

3/7

Biochemical neonatal hypothyroidism

6/7

‘‘Floppy infant’’ syndrome (hypotonia, lethargy, and sucking difficulties)

Altered neonatal thermogeneis Loss of beat-to-beat variability in fetal heart rate. Decreased fetal movements Transient fetal bradycardia

Newborn hyperglycemia Newborn alopecia, hypertrichosis, lanuginosa and decreased ossification of the wrist Cervical or vaginal structural changes Masculinization of the female infant Tumors of female and male reproductive systems Fetal intoxication and neonatal death Newborn respiratory depression

Comments/recommendations When administered by intra-amniotic injection with ethiodized oil Greater thyroid suppression, the longer the time interval between injection and delivery The frequency of newborn complications increases when doses exceed 30–40 mg or when diazepam is taken for long periods When administered in labor

After rapid drop in maternal blood pressure following intravenous bolus administration Following intravenous administration Following oral treatment during the last 19–69 days of pregnancy 22–58%

C/R

Ref. 200

201–204

205–208 209 210 211, 212

213, 214 215

180, 216– 218

Following maternal acute digitoxin overdose

219 220, 221, 175

Diav-Citrin and Koren

59. Diethylstilbestrol

Fetal/neonatal toxic effects

63. Docusate sodium

Fetal distress (i.e., bradycardia and loss of beat-to-beat variability) Neonatal hypomagnesemia

C/R

64. Doxepin

Paralytic ileus in a neonate

C/R

65. Enalapril

Oligohydramnios, IUGR, renal dysplasia, intrauterine fetal death, fetal hypotension, neonatal anuria, renal failure, craniofacial deformations, hypocalvaria/ acalvaria, pulmonary hypoplasia, limb contractures, neonatal death

66. Ephedrine

Increase in fetal heart rate and beat-tobeat variability Decrease in uterine blood flow and contribution to intrauterine anoxic insult resulting in neonatal death Abruptio placentae with resulting fetal death at 23 weeks

67. Epinephrine

68. Epoetin alfa

69. Ergotamine

C/R

C/R

Fetal death

C/R

Fetal distress

C/R

Possible result of uterine hyperstimulation

222

Chronic maternal overuse Resolved spontaneously Thought to be caused primarily by coadministered chlorpromazine When used in the second and third trimesters of pregnancy. Speculated mechanism related to drug-induced oligohydramnios, causing a mechanical insult, combined with drug-induced fetal hypotension and decreased renal blood flow. In cases in which maternal disease requires enalapril in late pregnancy, monitoring of amniotic fluid volume is advised during gestation, as well as close observation of blood pressure and renal function in the neonate When used to treat or prevent maternal hypotension following spinal anesthesia Following a large intravenous dose to reverse severe maternal hypotension secondary to anaphylaxis Erythropoietin could not be excluded as a contributing factor in a woman with severe hypertension and chronic renal failure Following acute overdose, in a suicide attempt at 35 weeks Accidental use of ergotamine and caffeine at 38 weeks

223 104 180

Direct Drug Toxicity to the Fetus

62. Dimenhydrinate

224–226 227

228

229 230

279

Table 1

Continued Drug

70. Esmolol

Fetal/neonatal toxic effects

Reported rate of occurrence C/R 2 cases

71. Ethacrynic acid

Ototoxicity observed in the newborn and mother

C/R

72. Ethosuximide 73. Etoposide

Spontaneous hemorrhage in the neonate Oligohydramnios, IUGR

C/R C/R

Anemia, leukopenia, and profound neutropenia and thrombocytopenia in the neonate

C/R

Marked leukopenia with neutropenia in the neonate on day 3 (10 days after in utero exposure to the chemotherapy), scalp hair loss and loss of lanugo at 10 days of age, moderate bilateral sensorineural hearing loss at 1 year follow-up

C/R

Resolved spontaneously by 60 hours of age Newborns exposed near delivery should be closely observed during the first 24– 48 hours for signs and symptoms of βadrenergic blockade Following the use during the third trimester Concomitant exposure to kanamycin Four cycles of chemotherapy consisting of etoposide and cisplatin given from 27 weeks’ gestation for dysgerminoma, initially treated surgically Normal hematological profile in the newborn Following combination chemotherapy consisting of two courses of etoposide, cytarabine, and daunorubicin given from 25 weeks’ gestation for acute myeloid leukemia Treated successfully; at 1 year of age blood counts were normal and treatment was stopped. Combination chemotherapy consisting of etoposide, cisplatin, and bleomycin given from 26 weeks’ gestation, the patient became profoundly neutropenic and developed septicemia By 12 weeks’ of age the infant had hair growth Aminoglycoside therapy also given to the mother and neonate

Ref. 231 232, 233

234

235 236

237

110

Diav-Citrin and Koren

Fetal bradycardia β-adrenergic blockade in the fetus and infant

Comments/recommendations

280

No.

Transient pronounced jaundice with elevated transaminases in the newborn

C/R

75. Fentanyl

Respiratory depression in the neonate Loss of fetal heart rate variability without causing fetal hypoxia

C/R

76. Flecainide

Loss of fetal heart rate variability and accelerations

C/R

Transient conjugated hyperbilirubinemia in the neonate

C/R

Cyanosis and jerking extremities in the newborn Minor extrapyramidal symptoms in the infant at 4 weeks of age

C/R

77. Fluorouracil 78. Fluphenazine

79. Gabapentin

C/R

Severe rhinorrhea and upper respiratory C/R distress at 8 hours of age, poor feeding, periodic vomiting, choreoathetoid movements, and intermittent arching of the body Jaundice and intermittent tremors that oc- C/R curred for about 5 days after birth

Following in utero exposure to etretinate, no other anomalies observed, resolved at 5 months follow-up Following epidural fentanyl during labor Was the only effect in a prospective study of 137 women receiving fentanyl in labor Following treatment of fetal supraventricular tachycardia during the third trimester Fetal heart rate returned to a reactive pattern 5 days after delivery Following successful treatment of fetal supraventricular tachycardia at 28 weeks’ gestation after a trial of digoxin and adenosine had failed Resolved at 2 months follow-up Following third-trimester exposure

238

239 240, 241

242

243

Direct Drug Toxicity to the Fetus

74. Etretinate

244

281

Following treatment in pregnancy with 245 fluphenazine decanoate injections every 3 weeks, readily responded to diphenhydramine, resolved at 2 months of age Following treatment with fluphenazine 246 throughout pregnancy for schizophrenia, symptoms improved with pseudoephedrine solution; rinorrhea and nasal congestion persisted for 3 months Following first-trimester exposure to lamo- 247 trigine in addition to administration of gabapentin before and throughout pregnancy; skin tags and ear anomalies also noted in the infant

Table 1

Drug

Fetal/neonatal toxic effects

Reported rate of occurrence

Neonatal hypoglycemia (blood glucose ⬍25 mg/dL)

4/15 (27%)

81. Hexamethonium

Paralytic ileus

3 cases

82. Hydralazine

Neonatal thrombocytopenia and bleeding

3 infants

Fetal premature atrial contractions

C/R

Lupus-like syndrome in the newborn, who died at 36 hours of age due to cardiac tamponade induced by pericardial effusion

C/R

Fetal distress

5/7 (71%)

83. Hydroxyprogesterone

Masculinization of the female fetus

This is seen 3.5 times more often than in newborns whose mothers were treated with insulin; in 1 newborn, the hypoglycemia persisted for more than 48 hours, insulin recommended when drug therapy indicated for diabetes The drug had been used in the treatment of preeclampsia and essential hypertension Daily exposure throughout the third trimester, may be related to severe maternal hypertension rather than to the drug Noted at 36 weeks gestational age, a week following the addition of hydralazine to the woman’s treatment with methyldopa, tachyarrhythmias not observed. The fetal arrhythmia resolved 24 hours after stopping hydralazine Treated with IV hydralazine during a 6day period in the 28th week of gestation for hypertension, IV methyldopa was administered on the sixth day of therapy, lupus-like syndrome developed in the mother on the fifth day of therapy and gradually resolved following discontinuation of hydralazine and delivery Associated with rapid uncontrolled decline in maternal blood pressure when given IV for hypertension

Ref. 248, 249

250, 251

252–254

255

256

257

258–260

Diav-Citrin and Koren

80. Glyburide

Comments/recommendations

282

No.

Continued

Decreased amniotic fluid volume Oligohydramnios

3/4 (75%) C/R of triplets 8/30 (27%) 3/61 (4.9%)

Low-normal amniotic fluid volume

4/61 (6.6%)

85. Idarubicin

Mild constriction of the ductus arteriosus (non-dose-related) Fetal death

C/R

86. Indomethacin

Reduction in the amniotic fluid volume

6/52 (11.5%)

Reduced fetal urine output Impaired renal function in preterm neonates Constriction of the fetal ductus arteriosus with or without tricuspid regurgitation

Periventricular leukomalacia in preterm neonates

In temporal relationship to therapy with ibuprofen Resolved on discontinuation of treatment, less than with indomethacin Resolved after discontinuation of ibuprofen, true oligohydramnios not observed Normal echocardiograms in all four cases within 1 week of discontinuing therapy Concomitant treatment with cytarabine for acute myeloblastic leukemia during the second trimester In preterm labor with intact membranes at gestational age of 32 weeks or less Usually transient

261 262 263 264 264 264 265

261, 263, 266–268 269 270

Ductal constriction is dependent on the 271–297, gestational age of the fetus, starting as 299, 300 early as 27 weeks and increasing markedly at 27–32 weeks, independent of fetal serum indomethacin levels; usually transient and reversible if therapy is stopped an adequate time prior to delivery

C/R

34% (26/76)

Most likely due to ductal constriction and shunting of the right ventricular outflow into the pulmonary vessels; Results in pulmonary arterial hypertrophy In one twin fetus after 28 days of indometha- 298 cin treatment, possibly due to ductal con- 301 striction, resolved completely within 48 hours of stopping the drug 302

283

Premature closure of the ductus arteriosus Primary pulmonary hypertension of the newborn, persistent fetal circulation after birth Neonatal death Unilateral pleural effusion

Used as a tocolytic agent

Direct Drug Toxicity to the Fetus

84. Ibuprofen

No.

284

Table 1

Continued Drug

Fetal/neonatal toxic effects

87. Insulin (bovine or porcine)

Fetal macrosomia

88. Iodides: Iodide or iodine-containing products Radioiodine

Hypothyroidism and goiter in the fetus and newborn, cardiomegaly

90. Isoproterenol

47% (21/46)

Partial or complete ablation of the fetal thy- Several C/Rs roid gland

Combined hypothyroidism and hypoparathyroidism Hemorrhagic disease of the newborn

An isolated late deceleration of 5 beats per minute in a fetus

Comments/recommendations Crosses the placenta as an insulin-antibody complex, the amount of transfer correlated with the amount of antiinsulin antibodies in the mother, high concentrations of animal insulin in cord blood were significantly associated with the development of fetal macrosomia. It is difficult to separate the effect of insulin from the mother’s glycemic control. Human insulin is the drug of choice for the control of diabetes mellitus in women who may become pregnant. When used for prolonged periods or close to term. Goiter may be large enough to cause tracheal compression and death The effect is dose-dependent (in the pregnancies terminating with a hypothyroid infant 131 I doses ranged from 10 to 225 mCi) and may occur only after 10 weeks gestation when the fetal thyroid begins concentrating iodine

C/R

303

304–309

310–316

317

C/R 2 cases

Ref.

Vitamin K supplementation of women taking the drug near term recommended prior to delivery for prevention Two minutes after the mother received 0.25 µg of the drug

318

319

Diav-Citrin and Koren

89. Isoniazid

Reported rate of occurrence

92. Kanamycin

93. Ketamine

94. Labetalol

Severe neonatal toxicity with respiratory depression and hypotension Hypotension Hypocalcemia Paralytic ileus Tachycardia Hypoglycemia Death ECG changes suggesting myocardial ischemia Ototoxicity: Deafness

C/R

Hearing impairment

2.3% (9/391)

89% 100% 33%

16% 6/9

Depression of the newborn Excessive neonatal muscle tone sometimes with apnea Newborn bradycardia

5 infants

Hypotension Mild transient hypotension in term newborns

1 infant 11 infants C/R

Neonatal hypoglycemia

19.1% (18/94) vs. 9.3% (9/97)

320, 321

Transient

323

The mother also experienced ototoxicity, ethacrinic acid also given during pregnancy In women who had received kanamycin, 50 mg/kg, for prolonged periods during pregnancy When used close to delivery for obstetrical anesthesia, transient dose-related toxicity, usually avoided with lower doses Bradycardia marked and persistent in one infant, all survived Delivered by C/S at 28 weeks’ gestation Other measures of β-adrenergic blockade did not differ when compared to matched controls The woman was also taking a thiazide diuretic Women were treated for mild preeclampsia presenting at 26–35 weeks’ gestation Newborns should be closely observed during the first 24–48 hours for signs and symptoms of α/β-adrenergic blockade if exposed close to delivery

234

320 322

324

325–335

336–341

285

Significantly more with IUGR after exposure to labetolol plus hospitalization compared with hospitalization alone

Rare if cord levels are less than 2 ng/mL, increased if cord blood serum levels exceeded 10 ng/mL When cord levels exceeded 10 ng/mL Not related to cord concentrations

Direct Drug Toxicity to the Fetus

91. Isoxsuprine

Continued

No.

Drug

95. Lidocaine

96. Lisinopril

286

Table 1

Fetal/neonatal toxic effects Abnormal fetal rhythm (tachycardia or bradycardia) Lower scores on tests of muscle strength and tone in newborns compared to controls CNS depression in the newborn (low Apgar scores) Chronic renal failure in the newborn Fatal hypocalvaria and chronic renal failure in the premature newborn

Severe oligohydramnios, IUGR, fetal distress, hypocalvaria, and persistent renal insufficiency Fetal calvarial hypoplasia ACE-inhibitor fetopathy 97. Lithium

50% (6/12)

Comments/recommendations Following paracervical block with lidocaine in 12 laboring women After continuous lumbar epidural blocks

Ref. 342 343

344 2 cases C/R

345–348 Fetal calvarial hypoplasia, open biopsy at 11 weeks of age showed extensive atrophy and loss of tubules with interstitial fibrosis

C/R

C/R See entries on captopril and enalapril for further details Most of the toxic effects self-limited (returning to normal within 2 weeks) Persisted for 2 months or longer

C/R

With high IV maternal doses The woman was also taking clozapine, resolved at 5 days of age

180, 349– 366

367, 368

Diav-Citrin and Koren

98. Lorazepam

Hypotonia, cyanosis, bradycardia, atrial flutter, T-wave inversion on ECG, cardiomegaly, thyroid depression with goiter, nephrogenic diabetes insipidus, polyhydramnios, shock, seizures, gastrointestinal bleeding, hepatomegaly, neonatal jaundice, stillbirth, fetal and newborn lithium intoxication ‘‘Floppy infant’’ syndrome Transient mild floppy infant syndrome

Reported rate of occurrence

Newborn depression and hypotonia

Series

Severe depression at birth

2 cases

Muscle contraction impairment Decreased gastrointestinal motility, ileus, hypotonia Cogenital rickets

Wide-spaced fontanelles and parietal bone thinning Fetal bradycardia 100. Medroxyprogesterone 101. Melphalan 102. Menadione

Ambiguous genitalia IUGR Marked hyperbilirubinemia and kernicterus in the newborn

103. Meperidine

Respiratory depression in the newborn

2/5

2/22 C/R

Especially in premature infants, phytonadione is considered the vitamin K of choice for administration during pregnancy Following exposure during labor, timeand dose-dependent Persisting for several days

105. Mercaptopurine

Myelosuppression Microangiopathic hemolytic anemia IUGR Chromosomal aberrations

2 cases C/R 40% C/R

369–375

376, 378

377 379 380–384

374 385 386 117 387–390

391, 392 393, 394

Mean decline in TcP O211.7 torr mean dura- 395 tion 4.6 minutes after injection Correlated with high fetal mepivacaine 396 blood levels after paracervical block 58, 177 After combination chemotherapy 397 Clinical significance unknown 117 64

287

Fetal bradycardia

Maternal hypothermia with maternal bradycardia

C/R

Impaired behavioral response and EEG changes in the neonate Transient decrease in oxygenation 104. Mepivacaine

Intrauterine hypoxia could not be ruled out as a potential cause or contributing factor Spontaneous remission in one infant, residual effects of anoxic encephalopathy in the second Up to 48 hours after birth Other drugs or maternal severe hypertension could have been the causes With long-term infusions of magnesium, probably due to sustained fetal hypocalcemia Effects returned to normal with time

Direct Drug Toxicity to the Fetus

99. Magnesium sulfate

No.

Continued Drug

106. Methadone 107. Methimazole/ carbimazole

108. Methotrexate

109. Methyldopa

110. Methylene blue

Fetal/neonatal toxic effects Neonatal hyperbilirubinemia Thrombocytosis Mild fetal hypothyroidism

Reported rate of occurrence

C/R

Hypothyroidism evident at 2 months of age Neonatal hypothyroidism with goiter Small goiters in the newborn Severe newborn myelosuppression

C/R

IUGR Chromosomal aberrations Mild reduction in neonatal systolic blood pressure in the first two days of life Transient neonatal nasal obstruction Newborn complications: hemolytic anemia, hyperbilirubinemia, methemoglobinemia, deep blue staining, intestinal obstruction Jejunal atresia

40% C/R 24 infants

Newborn respiratory depression

Respiratory depression on the day 1, hypoglycemia and jaundice on day 3

2 cases 2 cases 2 cases

Comments/recommendations In polydrug users Usually resolves spontaneously within a few days With subsequent mental retardation

In carbimazole-exposed newborns Concomitant administration of other antineoplastic agents Clinical significance unknown Reduction not clinically significant

Following intra-amniotic injection of methylene blue

In 19% (17/89) of twin pregnancies 19% (5/26)

When used in genetic amniocentesis in twins, possible mechanism: mesenteric vasoconstriction When used in combination with succinylcholine for rapid-sequence induction of anesthesia before cesarean section, higher proportion than the one found with thiopenthal When used before elective cesarean section

Ref. 398 399 400 401 402 403 58, 177 117 64 404 405 406–416

417, 418

419

420

Diav-Citrin and Koren

111. Midazolam

288

Table 1

Hypertrichosis Fetal death

114. Morphine

Respiratory newborn depression

115. Nadolol

IUGR, tachypnea, hypoglycemia, hypothermia, cardiorespiratory depression

C/R

116. Nalbuphine

Sinusoidal fetal heart rate pattern Fetal distress and neonatal respiratory depression Fatal respiratory failure in the newborn

C/R

117. Naloxone

2 cases C/R

C/R

118. Naproxen

Primary pulmonary hypertension of the 3 cases newborn with severe hypoxemia, coagulopathy, hyperbilirubinemia and impaired renal function

119. Nitroglycerin

Fetal heart changes (loss of beat-to-beat variability, late decelarations and bradycardia)

12

120. 121. 122. 123. 124.

Transient fetal bradycardia Masculinization of the female fetus Masculinization of the female fetus Neonatal urinary retention Neonatal hyperbilirubinemia ⫹/⫺ clinically significant jaundice Transient choreoathetosis

C/R 0.3–18.3% 25% (1/4) C/R

Nitroprusside Norethindrone Norethynodrel Nortriptyline Oral contraceptives

C/R

Less prominent at 2 months of age Following maternal poisoning in a suicide attempt at 31 weeks gestation When used in labor, more common than with meperidine Other possible contributing factors to some of the effects: concomitant administration of hydrochlorothiazide and maternal disease

421, 422 423 424, 425 426

427 428–432 When used at term to treat fetal heart rate baseline with low beat-to-beat variability Following exposure at 30 weeks for 2–6 days, attributed to prostaglandin inhibition, one infant died at 4 days of age and autopsy revealed a short and constricted ductus arteriosus Dose-dependent, following maternal IV treatment with nitroglycerin without volume expansion for severe pregnancy induced hypertension

433

Direct Drug Toxicity to the Fetus

112. Minoxidil 113. Misoprostol

434, 435

436, 437

438 439–442 439, 441 443 444 Resolved spontaneously

445

289

290

Table 1

Continued

No.

Drug

Fetal/neonatal toxic effects

Reported rate of occurrence

Fetal heart rate changes

126. Pentazocine 127. Phenazocine 128. Phenobarbital

Neonatal respiratory depression Neonatal respiratory depression Early hemorrhagic disease of the newborn

C/Rs

129. Phenytoin

Early hemorrhagic disease of the newborn

C/Rs

C/R

130. Pindolol

Neonatal hemorrhage secondary to thrombocytopenia Decrease in fetal heart rate

C/R

The woman used the nasal spray more frequently than the recommended dosage interval, speculated mechanism: α-adrenergic effect on the uterine vessels reducing the uterine blood flow and causing fetal hypoxia and bradycardia Following use during labor Following use during labor Probably as a result of induction of fetal microsomal enzymes that deplete the already low reserves of vitamin K, resulting in suppression of the vitamin K–dependent coagulation factors II, VII, IX, and X, vitamin K prophylaxis recommended prior to delivery. Probably as a result of induction of fetal microsomal enzymes that deplete the already low reserves of vitamin K, resulting in suppression of the vitamin K–dependent coagulation factors II, VII, IX, and X, vitamin K prophylaxis recommended prior to delivery

Ref. 446

447, 448 449, 450 451–460

451–454, 456, 458– 467

468 When compared before and after therapy in women with pregnancy induced hypertension in the third trimester

469

Diav-Citrin and Koren

125. Oxymetazoline

Comments/recommendations

Stillbirth

C/R

132. Prednisone/ prednisolone

Immunosuppression in the newborn

C/R

Stillbirth

Overactivity, tremors, jitteriness

23.5% (8/34) vs. 3% (1/34) in controls C/R

Hemorrhagic disease of the newborn

C/Rs

133. Primidone

134. Promazine

Neonatal hyperbilirubinemia

135. Promethazine

Neonatal respiratory depression Transient and behavioral and EEG changes

Small case series 100% (28/28)

Impaired platelet aggregation in the newborn

88.9% (16/18)

Podophyllum poisoning following topical administration of 7.5 mL of 25% podophyllum resin to florid vulval warts Following exposure to high doses of prednisone and azathioprine throughout gestation, resolved at 15 weeks of age Attributed to failure of placental function, controls had similar diagnoses

470

Malformed newborn, died at 3 weeks of age Probably as a result of induction of fetal microsomal enzymes that deplete the already low reserves of vitamin K, resulting in suppression of the vitamin K–dependent coagulation factors II, VII, IX, and X, vitamin K prophylaxis recommended prior to delivery From exposure to 100 mg or more during labor, mean bilirubin significantly higher in 317 exposed neonates compared to 272 controls

472, 473

Most women (27/28) received a combination of meperidine with promethazine or phenobarbital, persisted for less than 3 days

95

471

452, 454– 457

Direct Drug Toxicity to the Fetus

131. Podophyllum

474

475 476

477, 478

291

No.

Continued Drug

136. Propofol

137. Propranolol

Fetal/neonatal toxic effects Neonatal changes: Low Apgar scores Hypotonia Somnolence Irritability IUGR Hypoglycemia Bradycardia Respiratory depression Hyperbilirubinemia Polycythemia Thrombocytopenia Hyperirritability Hypocalcemia with convulsions Coagulopathy Respiratory depression Prematurity

Reported rate of occurrence

Comments/recommendations Following use in cesarean section

Ref. 479

5/20 1/20 5/20 14% 10% 7% 4% 4% 1% 0.6% 0.6% 0.6% 80% (4/5) 33% (3/9)

Transient fetal bradycardia 20% (2/10) Mild fetal hypothyroidism Fetal goiter

12% (29/241)

Newborn goiter with tracheal compression

2 cases

Clinical hypothyroidism with mental and physical delay in development Neonatal hepatitis

2 cases C/R

From analysis of 23 reports involving 167 180, 480– live-born infants exposed chronically to 501 propranolol in utero. Other possible factors that may have caused these effects or contributed are maternal disease, concomitant medications, or a combination of those factors. Newborn exposed to propranolol near delivery should be closely observed in the first 24–48 hours for signs and symptoms of β-adrenergic blockade 502 When given IV prior to cesarean section When administered for pregnancy-induced 503 hypertension 504 For dysfunctional labor When used close to term, resolves sponta- 505 neously within a few days 180 Usually smaller than iodide-induced goiters Resulted in death in one case and moder506, 507 ate respiratory distress in the second One of the infants also exposed to high 508, 509 doses of iodide during gestation 510 Resolved spontaneously

Diav-Citrin and Koren

138. Propylthiouracil

292

Table 1

Intrauterine and infantile convulsions

140. Quinidine 141. Quinine

Neonatal thrombocytopenia Auditory and optic nerve damage

142. Reserpine

Neonatal thrombocytopenic purpura Neonatal hemolytic anemia Nasal discharge, cyanosis, lethargy, and poor feeding in the newborn Hemorrhagic disease of the newborn

143. Rifampicin 144. Ritodrine

High doses of pyridoxine early in gestation presumably altered the normal metabolism of pyridoxine leading to intractable seizures in the newborn 0.9% (2/234) Usually in potentially toxic doses as an abortifacient N ⫽ 12 3 cases

Fetal and neonatal complications: Tachycardia Neonatal cardiac arrhythmias Disproportionate septal hypertrophy

145. Scopolamine

Prophylacic vitamin K is recommended prior to delivery

515 516–520 521 522 523 318 524–535

Following in utero exposure to ritodrine for 2 weeks or longer As a result of transient fetal hyperglycemia followed by hyperinsulinemia; more common following intravenous administration Following severe maternal ketoacidosis

Newborn hypoglycemia

Fetal death Neonatal hyperbilirubinemia Transient decrease in the glomerular filtration rate in the newborn Fetal tachycardia, decreased heart rate variability, and decreased deceleration of heart rate Newborn toxicity (fever, tachycardia, lethargy)

In G6PD deficient newborns From use near term

511–514

Direct Drug Toxicity to the Fetus

139. Pyridoxine

Noted between 12–36 hours of age, clinical significance unknown When administered at term

C/R

Reversed following treatment with physostigmine

536–538

539

293

Table 1

Continued Drug

Fetal/neonatal toxic effects

Reported rate of occurrence

146. Sotalol

Fetal bradycardia

5/6

147. Streptomycin

Ototoxicity (cochlear and vestibular)

3 C/Rs

148. Sulfonamides

Jaundice Hemolytic anemia

149. Sulindac 150. Tamoxifen 151. Terbutaline

Transient fetal tachycardia Neonatal hypoglycemia Myocardial necrosis in a newborn Cardiovascular decompensation

Yellow-gold fluorescence in the mineralized structures of a fetal skeleton Permanent yellow-brown discoloration of teeth

Lasting up to 24 hours, newborns exposed near delivery should be closely observed during the first 24–48 hours for signs and symptoms of β-adrenergic blockade From exposure to long-term high doses in late pregnancy When given close to delivery In the one case involving the fetus, resulting in hydrops fetalis and stillbirth

Ref. 540

541–543 544–549 544, 545, 549

550, 551 C/R

Following exposure during the first 20 weeks of pregnancy

C/R In 3 of a quadruplet pregnancy

C/R

Possibly due to downregulation of fetal βadrenergic receptors leading to decreased myocardial function and reduced cardiac output Following exposure prior to delivery Due to the chelating ability of the drug, forms a complex with calcium orthophosphate and becomes incorporated into bones and teeth undergoing calcification, effect on deciduous or permanent teeth depends on timing of exposure in pregnancy

552 553–556 557, 558 559 560

561 562–577

Diav-Citrin and Koren

152. Tetracyclines

Transient constriction of the fetal ductus arteriosus Ambiguous genitalia in a female newborn

3 cases (2 newborns and 1 fetus)

Comments/recommendations

294

No.

Transient tachycardia, irritability, and vom- 3 cases iting in the newborn

154. Thioguanine

Intrauterine fetal death

2 cases

IUGR Neonatal thrombocytopenia IUGR Neonatal hyperbilirubinemia Liver toxicity Neonatal afibrinogenemia with fatal hemorrhage Transient hyperglycinemia Fetal and neonatal distress Transient fetal bradycardia

40% C/R

155. Tolbutamide 156. Valproic acid

157. Vancomycin

158. Vinblastine 159. Vincristine 160. Zidovudine

IUGR IUGR IUGR Anemia

More likely to occur when maternal serum levels at term are in the high therapeutic range or above After antineoplastic therapy with thioguanine and other agents at 15 weeks’ gestation Persisted for ⬃2 weeks

3 cases 3 cases C/R

Causal relationship uncertain May be fatal

2 cases 43% (6/14) C/R

No adverse effects seen in the newborns

40% 40% 4% (2/45) 19% (6/31)

1 g IV dose was given in 3 minutes an hour before delivery, causing maternal hypotension

Difference in hemoglobin between exposed and unexposed infants was 1 g/ dL, occurring at 3 weeks of age, with similar values by 12 weeks of age

578–580

581, 582

117 583 180 583–586 586, 587 588

Direct Drug Toxicity to the Fetus

153. Theophylline

589 590 591

117 117 592 592, 593

C/R ⫽ case reports; IUGR ⫽ intrauterine growth retardation; SGA ⫽ small for gestational age; CNS ⫽ central nervous system; DIC ⫽ disseminated intravascular coagulation; ACE ⫽ angiotensin-converting enzyme; ECG ⫽ electrocardiogram.

295

296

Diav-Citrin and Koren

nancy. Complications of oligohydramnios (i.e., fetal limb contractures, lung hypoplasia, and craniofacial anomalies), prematurity, intrauterine growth restriction, and fetal death have also been reported with the use of these agents late in pregnancy. The adverse effects are related to the hemodynamic effects of ACE inhibitors on the fetus. These adverse effects are termed ACE-inhibitor fetopathy. The teratogenic risk with first-trimester exposure to these agents appears to be low. These agents are frequently used in women of reproductive age due to their efficacy and few side effects. This woman should be offered reassurance of unlikely first-trimester effects and may benefit from a review the fetopathy associated with the use of these agents during late pregnancy (494–596). Based on the adverse effects reported in the literature, she should discontinue the captopril, if possible. An alternative antihypertensive regimen should be substituted for the drug prior to her entry into the second trimester if drug therapy is required during pregnancy. In cases where all other options have been exhausted and captopril is required to treat the maternal disease, careful monitoring of maternal and fetal renal function is prudent. Serial sonograms are important for assessment of fetal growth and amniotic fluid volume. This should be followed by close observation of renal function and blood pressure in the newborn.

REFERENCES 1. Dubois D, Petitcolas J, Temperville B, et al. Treatment of hypertension in pregnancy with β-adrenoceptor antagonists. Br J Clin Pharmacol 1982; 13(suppl):375S–378S. 2. Williams ER, Morrissey JR. A comparison with methyldopa in hypertensive pregnancy. Pharmatherapeutica 1983; 3:487–491. 3. Dumez Y, Tchobroutsky C, Hornych H, Amiel-Tison C. Neonatal effects of maternal administration of acebutolol. BMJ 1981; 283:1077–1079. 4. Char VC, Chandra R, Fletcher AB, Avery GB. Polyhydramnios and neonatal renal failure— a possible association with maternal acetaminophen ingestion. J Pediatr 1975; 86:638–639. 5. Haibach H, Akhter JE, Muscato MS, et al. Acetaminophen overdose with fetal demise. Am J Clin Pathol 1984; 82:240–242. 6. Merlob P, Litwin A, Mor N. Possible association between acetazolamide administration during pregnancy and metabolic disorders in the newborn. Eur J Obstet Gynecol Reprod Biol 1990; 35:85–88. 7. Kemball ML, McIver C, Milnar RDG, et al. Neonatal hypoglycaemia in infants with diabetic mothers given sulphonylurea drugs in pregnancy. Arch Dis Chil 1970; 45:696–701. 8. Harris EL. Adverse reactions to oral antidiabetic agents. BMJ 1971; 3:29–30. 9. Liggins GC, Vaughan GS. Intravenous infusion of salbutamol in the management of premature labor. J Obstet Gynaecol Br Commonw 1973; 80:29–33. 10. Korda AR, Lynerum RC, Jones WR. The treatment of premature labor with intravenous administered salbutamol. Med J Austr 1974; 1:744–746. 11. Hastwell G. Salbutamol aerosol in premature labor. Lancet 1975; 2:1212–1213. 12. Eggers TR, Doyle LW, Pepperell RJ. Premature labour. Med J Austr 1979; 1:213–216. 13. Wager J, Fredholm B, Lunell NO, Persson B. Metabolic and circulatory effects of intravenous and oral salbutemol in late pregnancy in diabetic and non-diabetic women. Acta Obstet Gynecol Scand 1982; 108(suppl):41–46. 14. Baker ER, Flanagan MF. Fetal atrial flutter associated with maternal beta-sympathomimetic drug exposure. Obstet Gynecol 1997; 89:861. 15. Hastwell GB, Halloway CP, Taylor TLY. A study of 208 patients in premature labor treated with orally administered salbutamol. Med J Austr 1978; 1:465–469. 16. Thomas DJB, Dove AF, Alberti KGMM. Metabolic effects of salbutamol infusion during premature labor. Br J Obstet Gynaecol 1977; 84:497–499.

Direct Drug Toxicity to the Fetus

297

17. Wager J, Lunell NO, Nadal M, Ostman J. Glucose tolerance following oral salbutamol treatment in late pregnancy. Acta Obstet Gynecol Scand 1981; 60:291–294. 18. Lunell NO, Joelsson I, Larsson A, Parsson B. The immediate effect of β-adrenergic agonist (salbutamol) on carbohydrate and lipid metabolism during the third trimester of pregnancy. Acta Obstet Gynecol Scand 1977; 56:475–478. 19. Procianoy RS, Pinheiro CEA. Neonatal hyperinsulinism after short-term maternal beta sympathomimetic therapy. J Pediatr 1982; 101:612–614. 20. Barnett AH, Stubbs SM, Mander AM. Management of premature labour in diabetic pregnancy. Diabetologia 1980; 18:365–368. 21. Wager J, Fredholm BB, Lunell NO, Persson B. Metabolic and circulatory effects of oral salbutamol in the third trimester of pregnancy in diabetic and nondiabetic women. Br J Obstet Gynaecol 1981; 88:352–361. 22. Desgranges MF, Moutquin JM, Peloquin A. Effects of maternal oral salbutamol therapy on neonatal endocrine status at birth. Obstet Gynecol 1987; 69:582–584. 23. Redfern N, Bower S, Bullock RE, Hull CJ. Alfentanil for caesaean section complicated by severe aortic stenosis: a case report. Br J Anaesth 1987; 59:1309–1312. 24. Heytens L, Cammu H, Camu F. Extradural analgesia during labour using alfentanil. Br J Anaesth 1987; 59:331–337. 25. Smith EJ, Nagyfy SF. A report on comparative studies of new drugs used for obstetrical analgesia. Am J Obstet Gynecol 1949; 58:695–702. 26. Hapke FB, Barnes AC. The obstetric use and effect on fetal respiration of nisentil. Am J Obstet Gynecol 1949; 58:799–801. 27. Kane WM. The results of nisentil in 1,000 obstetrical cases. Am J Obstet Gynecol 1953; 65:1020–1026. 28. Backner DD, Foldes FF, Gordon EH. The combined use of alphaprodine (nisentil) hydrochloride and levallorphan (lorifan) tartate for analgesia in obstetrics. Am J Obstet Gynecol 1957; 74:271–282. 29. Gillan JS, Hunter GW, Dorner CB, Thompson GR. Meperidine hydrochloride and alphaprodine as obstetric analgesic agents; a double blind study. Am J Obstet Gynecol 1958; 75: 1105–1110. 30. Roberts H, Kuck MAC. Use of alphaprodine and levallorphan during labor. Can Med Assoc J 1960; 83:1088–1093. 31. Burnett RG, White CA. Alphaprodine for continuous intravenous obstetric analgesia. Obstet Gynecol 1966; 27:472–477. 32. Anthinarayanan PR, Mangurthen HH. Unusually prolonged action of maternal alphaprodine causing fetal depression. Q Pediatr Bull 1977; 3:14–16. 33. Gray JH, Cudmore DW, Luther ER, et al. Sinusoidal fetal heart rate pattern associated with alphaprodine administration. Obstet Gynecol 1978; 52:678–681. 34. Corby DG, Schulman I. The effects of antenatal drug administration on aggregation of platelets of newborn infants. J Pediatr 1971; 79:307–313. 35. McNall PG, Jafarnia MR. Management of myasthenia gravis in the obstetrical patient. Am J Obstet Gynecol 1965; 92:518–525. 36. Plauche WC. Myasthenia gravis in pregnancy: an update. Am J Obstet Gynecol 1979; 135: 691–697. 37. Iffy L, Ansell JS, Bryant FS, Hermann WL. Nonadrenal female pseudohermaphroditism: an unusual case of masculinization. Obstet Gynecol 1965; 26:59–65. 38. Marek J, Horky K. Aminoglutethimide administration in pregnancy. Lancet 1970; 2:1312– 1313. 39. Candlepergher G, Buchberger R, Suzzi GL, Padrini R. Trans-placental passage of aniodarone: electrocardiographic and pharmacologic evidence in the newborn. G Ital Cardiol 1982; 12:79–82. 40. McKanna WJ, Harris L, Rowland E, et al. Amiodarone therapy during pregnancy. Am J Cardiol 1983; 51:1231–1233.

298

Diav-Citrin and Koren

41. Robson DJ, Jeeva Raj MV, Storey GAC, Holt DW. Use of amiodarone during pregnancy. Postgrad Med J 1985; 61:75–77. 42. Rey E, Bachrach LK, Burrow GN. Effects of amiodarone during pregnancy. Can Med Assoc J 1987; 136:959–960. 43. Widerhorn J, Bhandari AK, Bughi S, et al. Fetal and neonatal adverse effects profile of amiodarone treatment during pregnancy. Am Heart J 1991; 122:1162–1168. 44. Plomp TA, Vulsma T, de Vijlder JJM. Use of amiodarone during pregnancy. Eur J Obstet Gynecol Reprod Biol 1992; 43:201–207. 45. De Wolf D, De Schepper J, Verhaaren H, et al. Congenital hypothyroid goiter and amiodarone. Acta Paediatr Scand 1988; 77:616–618. 46. Laurent M, Betremieux P, Biron Y, LeHelloco A. Neonatal hypothyroidism after treatment by amiodarone during pregnancy (letter). Am J Cardiol 1987; 60:942. 47. Magee LA, Downar E, Sermer M, Pregnancy outcome after gestational exposure to amiodarone in Canada. Am J Obstet Gynecol 1995; 172:1307–1311. 48. De Catte L, De Wolf D, Smitz J, et al. Fetal hypothyroidism as a complication of amiodarone treatment for persistent fetal supraventricular tachycardia. Prenatal Diagn 1994; 14:762–765. 49. Goodlin RC, Kaiser IH. The effect of ammonium chloride induced maternal acidosis on the human fetus at term: I. pH, hemoglobin, blood gases. Am J Med Sci 1957; 233:666–674. 50. Kaiser IH, Goodlin RC. The effect of ammonium chloride induced maternal acidosis on the human fetus at term: II. Electrolytes. Am J Med Sci 1958; 235:549–554. 51. Larsson G. The amphetamine addicted mother and her child. Acta Paediatr Scand 1980; 278(suppl):7–24. 52. Little BB, Snell LM, Gilstrap LC III. Metamphetamine abuse during pregnancy: outcome and fetal effects. Obstet Gynecol 1988; 72:541–544. 53. Dixon SD, Bejar R. Echoencephalographic findings in neonates associated with maternal cocaine and methamphetamine use: incidence and clinical correlates. J Pediatr 1989; 115:770–778. 54. Dearlove JC, Betteridge T. Stillbirth due to intravenous amphetamine. BMJ 1992; 304:548. 55. Heim K, Alge A, Marth C. Anaphylactic reaction to ampicillin and severe complication in the fetus. Lancet 1991; 337:859. 56. Zierler S, Purohit D. Prenatal antihistamine exposure and retrolental fibroplasia. Am J Epidemiol 1986; 123:192–196. 57. Hoffhaver H, Dobbeck P. Untersuchungen uber die Plactapassage des kallikrein Inhibitors. Klin Wochenschr 1970; 48:183–184. 58. Okun DB, Groncy PK, Sieger L, Tanaka KR. Acute leukemia in pregnancy: transient neonatal myelosuppression after combination chemotherapy in the mother. Med Pediatr Oncol 1979; 7:315–319. 59. Khurshid M, Saleem M. Acute leukaemia in pregnancy. Lancet 1978; 2:534–535. 60. Karp GI, Von Oeyen P, Valone F, et al. Doxorubicin in pregnancy: Possible transplacental passage. Cancer Treat Dep 1983; 67:773–777. 61. Garnick MB. Doxorubicin in pregnancy: possible transplacental passage. Cancer Treat Rep 1983; 67:773–777. 62. Awidi AS, Tarawneh MS, Shubair KS, Issa AA, Dajani YF. Acute leukemia in pregnancy: report of five cases treated with a combination which included a low dose of Adriamycin. Eur J Cancer Clin Oncol 1983; 19:881–884. 63. Turchi JJ, Villasis C. Anthracyclines in the treatment of malignancy in pregnancy. Cancer 1988; 61:435–440. 64. Schleuning M, Clemm C. Chromosomal aberrations in a newborn whose mother received cytotoxic treatment during pregnancy. N Eng J Med 1987; 317:1666–1667. 65. Collins E, Turner G. Maternal effects of regular salicylate ingestion in pregnancy. Lancet 1975; 2:335–337. 66. Turner G, Collins E. Fetal effects of regular salicylate ingestion in pregnancy. Lancet 1975; 2:338–339.

Direct Drug Toxicity to the Fetus

299

67. Shapiro S, Monson RR, Kaufman DW, et al. Perinatal mortality in relation to aspirin taken during pregnancy. Lancet 1976; 1:1375–1376. 68. Arcilla RA, Thilenius OG, Ranniger K. Congestive heart failure from suspected ductal closure in utero. J Pediatr 1969; 75:74–78. 69. Bleyer WA, Breckenridge RJ. Studies on the detection of adverse drug reactions in the newborn: II. The effects of prenatal aspirin on newborn hemostasis. JAMA 1970; 213:2049– 2053. 70. Corby DG, Schulman I. The effects of antenatal drug administration on aggregation of platelets of newborn infants. J Pediatr 1971; 79:307–313. 71. Casteels-Van Daele M, de Gaetano G, Vermijlen J. More on the effects of antenatally adminitered aspirin on aggregation of platelets of neonates. J Pediatr 1972; 80:685–686. 72. Haslam RR, Ekert H, Gillam GL. Hemorrhage in a neonate possibly due to maternal ingestion of salicilate. J Pediatr 1974; 84:556–557. 73. Ekert H, Haslam RR. Maternal ingested salicylate as a cause of neonatal hemorrhage. (reply). J Pediatr 1974; 85:738. 74. Pearson H. Comparative effects of aspirin and acetaminophen on hemostasis. Pediatrics 1978; 62(suppl):962–969. 75. Haslam RR. Neonatal purpura secondary to maternal salicylism. J Pediatr 1975; 86:653. 76. Stuart MJ, Gross SJ, Elrad H, Graeber JE. Effects of acetylsalicylic acid ingestion on maternal and neonatal hemostasis. N Engl J Med 1982; 307:909–912. 77. Stuart MJ. Aspirin and maternal or neonatal hemostasis. N Engl J Med 1983; 308:281. 78. Rumack CM, Guggenheim MA, Rumack BH, et al. Neonatal intracranial hemorrhage and maternal use of aspirin. Obstet Gynecol 1981; 58(suppl):52S–56S. 79. Soller RW, Stander H. Maternal drug exposure and perinatal intracranial hemorrhage. Obstet Gynecol 1981; 58:735–737. 80. Corby DG. Editorial comment. Obstet Gynecol 1981; 58:737–740. 81. Jackson AV. Toxic effects of salicylate on the foetus and mother. J Pathol Bacteriol 1948; 60:587–593. 82. Aterman K, Holzbecker M, Ellenberger HA. Salicylate levels in a stillborn to a drug-addicted mother, with comments on pathology and analytical methodology. Clin Toxicol 1980; 16: 263–268. 83. Earle R Jr. Congenital Salicilate intoxication-report of a case. N Engl J Med 1961; 265: 1003–1004. 84. Lynd PA, Andeasen AC, Wyatt RJ. Intrauterine salicilate intoxication in a newborn: a case report. Clin Pediatr (Phila) 1976; 15:912–913. 85. Palmisano PA, Cassady G. Salicylate exposure in the perinate. JAMA 1969; 209:556– 558. 86. Dubois D, Peticolas J, Temperville B, Klepper A. Treatment with atenolol of hypertension in pregnancy. Drugs 1983; 25(suppl 2):215–218. 87. Butters L, Kennedy S, Rubin PC. Atenolol in essential hypertension during pregnancy. BMJ 1990; 301:587–589. 88. Lip GYH, Beevers M, Churchill D, et al. Effect of atenolol on birth weight. Am J Cardiol 1997; 79:1436–1438. 89. Al Kasab SM, Sabag T, Al Zaibag M. β-Adrenergic receptor blockade in the management of pregnant women with mitral stenosis. Am J Obstet Gynecol 1990; 163:37–40. 90. Lardoux H, Gerard J, Blazquez G, et al. Hypertension in pregnancy: evaluation of two beta blockers atenolol and labetolol. Eur Heart J 1983; 4(suppl G):35–40. 91. Tuimala R, Hartikainen Sorri AL. Randomised comparison of atenolol and pindolol for treatment of hypertension in pregnancy. Curr Ther Res 1988; 44:579–584. 92. Ingemansson I, Liedholm H, Montan S, et al. Fetal heart rate during treatment of maternal hypertension with beta adrenergic antagonists. Acta Obstet Gynecol Scand 1984; 118(suppl): 95–97.

300

Diav-Citrin and Koren

93. Woods DL, Morrell DF. Atenolol: side effects in a newborn infant. BMJ 1982; 285:691– 692. 94. Rubin PC, Butters L, Clark DM, et al. Placebo controlled trial of atenolol in treatment of pregnancy associated hypertension. Lancet 1983; 1:431–434. 95. Cote CJ, Meuwissen HJ, Povkering RJ. Effects on the neonate of prednisone and azathioprine administered to the mother during pregnancy. J Pediatr 1974; 85:324–328. 96. DeWitte DB, Buick MK, Cyran SE, Maisels MJ. Neonatal pancytopenia and severe combined immunodeficiency associated with antenatal administration of azathioprine and prednisone. J Pediatr 1984; 105:625–628. 97. Davison JM, Dellagrammatikas H, Parkin JM. Maternal azathioprine therapy and depressed haemopoiesis in the babies of renal allograft patients. Br J Obstet Gynaecol 1985; 92:233– 239. 98. Scott JR. Fetal growth retardation associated with maternal administration of immunosuppressive drugs. Am J Obstet Gynecol 1977; 128:668–676. 99. Pirson Y, Van Lierde M, Ghysen J, et al. Retardation of fetal growth in patients receiving immunosuppressive therapy. N Engl J Med 1985; 313:328. 100. Marushak A, Weber T, Bock J, et al. Pregnancy following kidney transplantation. Acta Obstet Gynecol Scand 1986; 65:557–559. 101. Davison JM, Lindheimer MD. Pregnancy in renal transplant recipients. J Reprod Med 1982; 27:613–621. 102. The Registration Committee of the European Dialysis and Transplant association. Successful pregnancies in women treated by dialysis and kidney transplantation. Br J Obstet Gynaecol 1980; 87:839–845. 103. Leb DE, Weisskopf B, Kanovitz BS. Chromosome abberations in a child of a kidney transplant recipient. Arch Intern Med 1971; 128:441–444. 104. Falterman CG, Richardson CJ. Small left colon syndrome associated with maternal ingestion of psychotropic drugs. J Pediatr 1980; 97:308–310. 105. Papageorgiou AN, Desgranges MF, Masson M, et al. The antenatal use of betamethasone in the prevention of respiratory distress syndrome: a controlled double blind study. Pediatrics 1979; 63:73–79. 106. Bielawski D, Hiatt IM, Hegyi T. Betamethasone induced leukemoid reaction in preterm infant. Lancet 1978; 1:218–219. 107. Ferguson JE, Hensleigh PA, Gill P. Effects of betamethasone on white blood cells in patients with premature rupture of membranes and preterm labor. Am J Obstet Gynecol 1984; 150: 439–441. 108. Wasserstrum N, Huhta JC, Mari G, et al. Betamethasone and the human fetal ductus arteriosus. Obstet Gynecol 1989; 74:897–900. 109. Dorr HG, Versmold HT, Sippell WG, et al. Antenatal betamethasone therapy: effects on maternal, fetal, and mineralocorticoids, glucocorticoids, and progestins. J Pediatr 1986; 108: 990–993. 110. Raffles A, Williams J, Costeloe K, Clark P. Transplacental effects of maternal cancer chemotherapy: case report. Br J Obstet Gynaecol 1989; 96:1099–1100. 111. Bornstein RS, Hungerford DA, Haller G, et al. Cytogenic effects of bleomycin therapy in man. Cancer Res 1971; 31:2004–2007. 112. Opitz JM, Grosse RF, Haneberg B. Congenital effects of bromism? Lancet 1972; 1:91–92. 113. Rossiter EJR, Rendel Short TJ. Congenital effects of bromism? Lancet 1972; 2:705. 114. Finken RL, Robertson WO. Transplacental bromism. Am J Dis Child 1963; 106:224–226. 115. Mangurten HH, Ban R. Neonatal hypotonia secondary to transplacental bromism. J Pediatr 1974; 85:426–428. 116. Pleasure JR, Blackburn MG. Neonatal bromide intoxication: prenatal ingestion of a large quantity of bromides with transplacental accumulation in the fetus. Pediatrics 1975; 55:503– 506.

Direct Drug Toxicity to the Fetus

301

117. Nicholson HO. Cytotoxic drugs in pregnancy: review of reported cases. J Obstet Gynaecol Br Commonw 1968; 75:307–312. 118. Boros SJ, Reynolds JW. Intrauterine growth retardation following third trimester exposure to busulfan. Am J Obstet Gynecol 1977; 129:111–112. 119. Gebhart E, Schwanitz G, Hartwich G. Chromosomal abberations during busulfan therapy. Dtsch Med Wochenschr 1974; 99:52–56. 120. Maduska AL, Hajghassemali M. A double blind comparison of butorphanol and meperidine in labour: maternal pain relief and effect on the newborn. Can Anaesth Soc J 1978; 25:398– 404. 121. Pittman KA, Smyth RD, Losada M, et al. Human perinatal distribution of butorphanol. Am J Obstet Gynecol 1980; 138:797–800. 122. Quilligan EJ, Keegan KA, Donahue MJ. Double blind comparison of intravenously injected butorphanol and meperidine in parturients. Int J Gynaecol Obstet 1980; 18:363–367. 123. Angel JL, Knuppel RA, Lake M. Sinusoidal fetal heart rate pattern associated with intravenous butorphanol administration: a case report. Am J Obstet Gynecol 1984; 149:465– 467. 124. Hatjis CG, Meis PJ. Sinusoidal fetal heart rate pattern associated with intravenous butorphanol administration. Obstet Gynecol 1986; 67:377–380. 125. Mau G, Netter P. Kaffee- und Alkoholkonsum-Riskofaktoren in der Schwangerschaft? Geburtsh Frauenheilkd 1974; 34:1018–1022. 126. Hadeed A, Siegel S. Newborn cardiac arrhythmias associated with maternal caffeine use during pregnancy. Clin Pediatr 1993; 32:45–47. 127. Devoe LD, Murray C, Youssif A, Arnaud M. Maternal caffeine consumption and fetal behavior in normal third trimester pregnancy. Am J Obstet Gynecol 1993; 168:1105–1112. 128. Kovarik J, Woloszczuk W, Linkesch W, Pavelka R. Calcitonin in pregnancy. Lancet 1980; 1:199–200. 129. Figgs J, Hamilton R, Homel S, McCabe J. Camphorated oil intoxication in pregnancy: report of a case. Obstet Gynecol 1965; 25:255–258. 130. Weiss J, Catalano P. Camphorated oil intoxication during pregnancy. Pediatrics 1973; 52: 713–714. 131. Blackman WB, Curry HB. Camphor poisoning: report of case occurring during pregnancy. J Fla Med Assoc 1957; 43:99. 132. Jacobziner H, Raybin HW. Camphor poisoning. Arch Pediatr 1962; 79:28. 133. Hanssens M, Keirse MJNC, Vankelecom F, Van Assche FA. Fetal and neonatal effects of treatment with angiotensin converting enzyme inhibitors in pregnancy. Obstet Gynecol 1991; 78:128–135. 134. Markestad T, Ulstein M, Strandjord RE, et al. Anticonvulsant drug therapy in human pregnancy: effects on serum concentrations of vitamin D metabolites in maternal and cord blood. Am J Obstet Gynecol 1984; 150:245–258. 135. Mountain KR, Hirsh J, Gallus AS. Neonatal coagulation defect due to anticonvulsant drug treatment in pregnancy. Lancet 1970; 1:265–268. 136. Cornelissen M, Steegers-Theunissen R, Kollee L, et al. Increased incidence of neonatal vitamin K deficiency resulting from maternal anticonvulsant therapy. Am J Obstet Gynecol 1993; 168:923–928. 137. Cornelissen M, Steegers-Theunissen R, Kollee L, et al. Supplementation of vitamin K in pregnant women receiving anticonvulsant therapy prevents neonatal vitamin K deficiency. Am J Obstet Gynecol 1993; 168:884–888. 138. Frey B, Schuigr G, Musy JP. Transient cholestatic hepatitis in a neonate associated with carbamazepine exposure during pregnancy and breast feeding. Eur J Pediatr 1990; 150:136– 138. 139. Lawler SD, Lele KP. Chromosomal damage induced by chlorambucil and lymphocytic leukemia. Scand J Haematol 1972; 9:603–612.

302

Diav-Citrin and Koren

140. Westin J. Chromosome abnormalities after chlorambucil therapy of polycythemia vera. Scand J Haematol 1976; 17:197–204. 141. Catovsky D, Galton DAG. Myelomonocytic leukaemia supervening on chronic lymphocytic leukemia. Lancet 1971; 1:478–479. 142. Rosner R. Acute leukemia as a delayed consequence of cancer chemotherapy. Cancer 1976; 37:1033–1036. 143. Reimer RR, Hover R, Fraumeni JF, Young RC. Acute leukemia after alkylating agent therapy of ovarian cancer. N Engl J Med 1977; 297:177–181. 144. Oberheuser F. Praktische Erfahrungen mit Medikamenten in der Schwangerschaft. Therapiewoche 1971; 31:2200. As reported in Manten A. Antibiotic drugs. In: Dukes MNG, ed. Meyler’s Side Effects of Drugs, Vol VIII. New York: American Elsevier, 1975, p. 604. 145. Stirrat GM, Edinston PT, Berry DJ. Transplacental passage of chlordiazepoxide. Br Med J 1974; 2:729. 146. Hart CW, Naunton RF. The ototoxicity of chloroquine phosphate. Arch Otolarygol 1964; 80:407–412. 147. Senior B, Slone D, Shapiro S, et al. Benzothiadiazides and neonatal hypoglycaemia. Lancet 1976; 2:377. 148. Gray MJ. Use and abuse of thiazides in pregnancy. Clin Obstet Gynecol 1968; 11:568–578. 149. Menzies DN. Controlled trial of chlorothiazide in treatment of early pre-eclampsia. BMJ 1964; 1:739–742. 150. Harley JD, Robin H, Robertson SEJ. Thiazide induced neonatal haemolysis? BMJ 1964; 1: 696–697. 151. Rodriguez SU, Leikin SL, Hiller MC. Neonatal thrombocytopenia associated with ante-partum administration of thiazide drugs. N Engl J Med 1964; 270:881–884. 152. Leikin SL. Thiazide and neonatal thrombocytopenia. N Engl J Med 1964; 271:161. 153. Prescott LF. Neonatal thrombocytopenia and thiazide drugs. BMJ 1964; 1:1438. 154. Jones JE, Reed JF Jr. Renal vein thrombosis and thrombocytopenia in the newborn infant. J Pediatr 1965; 67:681–682. 155. Karpatkin S, Strick N, Karpatkin MB, Siskind GW. Cumulative experience in the detection of antiplatelet antibody in 234 patients with idiopathic thrombocytopenic purpura, systemic lupus erythematosus and other clinical disorders. Am J Med 1972; 52:776–785. 156. Pritchard JA, Walley PJ. Severe hypokalemia due to prolonged administration of chlorothiazide during pregnancy. Am J Obstet Gynecol 1961; 81:1241. 157. Alstatt LB. Transplacental hyponatremia in the newborn infant. J Pediatr 1965; 66:985–988. 158. Anderson GG, Hanson TM. Chronic fetal bradycardia: possible association with hypokalemia. Obstet Gynecol 1974; 44:896–898. 159. Minkowitz S, Soloway HB, Hall JE, Yermakov V. Fatal hemorrhagic pancreatitis following chlorothiazide administration in pregnancy. Obstet Gynecol 1964; 24:337–342. 160. Hammond JE, Toseland PA. Placental transfer of chlorpromazine. Arch Dis Child 1970; 45: 139–140. 161. Hill RM, Desmond MM, Kay JL. Extrapyramidal dysfunction in newborn infant of a schizophrenic mother. J Pediatr 1966; 69:589–595. 162. Ayd FJ Jr, ed. Phenothiazine therapy during pregnancy-effects on the newborn infant. Int Drug Ther Newslett 1968; 3:39–40. 163. Tamer A, McKay R, Arias D, et al. Phenothiazine-induced extrapyramidal dysfunction in the neonate. J Pediatr 1969; 75:479–480. 164. Levy W, Wisniewski K. Chlorpromazine causing extrapyramidal dysfunction in newborn infant of psychotic mother. NY State J Med 1974; 74:684–685. 165. O’Connor M, Johnson GH, James DI. Intrauterine effect of phenothiazines. Med J Aust 1981; 1:416–417. 166. Zucker P, Simon G. Prolonged symptomatic neonatal hypoglycemia associated with maternal chlorpropamide therapy. Pediatrics 1968; 42:824–825.

Direct Drug Toxicity to the Fetus

303

167. Kemball ML, McIver C, Milnar RDG, et al. Neonatal hypoglycemia in infants of diabetic mothers given sulphonylurea drugs in pregnancy. Arch Dis Child 1970; 45:696–701. 168. Harris EL. Adverse reactions to oral antidiabetic agents. BMJ 1971; 3:29–30. 169. Piacquadio K, Hollingworth DR, Murphy H. Effects of in-utero exposure to oral hypoglycemic drugs. Lancet 1991; 338:866–869. 170. Sadler LC, Lane M, North R. Severe fetal intracranial haemorrhage during treatment with cholestyramine for intrahepatic cholestasis of pregnancy. Br J Obstet Gynecol 1995; 102: 169–170. 171. Glade G, Saccar CL, Pereira GR. Cimetidine in pregnancy: apparent transient liver impairment in the newborn. Am J Dis Child 1980; 134:87–88. 172. Karat AB. Long-term follow up of clofazimine (Lamprene) in the management of reactive phases of leprosy. Lepr Rev 1975; 46(suppl):105–109. 173. Farb H, West DP, Pedvis-Leftick A. Clofazimine in pregnancy complicated by leprosy. Obstet Gynecol 1982; 59:122–123. 174. Fisher JB, Edgren BE, Mammel MC, Coleman JM. Neonatal apnea associated with maternal clonazepam therapy: a case report. Obstet Gynecol 1985; 66(suppl):34S–35S. 175. Bonica JJ. Principles and Practice of Obstetric Analgesia and Anesthesia. Philadelphia, PA: FA Davis, 1967, p 245. 176. Hall JG, Pauli RM, Wilson KM. Maternal and fetal sequelae of anticoagulation during pregnancy. Am J Med 1980; 68:122–140. 177. Pizzuto J, Aviles A, Noriega L, et al. Treatment of acute leukemia during pregnancy: presentation of nine cases. Cancer Treat Rep 1980; 64:697–683. 178. Klintmalm G, Althoff P, Appleby G, Segerbrandt E. Renal function in a newborn baby delivered of a renal transplant patient taking cyclosporine. Transplantation 1984; 38:198–199. 179. Burrows DA, O’Neil TJ, Sorrells TL. Successful twin pregnancy after renal transplant maintained on cyclosporine A immunosuppression. Obstet Gynecol 1988: 72:459–461. 180. Briggs GG, Freeman Rk, Yaffe SJ. Drugs in Pregnancy and Lactation, 5th ed. Baltimore: Williams & Wilkins, 1998. 181. Lau RJ, Scott JR. Pregnancy folowing renal transplantation. Clin Obstet Gynecol 1985; 28: 339–350. 182. Maurer LH, Forcier RJ, McIntyre OR, Benirschke K. Fetal group C trisomy after cytosine arabinoside and thioguanine. Ann Intern Med 1871; 75:809–810. 183. Duck SC, Katayama KP. Danazol may cause female pseudohermaphroditism. Fertil Steril 1981; 35:230–231. 184. Castro-Magana M, Cheruvanky T, Collipp PJ, et al. Transient adrenogenital syndrome due to exposure to danazol in utero. Am J Dis Child 1981; 135:1032–1034. 185. Peress MR, Kreutner AK, Mathur RS, Williamson HO. Female pseudohermaphroditism with somatic chromosomal anomaly in association with in utero exposure to danazol. Am J Obstet Gynecol 1982; 142:708–709. 186. Schwartz RP. Ambiguous genitalia in a term female infant due to exposure to danazol in utero. Am J Dis Child 1982; 136:474. 187. Wentz AC. Adverse effects of danazol in pregnancy. Ann Intern Med 1982; 96:672–673. 188. Shaw RW, Farquar JW. Female pseudohermaphroditism associated with danazol exposure in utero: case report. Br J Obstet Gynaecol 1984; 91:386–389. 189. Rosa FW. Virilization of the female fetus with maternal danazol exposure. Am J Obstet Gynecol 1984; 149:99–100. 190. Kingsbury AC. Danazol and fetal masculinization: a warning. Med J Aust 1985; 143:410– 411. 191. Quagliarello J, Greco MA. Danazol and urogenital sinus formation in pregnancy. Fertil Steril 1985; 43:939–942. 192. Brunskill PJ. The effects of fetal exposure to danazol. Br J Obstet Gynaecol 1992; 99:212– 215.

304

Diav-Citrin and Koren

193. Hocking DR. Neonatal haemolytic disease due to dapsone. Med J Aust 1968; 1:1130–1131. 194. Thornton YS, Bowe ET. Neonatal hyperbilirubinemia after treatment of maternal leprosy. South Med J 1989; 82:668. 195. Gililland J, Weinstein L. The effects of cancer chemotherapeutic agents on the developing fetus. Obstet Gynecol Surv 1983; 38:6–13. 196. Colbert N, Najman A, Gorin NC, Blum F. Acute leukaemia during pregnancy: favourable course of pregnancy in two patients treated with cytosine arabinoside and anthracyclines. Nouv Presse Med 1980; 9:175–178. 197. Rayburn WF, Donn SM, Wulf ME. Iron overdose during pregnancy: successful therapy with deferoxamine. Am J Obstet Gynecol 1983; 147:717–718. 198. Otero L, Conlon C, Reynolds P, et al. Neonatal leukocytosis associated with prenatal administration of dexamethasone. Pediatrics 1981; 68:778–780. 199. Anday EK, Harris MC. Leukemoid reaction associated with antenatal dexamethasone administration. J Pediatr 1982; 101:614–616. 200. Rodesch F, Camus M, Ermans AM, et al. Adverse effect of amniofetography on fetal thyroid function. Am J Obstet Gynecol 1976; 126:723–726. 201. Scanlon JW. Effect of benzodiazepines in neonates. N Engl J Med 1975; 292:649. 202. Gillberg C. ‘‘Floppy infant syndrome’’ and maternal diazepam. Lancet 1977; 2:244. 203. Haram K. ‘‘Floppy infant syndrome’’ and maternal diazepam. Lancet 1977; 2:612–613. 204. Speight AN. Floppy-infant syndrome and maternal diazepam and/or nitrazepam. Lancet 1977; 1:878. 205. Cree JE, Meyer J, Haily DM. Diazepam in labour: its metabolism and effect on the clinical condition and thermogenesis of the newborn. BMJ 1973; 4:251–255. 206. McAllister CB. Placental transfer and neonatal effects of diazepam when administered to women just before delivery. Br J Anaesth 1980; 52:423–427. 207. Owen JR, Irani SF, Blair AW. Effect of diazepam administered to mothers during labour on temperature regulation of the neonate. Arch Dis Child 1972; 47:107–110. 208. Scher J, Hailey DM, Beard RW. The effects of diazepam on the fetus. J Obstet Gynaecol Br Commonw 1972; 79:635–638. 209. Van Geijn HP, Jongsma HW, Doesberg WH, et al. The effect of diazepam administration during pregnancy or labor on the heart rate variability of the newborn infant. Eur J Obstet Gynaecol Reprod Biol 1980; 10:187–201. 210. Birger M, Homberg R, Insler V. Clinical evaluation of fetal movements. Int J Gynaecol Obstet 1980; 183:377–382. 211. Morris JA, Arce JJ, Hamilton CJ, et al. The management of severe preeclampsia and eclampsia with intravenous diazoxide. Obstet Gynecol 1977; 49:675–680. 212. Michael CA. Intravenous diazoxide in the treatment of severe preeclamptic toxaemia and eclampsia. Aust N Z J Obstet Gynaecol 1973; 13:143–146. 213. Neuman J, Weiss B, Rabello Y, et al. Diazoxide for acute control of severe hypertension complicating pregnancy: a pilot study. Obstet Gynecol 1979; 53(suppl):50S–55S. 214. Milsap RL, Auld PAM. Neonatal hyperglycemia following maternal diazoxide administration. JAMA 1980; 243:144–145. 215. Milner RDG, Chouksey SK. Effects of fetal exposure to diazoxide in man. Arch Dis Child 1972; 47:537–543. 216. Jefferies JA, Robboy SJ, O’Brien PC, et al. Structural anomalies of the cervix and vagina in women enrolled in the Diethylstilbestrol Adenosis (DESAD) Project. Am J Obstet Gynecol 1984; 148:59–66. 217. Peress MR, Tsai CC, Mathur RS, Williamson HO. Hirsutism and menstrual patterns in women exposed to diethylstilbestrol in utero. Am J Obstet Gynecol 1982; 144:135–140. 218. Vessey MP, Fairweather DVI, Norman-Smith B, Buckley J. A randomized double-blind controlled trial of the value of stilboestrol therapy in pregnancy: long-term follow-up of mothers and their offspring. Br J Obstet Gynaecol 1983; 90:1007–1017.

Direct Drug Toxicity to the Fetus

305

219. Sherman JL Jr, Locke RV. Transplacental neonatal digitalis intoxication. Am J Cardiol 1960; 6:834–837. 220. Ruch WA, Ruch RM. A preliminary report on dihydrocodeine-scopolamine in obstetrics. Am J Obstet Gynecol 1957; 74:1125–1127. 221. Myers JD. A preliminary clinical evaluation of dihydrocodeine bitartrate in normal parturition. Am J Obstet Gynecol 1958; 75:1096–1100. 222. Hara GS, Carter RP, Kranz KE. Dramamine in labor: potential boon or a possible bomb? J Kans Med Soc 1980; 81:134–136, 155. 223. Scindler AM. Isolated neonatal hypomagnesaemia associated with maternal overuse of stool softener. Lancet 1984; 2:822. 224. Wright RG, Shnider SM, Levinson G, et al. The effect of maternal administration of ephedrine on fetal heart rate and variability. Obstet Gynecol 1981; 57:734–738. 225. Antoine C, Young BK. Fetal lactic acidosis with epidural anesthesia. Am J Obstet Gynecol 1982; 142:55–59. 226. Datta S, Alper MH, Ostheimer GW, Weiss JB. Method of ephedrine administration and nausea and hypotension during spinal anesthesia for cesarean section. Anesthesiology 1982; 56:68–70. 227. Entman SS, Moise KJ. Anaphylaxis in pregnancy. South Med J 1984; 77:402. 228. Braga J, Marques R, Branco A, et al. Maternal and perinatal implications of the use of human recombinant erythropoeitin. Acta Obstet Gynecol Scand 1996; 75:449–453. 229. Au KL, Woo JSK, Wong VCW. Intrauterine death from ergotamine overdosage. Eur J Obstet Gynecol Reprod Biol 1985; 19:313–315. 230. de Groot ANJA, van Dongen PWJ, van Roosmalen J, Eskes TKAB. Ergotamine-induced fetal stress: review of side effects of ergot alkaloids during pregnancy. Eur J Obstet Gynecol Reprod Biol 1993; 51:73–77. 231. Ducey JP, Knape KG. Maternal esmolol administration resulting in fetal distress and cesarean section in a term pregnancy. Anesthesiology 1992; 77:829–832. 232. Gilson GJ, Knieriem KJ, Smith JF, et al. Short-acting beta-adrenergic blockade and the fetus: a case report. J Reprod Med 1992; 37:277–279. 233. Fairly CJ, Clarke JT. Use of esmolol in a parturient with hpertrophic obstructive cardiomyopathy. Br J Anaesth 1995; 75:801–804. 234. Jones HC. Intrauterine ototoxicity: a case report and review of literature. J Natl Med Assoc 1973; 65:201–203. 235. Speidel BD, Meadow SR. Epilepsy, anticonvulsants and congenital malformations. Drugs 1974; 8:354–365. 236. Buller RE, Darrow V, Manetta A, et al. Conservative surgical management of dysgerminoma concomitant with pregnancy. Obstet Gynecol 1992; 79:887–890. 237. Murray NA, Acolet D, Deane M, et al. Fetal marrow suppression after maternal chemotherapy for leukemia. Arch Dis Child 1994; 71:F209–F210. 238. Jager K, Schiller F, Stech P. Congenital ichthyosiforme erythroderma, pregnancy under aromatic retinoid treatment. Hautarzt 1985; 36:150–153. 239. Carrie LES, O’Sullivan GM, Seegobin R. Epidural fentanyl in labour. Anaesthesia 1981; 36: 965–969. 240. Rayburn W, Rathke A, Leuschen MP, et al. Fentanyl citrate analgesia during labor. Am J Obstet Gynecol 1989; 161:202–206. 241. Johnson ES, Colley PS. Effects of nitrous oxide and fentanyl anesthesia on fetal heart-rate variability intra- and postoperatively. Anesthesiology 1980; 52:429–430. 242. Van Gelder-Hasker MR, de Jong CLD, de Vries JIP, van Geijn HP. The effect of flecainide acetate on fetal heart rate variability: a case report. Obstet Gynecol 1995; 86:667–669. 243. Vanderhal AL, Cocjin J, Santulli TV, et al. Conjugated hyperbilirubinemia in a newborn infant after maternal (transplacental) treatment with flecainide acetate for fetal tachycardia and fetal hydrops. J Pediatr 1995; 126:988–990.

306

Diav-Citrin and Koren

244. Stadler HE, Knowles J. Fluorouracil in pregnancy: effect on the neonate. JAMA 1971; 217: 214–215. 245. Cleary MF. Fluphenazine decanoate during pregnancy. Am J Psychiatry 1977; 134:815–816. 246. Nath SP, Miller DA, Muraskas JK. Severe rinorrhea and respiratory distress in a neonate exposed to fluphenazine hydrochloride prenatally. Ann Pharmacother 1996; 30:35–37. 247. Lamotrigine Pregnancy Registry. Interim Report. 1 September 1992 through 30 September 1996. Glaxo Wellcome Inc, 1997. 248. Coetzee EJ, Jackson WPU. Pregnancy in established non-insulin-dependent diabetics: a fiveand-a-half year study at Groote Schuur Hospital. S Afr Med J 1980; 58:795–802. 249. Coetzee EJ, Jackson WPU. Oral hypoglycemics in the first trimester and fetal outcome. S Afr Med J 1984; 65:635–637. 250. Morris N. Hexamethonium in the treatment of pre-eclampsia and essential hypertension during pregnancy. Lancet 1953; 1:322–324. 251. Hallum JL, Hatchuel WLF. Congenital paralytic ileus in a premature baby as a complication of hexamethonium bromide therapy for toxemia of pregnancy. Arch Dis Child 1954; 29: 354–356. 252. Widerlov E, Karlman I, Storsater J. Hydralazine-induced neonatal thrombocytopenia. N Engl J Med 1980; 303:1235–1238. 253. Brazy JE, Grimm JK, Litte VA. Neonatal manifestations of severe maternal hypertension occurring before the thirty-sixth week of pregnancy. J Pediatr 1982; 100:265–271. 254. Sibai BM, Anderson GD. Pregnancy outcome of intensive therapy in severe hypertension in first trimester. Obstet Gynecol 1986; 67:517–522. 255. Lodeiro JG, Feinstein SJ, Lodeiro SB. Fetal premature atrial contractions associated with hydralazine. Am J Obstet Gynecol 1989; 160:105–107. 256. Yemini M, Shoham (Schwartz) Z, Dgani R, et al. Lupus-like syndrome in a mother and newborn following administration of hydralazine: a case report. Eur J Obstet Gynecol Reprod Biol 1989; 30:193–197. 257. Kirshon S, Wasserstrum N, Cotton DB. Should continuous hydralazine infusions be utilized in severe pregnancy-induced hypertension? Am J Perinatol 1991; 8:206–208. 258. Wilkins L. Masculinization of female fetus due to use of orally given progestins. JAMA 1960; 172:1028–1032. 259. Wilkins L, Jones HW, Holman GH, Stempfel RS Jr. Masculinization of the female fetus associated with administration of oral and intramuscular progestins during gestation: nonadrenal female pseudohermaphrodism. J Clin Endocrinol Metab 1958; 68:559–585. 260. Dayan E, Rosa FW. Fetal ambiguous genitalia associated with sex hormone use early in pregnancy. Food and Drug Administration, Division of Drug Experience. ADR Highlights 1981; 1–14. 261. Hickok DE, Hollenbach KA, Reilley SF, Nyberg DA. The association between decreased amniotic fluid volume and treatment with nonsteroidal anti-inflammatory agents for preterm labor. Am J Obstet Gynecol 1989; 160:1525–1531. 262. Wiggins DA, Elliott JP. Oligohydramnios in each sac of a triplet gestation caused by Motrin—fulfilling Kock’s postulates. Am J Obstet Gynecol 1990; 162:460–461. 263. Hendricks SK, Smith JR, Moore DE, Brown ZA. Oligohydramnios associated with prostaglandin synthetase inhibitors in preterm labour. Br J Obstet Gynaecol 1990; 97:312–316. 264. Hennessy MD, Livingston EC, Papagianos J, Killam AP. The incidence of ductal constriction and oligohydramnios during tocolytic therapy with ibuprofen (abstr). Am J Obstet Gynecol 1992; 166:324. 265. Reynoso EE, Huerta F. Acute leukemia and pregnancy—fatal fetal outcome after exposure to idarubicin during the second trimester. Acta Oncol 1994; 33:703–716. 266. Morales WJ, Smith SG, Angel JL, et al. Efficacy and safety of indomethacin compared versus ritodrine in the management of preterm labor: a randomized study. Obstet Gynecol 1989; 74:567–572.

Direct Drug Toxicity to the Fetus

307

267. De Wit W, Van Mourik I, Wiesenhaan PF. Prolonged maternal indomethacin therapy associated with oligohydramnios: case reports. Br J Obstet Gynaecol 1988; 95:303–305. 268. Goldenberg RL, Davis RO, Baker RC. Indomethacin-induced oligohydramnios. Am J Obstet Gynecol 1989; 160:1196–1197. 269. Kirshon B, Moise KJ Jr, Wasserstrum N, et al. Influence of short-term indomethacin therapy on fetal urine output. Obstet Gynecol 1988; 72:51–53. 270. Heijden AJ, Provost AP, Nauta J, et al. Renal functional impairment in preterm neonates related to intrauterine indomethacin exposure. Pediatr Res 1988; 24:644–648. 271. Atad J, David A, Moise J, Abramovici H. Classification of threatened premature labor related to treatment with a prostastaglandin inhibitor: indomethacin. Biol Neonate 1980; 37:291– 296. 272. Sureau C, Piovani P. Clinical study of indomethacin for prevention of prematurity. Eur J Obstet Gynecol Reprod Biol 1983; 46:400–402. 273. Van Kets H, Thiery M, Derom R, et al. Perinatal hazards of chronic antenatal tocolysis with indomethacin. Prostaglandins 1979; 18:893–907. 274. Van Kets H, Thiery M, Derom R, et al. Prostaglandin synthase inhibitors in preterm labor. Lancet 1980; 2:693. 275. Leonardi MR, Hankins GDV. What’s new in tocolytics. Clin Perinatol 1992; 19:367–384. 276. Higby K, Xenakis EM-J, Pauerstein CJ. Do tocolytic agents stop preterm labor? A critical and comprehensive review of efficacy and safety. Am J Obstet Gynecol 1993; 168:1247– 1259. 277. Levin DL. Effects of inhibition of prostaglandin synthesis on fetal development, oxygenation, and the fetal circulation. Semin Perinatol 1980; 4:35–44. 278. Csaba IF, Sulyok E, Ertl T. Relationship of maternal treatment with indomethacin to persistence of fetal circulation syndrome. J Pediatr 1978; 92:484. 279. Levin DL, Fixler DE, Morriss FC, Tyson J. Morphologic analysis of the pulmonary vascular bed in infants exposed in utero to prostaglandin synthetase inhibitors. J Pediatr 1978; 92: 478–483. 280. Rubaltelli FF, Chiozza ML, Zanardo V, Cantarutti F. Effect on neonate of maternal treatment with indomethacin. J Pediatr 1979; 94:161. 281. Manchester D, Margolis HS, Sheldon RE. Possible association between maternal indomethacin therapy and primary pulmonary hypertension of the newborn. Am J Obstet Gynecol 1976; 126:467–469. 282. Besinger RE, Niebyl JR, Keyes WG, Johnson TRB. Randomized comparative trial of indomethacin and ritodrine for the long-term treatment of preterm labor. Am J Obstet Gynecol 1991; 164:981–988. 283. Demandt E, Legius E, Devlieger H, et al. Prenatal indomethacin toxicity in one member of monozygous twins: a case report. Eur J Obstet Gynecol Reprod Biol 1990; 35:267–269. 284. Goudie BM, Dossetor JFB. Effect on the fetus of indomethacin given to suppress labour. Lancet 1979; 2:1187–1188. 285. Mogilner BM, Ashkenazy M, Borenstein R, Lancet M. Hydrops fetalis caused by maternal indomethacin treatment. Acta Obstet Gynecol Scand 1982; 61:183–185. 286. Moise KJ Jr, Huhta JC, Sharif DS, et al. Indomethacin in the treatment of premature labor: effects on the fetal ductus arteriosus. N Engl J Med 1988; 319:327–331. 287. Van Den Veyver I, Moise K Jr, et al. The effect of gestational age and fetal indomethacin levels on the incidence of constriction of the fetal ductus arteriosus (abstr). Am J Obstet Gynecol 1993; 168:373. 288. Van Den Veyver IB, Moise K Jr, et al. The effect of gestational age and fetal indomethacin levels on the incidence of constriction of the fetal ductus arteriosus. Obstet Gynecol 1993; 82:500–503. 289. Moise KJ Jr. Effect of advancing gestational age on the frequency of fetal ductal constriction in association with maternal indomethacin use. Am J Obstet Gynecol 1993; 168:1350–1353.

308

Diav-Citrin and Koren

290. Evans DJ, Kofinas AD, King K. Intraoperative amniocentesis and indomethacin treatment in the management of an immature pregnancy with completely dialated cervix. Obstet Gynecol 1992; 79:881–882. 291. Eronen M, Pesonen E, Kurki T, et al. The effects of indomethacin and a β-sympathomimetic agent on the fetal ductus arteriosus during treatment of premature labor: a randomized doubleblind study. Am J Obstet Gynecol 1991; 164:141–146. 292. Hallak M, Reiter AA, Ayres NA, Moise KJ Jr. Indomethacin for preterm labor: fetal toxicity in a dizygotic twin gestation. Obstet Gynecol 1991; 78:911–913. 293. Rosemond RL, Boehm FH, Moreau G, Karmo H. Tricuspid regurgitation: a method of monitoring patients treated with indomethacin (abstr). Am J Obstet Gynecol 1992; 166: 336. 294. Bivins HA Jr, Newman RB, Fyfe DA, et al. Randomized comparative trial of indomethacin and terbutaline for the long term treatment of preterm labor (abstr). Am J Obstet Gynecol 1993; 168:375. 295. Mari G, Moise KJ Jr, Deter RL, et al. Doppler assessment of the renal blood flow velocity waveform during indomethacin therapy for preterm labor and polyhydramnios. Obstet Gynecol 1990; 75:199–201. 296. Kirshon B, Mari G, Moise KJ Jr, Wasserstrum N. Effect of indomethacin on the fetal ductus arteriosus during treatment of symptomatic polyhydramnios. J Reprod Med 1990; 35:529– 532. 297. Buderus S, Thomas B, Fahneestich H, Kowalewski S. Renal failure in two preterm infants: toxic effect of prenatal maternal indomethacin treatment? Br J Obstet Gynaecol 1993; 100: 97–98. 298. Itskovitz J, Abramovici H, Brandes JM. Oligohydramnion, meconium and perinatal death concurrent with indomethacin treatment in human pregnancy. J Reprod Med 1980; 24:137– 140. 299. Veersema D, de Jong PA, van Wijck JAM. Indomethacin and the fetal renal nonfunction syndrome. Eur J Obstet Gynecol Reprod Biol 1983; 16:113–121. 300. Vanhaesebrouck P, Thiery M, Leroy GJ, et al. Oligohydramnios, renal insufficiency, and ileal perforation in preterm infants after intrauterine exposure to indomethacin. J Pediatr 1988; 113:738–743. 301. Murray HG, Stone PR, Strand L, Flower J. Fetal pleural effusion following maternal indomethacin therapy. Br J Obstet Gynaecol 1993; 100:277–282. 302. Baerts W, Fetter WPF, Hop WCJ, et al. Cerebral lesions in preterm infants after tocolytic indomethacin. Dev Med Child Neurol 1990; 32:910–918. 303. Menon RK, Cohen RM, Sperling MA, et al. Transplacental passage of insulin in pregnant women with insulin-dependent diabetes mellitus: Its role in fetal mcrosomia. N Engl J Med 1990; 323:309–315. 304. Mehta PS, Mehta SJ, Virherr H. Congenital iodide goiter and hypothyroidism: a review. Obstet Gynecol Surv 1983; 38:237–247. 305. I’Allemand D, Gruters A, Heidemann P, Schurnbrand P. Iodine-induced alterations of thyroid function in newborn infants after prenatal and perinatal exposure to povidone iodide. J Pediatr 1983; 102:935–938. 306. Bachrach LK, Burrow GN, Gare DJ. Maternal-fetal absorption of povidone-iodide. J Pediatr 1984; 102:158–189. 307. Jacobson JM, Hankins GV, Young RL, Hauth JC. Changes in thyroid function and serum iodine levels after prepartum use of a povidone-iodine vaginal lubricant. J Reprod Med 1984; 29:98–100. 308. Danziger Y, Perzelan A, Mimouni M. Transient congenital hypothyroidism after topical iodine in pregnancy and lactation. Arch Dis Child 1987; 62:295–296. 309. Wolff J. Iodide goiter and the pharmacologic effects of excess iodide. Am J Med 1969; 47: 101–124.

Direct Drug Toxicity to the Fetus

309

310. Russell KP, Rose H, Starr P. The effects of radioactive iodine on maternal and fetal thyroid function during pregnancy. Surg Gynecol Obstet 1957; 104:560–564. 311. Ray EW, Sterling K, Gardner LI. Congenital cretinism associated with I 131 therapy of the mother. Am J Dis Child 1959; 98:506–507. 312. Hamill GC, Jarman JA, Wynne MD. Fetal effects of radioactive iodine therapy in a pregnant woman with thyroid cancer. Am J Obstet Gynecol 1961; 81:1018–1023. 313. Fisher WD, Voorhess ML, Gardner LI. Congenital hypothyroidism in infant following maternal 131 I therapy. J Pediatr 1963; 62:132–146. 314. Green HG, Garies FJ, Shepard TH, Kelley VC. Cretinism associated with maternal sodium iodide I 131 therapy during pregnancy. Am J Dis Child 1971; 122:247–249. 315. Jafek BW, Small R, Lillian DL. Congenital radioactive iodine–induced stridor and hypothyroidism. Arch Otolaryngol 1974; 99:369–371. 316. Exss R, Graewe B. Congenital athyroidism in the newborn infant from intra-uterine radioiodine action. Biol Neonate 1974; 24:289–291. 317. Richards GE, Brewer ED, Conley SB, Saldana LR. Combined hypothyroidism and hypoparathyoidism in an infant after maternal 131 I administration. J Pediatr 1981; 99:141–143. 318. Eggermont E, Logghe N, Van De Casseye W, et al. Haemorrhagic disease of the newborn in the offspring of rifampicin and isoniazid treated mothers. Acta Paediatr Belg 1976; 29: 87–90. 319. DeSimone CA, Leighton BL, Norris MC, et al. The chronotropic effect of isoproterenol is reduced in term pregnant women. Anesthesiology 1988; 69:626–628. 320. Brazy JE, Little V, Grimm J, Pupkin M. Risk:benefit considerations for the use of isoxsuprine in the treatment of premature labor. Obstet Gynecol 1981; 58:297–303. 321. Brazy JE, Pupkin MJ. Effects of maternal isoxsuprine administration on preterm infants. J Pediatr 1979; 94:444–448. 322. Brazy JE, Little V, Grimm J. Isoxsuprine in the perinatal period: II. Relationships between neonatal symptoms, drug exposure, and drug concentration at the time of birth. J Pediatr 1981; 98:146–151. 323. Gemelli M, De Luca F, Manganaro R, et al. Transient electrocardiographic changes suggesting myocardial ischemia in newborn infants following tocolysis with beta-sympathomimetics. Eur J Pediatr 1990; 149:730–733. 324. Nishimura H, Tanimura T. Clinical Aspects of Teratogenicity of Drugs. Amsterdam: Experta Medica, 1976, pp 131–145. 325. Little B, Chang T, Chucot L, et al. Study of ketamine as an obstetric anesthetic agent. Am J Obstet Gynecol 1972; 113:247–260. 326. Meer FM, Downing JW, Coleman AJ. An intravenous method of anaesthesia for caesarean section: Part II. Ketamine. Br J Anaesth 1973; 45:191–196. 327. Galbert MW, Gardner AE. Ketamine for obstetrical anesthesia. Anesth Analg 1973; 52:926– 930. 328. Corssen G. Ketamine in obstetric anesthesia. Clin Obstet Gynecol 1974; 17:249–258. 329. Janeczko GF, El-Etr AA, Younes S. Low-dose ketamine anesthesia for obstetrical delivery. Anesth Analg 1974; 53:828–831. 330. Downing JW, Mahomedy MC, Jeal DE, Allen PJ. Anaesthesia for caesarean section with ketamine. Anaesthesia 1976; 31:883–892. 331. Ellingson A, Haram K, Sagen N. Ketamine and diazepam as anaesthesia for forceps delivery: a comparative study. Acta Anaesth Scand 1977; 21:37–40. 332. White PF, Way WL, Trevor AJ. Ketamine—its pharmacology and therapeutic uses. Anesthesiology 1982; 56:119–136. 333. Baraka A, Louis F, Dalleh R. Maternal awareness and neonatal outcome after ketamine induction of anaesthesia for caesarean section. Can J Anaesth 1990; 37:641–644. 334. Bovill JG, Coppel DL, Dundee JW, Moore J. Current status of ketamine anaesthesia. Lancet 1971; 1:1285–1288.

310

Diav-Citrin and Koren

335. Moore J, McNabb TG, Dundee JW. Preliminary report on ketamine in obstetrics. Br J Anaesth 1971; 43:779–782. 336. Michael CA, Potter JM. A comparison of labetolol with other antihypertensive drugs in the treatment of hypertensive disease of pregnancy. In: Riley A, Symonds EM, eds. The Investigation of Labetolol in the Management of Hypertension in Pregnancy. Amsterdam: Excerpta Medica, 1982, pp 111–122. 337. Davey DA, Dommisse J, Garden A. Intravenous labetolol and intravenous dihydralazine in severe hypertension in pregnancy. In: Riley A, Symonds EM, eds. The Investigation of Labetolol in the Management of Hypertension in Pregnancy. Amsterdam: Excerpta Medica, 1982, pp 52–61. 338. Michael CA. Use of labetolol in the treatment of severe hypertension during pregnancy. Br J Clin Pharmacol 1979; 8(suppl 2):211S–215S. 339. MacPherson M, Broughton Pipkin F, Rutter N. The effect of maternal labetolol on the newborn infant. Br J Obstet Gynaecol 1986; 93:539–542. 340. Riley AJ. Clinical pharmacology of labetolol in pregnancy. J Cardiovasc Pharmacol 1981; 3(suppl 1):S53–S59. 341. Sibai BM, Gonzalez AR, Mabie WC, Moretti M. A comparison of labetolol plus hospitalization versus hospitalization alone in the management of preeclampsia remote from term. Obstet Gynecol 1987; 70:323–327. 342. Liston WA, Adjepon-Yamoah KK, Scott DB. Foetal and maternal lignocaine levels after paracervical block. Br J Anaesth 1973; 45:750–754. 343. Scanlon JW, Brown WU Jr, Weiss JB, Alper MH. Neurobehavioral responses of newborn infants after maternal epidural anesthesia. Anesthesiology 1974; 40:121–128. 344. Shnider SM, Way EL. Plasma levels of lidocaine (Xylocaine) in mother and newborn following obstetrical conduction anesthesia: clinical applications. Anesthesiology 1968; 29:951– 958. 345. Rosa F, Bosco L. Infant renal failure with maternal ACE inhibition (abstr). Am J Obstet Gynecol 1991; 164:273. 346. Bhatt-Mehta V, Deluga KS. Chronic renal failure (CRF) in a neonate due to in-utero exposure to lisinopril. Presented at the 12th Annual Meeting of the American College of Clinical Pharmacy, Minneapolis, MN, August 20, 1991. Abstr, no. 43. 347. Pryde PG, Nugent CE, Sedman AB, Barr M Jr. ACE inhibitor fetopathy (abstr). Am J Obstet Gynecol 1992; 166:348. 348. Barr M Jr, Cohen MM Jr. ACE inhibitor fetopathy and hypocalvaria: the kidney–skull connection. Teratology 1991; 44:485–495. 349. Mizrahi EM, Hobbs JF, Goldsmith DI. Nephrogenic diabetes insipidus in transplacental lithium intoxication. J Pediatr 1979; 94:493–495. 350. Rane A, Tomson G, Bjarke B. Effects of maternal lithium therapy in a newborn infant. J Pediatr 1978; 93:296–297. 351. Woody JN, London WL, Wilbanks GD Jr. Lithium toxicity in a newborn. Pediatrics 1971; 47:94–96. 352. Tunnessen WW Jr, Hertz CG. Toxic effects of lithium in newborn infants: a commentary. J Pediatr 1972; 81:804–807. 353. Piton M, Barthe ML, Laloum D, et al. Acute lithium intoxication: report of two cases: mother and her newborn. Therapie 1973; 28:1123–1144. 354. Wilbanks GD, Bressler B, Peete CH Jr, et al. Toxic effects of lithium carbonate in a mother and newborn infant. JAMA 1970; 213:865–867. 355. Morrell P, Sutherland GR, Baumah PK, et al. Lithium toxicity in a neonate. Arch Dis Child 1982; 58:539–541. 356. Schou M, Goldfield MD, Weinstein MR, Villeneuve A. Lithium and pregnancy: I. Report from the register of lithium babies. BMJ 1973; 2:135–136. 357. Silverman JA, Winters RW, Strande C. Lithium carbonate therapy during pregnancy: apparent lack of effect upon the fetus. Am J Obstet Gynecol 1971; 109:934–936.

Direct Drug Toxicity to the Fetus

311

358. Strothers JK, Wilson DW, Royston N. Lithium toxicity in the newborn. BMJ 1973; 3:233– 234. 359. Karlsson K, Lindstedt G, Lundberg PA, Selstam U. Transplacental lithium poisoning: reversible inhibition of fetal thyroid. Lancet 1975; 1:1295. 360. Krause S, Ebbesen F, Lange AP. Polyhydramnios with maternal lithium treatment. Obstet Gynecol 1990; 75:504–506. 361. Stevens D, Burman D, Midwinter A. Transplacental lithium poisoning. Lancet 1974; 2:595. 362. Wilson N, Forfar JC, Goodman MJ. Atrial flutter in the newborn resulting from maternal lithium intoxication. Arch Dis Child 1983; 58:538–539. 363. Ang MS, Thorp JA, Parisi VM. Maternal lithium therapy and polyhydramnios. Obstet Gynecol 1990; 76:517–519. 364. Connoley G, Menahem S. A possible association between neonatal jaundice and long-term maternal lithium ingestion. Med J Aust 1990; 152:272–272. 365. Khandelwal SK, Sagar RS, Saxena S. Lithium in pregnancy and stillbirth: a case report. Br J Psychiatry 1989; 154:114–116. 366. Nishiwaki T, Tanaka K, Sekiya S. Acute lithium intoxication in pregnancy. Int J Gynecol Obstet 1996; 52:191–192. 367. McBride RJ, Dundee JW, Moore J, et al. A study of the plasma concentrations of lorazepam in mother and neonate. Br J Anaesth 1979; 51:971–978. 368. Di Michele V, Ramenghi LA, Sabatino G. Clozapine and lorazepam administration in pregnancy. Eur Psychiatry 1996; 11:214. 369. Dangman BC, Rosen TS. Magnesium levels in infants of mothers treated with MgSO 4 (abstr). Pediatr Res 1977; 11:415. 370. Lipsitz PJ, English IC. Hypermagnesemia in the newborn infant. Pediatrics 1967; 40:856– 862. 371. Lipsitz PJ. The clinical and biochemical effects of excess magnesium in the newborn Pediatrics 1971; 47:501–509. 372. Lamm CI, Norton KI, Murphy RJC, et al. Congenital rickets associated with magnesium sulfate infusion for tocolysis. J Pediatr 1988; 113:1078–1082. 373. Wilkins IA, Goldberg JD, Phillips RN, et al. Long-term use of magnesium sulfate as a tocolytic agent. Obstet Gynecol 1986; 67:38S–40S. 374. Dudley D, Gagnon D, Varner M. Long-term tocolysis with intravenous magnesium sulfate. Obstet Gynecol 1989; 73:373–378. 375. Pruett KM, Kirshon B, Cotton DB, et al. The effects of magnesium sulfate therapy on Apgar scores. Am J Obstet Gynecol 1988; 159:1047–1048. 376. Savory J, Monif GRG. Sreum calcium levels in cord sera of the progeny of mothers treated with magnesium for toxemia. Am J Obstet Gynecol 1971; 110:556–559. 377. Rasch DK, Huber PA, Richardson CJ, et al. Neurobehavioural effects of neonatal hypermagnesemia. J Pediatr 1982; 100:272–276. 378. Brady JP, Williams HC. Magnesium intoxication in a premature infant. Pediatrics 1967; 40: 100–103. 379. Brazy JE, Grimm JK, Little VA. Neonatal manifestations of severe maternal hypertension occurring before the thirty-sixth week of pregnancy. J Pediatr 1982; 100:265–271. 380. Holocomb WL Jr, Shackelford GD, Petrie RH. Prolonged magnesium therapy affects fetal bone (abstr). Am J Obstet Gynecol 1991; 164:386. 381. Smith LG Jr, Schanler RJ, Burns P, Moise KJ Jr. Effect of magnesium sulfate therapy (MgSO 4 ) on the bone mineral content of women and their newborns (abstr). Am J Obstet Gynecol 1991; 164:427. 382. Holocomb WL Jr, Shackelford GD, Petrie RH. Magnesium tocolysis and neonatal bone abnormalities: a controlled study. Obstet Gynecol 1991; 78:611–614. 383. Smith LG Jr, Burnes PA, Schanler RJ. Calcium homeostasis in pregnant women receiving long-term magnesium sulfate therapy for preterm labor. Am J Obstet Gynecol 1992; 167: 45–51.

312

Diav-Citrin and Koren

384. Cruishank DP, Chan GM, Doerrfeld D. Alterations in vitamin D and calcium metabolism with magnesium sulfate treatment of preeclampsia. Am J Obstet Gynecol 1993; 168:1170– 1177. 385. Rodis JF, Vinzileos AM, Campbell WA, et al. Maternal hypothermia: an unusual complication of magnesium sulfate therapy. Am J Obstet Gynecol 1987; 156:435–436. 386. Dayan E, Rosa FW. Fetal ambiguous genitalia associated with sex hormones use early in pregnancy. Food and Drug Administration, Division of Drug Experience. ADR Highlights 1981:1–14. 387. Lane PA, Hathaway WE. Vitamin K in infancy. J Pediatr 1985; 106:351–359. 388. Payne NR, Hasegawa DK. Vitamin K deficiency in newborns: a case report in α-1-antitrypsin deficiency and a review of factors predisposing to hemorrhage. Pediatrics 1984; 73:712– 716. 389. Wynn RM. The obstetric significance of factors affecting the metabolism of bilirubin, with particular reference to the role of vitamin K. Obstet Gynecol Surv 1963; 18:333–354. 390. Finkel MJ. Vitamin K 1 and vitamin K analogues. Clin Pharmacol Ther 1961; 2:795– 814. 391. Morrison JC, Wiser WL, Rosser SI, et al. Metabolites of meperidine related to fetal depression. Am J Obstet Gynecol 1973; 115:1132–1137. 392. Belfrage P, Boreus LO, Hartvig P, et al. Neonatal depression after obstetrical analgesia with pethidine: the role of the injection–delivery time interval and the plasma concentrations of pethidine and norpethidine. Acta Obstet Gynecol Scand 1981; 60:43–49. 393. Borgetedt AD, Rosen MG. Medication during labor correlated with behavior and EEG of the newborn. Am J Dis Child 1968; 115:21–24. 394. Hodgkinson R, Bhatt M, Wang CN. Double-blind comparison of the neurobehavior of neonates following the administration of different doses of meperidine to the mother. Can Anaesth Soc J 1978; 25:405–411. 395. Baxi LV, Petrie RH, James LS. Human fetal oxygenation (TcPO 2 ), heart rate and uterine activity following maternal administration of meperidine. J Perinat Med 1988; 16:23–30. 396. Gordon HR. Fetal bradycardia after paracervical block: correlation with fetal and maternal blood levels of local anesthetic (mepivacaine). N Engl J Med 1968; 279:910–914. 397. McConnell JF, Bhoola R. A neonatal complication of maternal leukemia treated with 6mercaptopurine. Postgrad Med 1973; 49:211–213. 398. Zelson C, Lee SJ, Casalino M. Neonatal narcotic addiction. N Engl J Med 1973; 289:1216– 1220. 399. Burstein Y, Giarrdina PJV, Rausen AR, et al. Thrombocytosis and increased circulating platelet aggregates in newborn infanst of polydrug users. J Pediatr 1979; 94:895–899. 400. Low L, Ratcliffe W, Alexander W. Intrauterine hypothyroidism due to antithyroid drug therapy for thyrotoxicosis during pregnancy. Lancet 1978; 2:370–371. 401. Hawe P, Francis HH. Pregnancy and thyrotoxicosis. BMJ 1962; 2:817–822. 402. Refetoff S, Ochi Y, Selenkow HA, Rosenfield RL. Neonatal hypothyroidism and goiter in one infant of each of two sets of twins due to maternal therapy with antithyroid drugs. J Pediatr 1974; 85:240–244. 403. Sugrue D, Drury MI. Hyperthyroidism complicating pregnancy: results of treatment by antithyroid drugs in 77 pregnancies. Br J Obstet Gynaecol 1980; 87:970–975. 404. Whitelaw A. Maternal methyldopa and neonatal blood pressure. BMJ 1981; 283:471. 405. Le Grass MD, Seifert B, Casiro O. Neonatal nasal obstruction associated with methyldopa treatment during pregnancy (letter). Am J Dis Child 1990; 144:143–144. 406. Plunkett GD. Neonatal complications. Obstet Gynecol 1973; 41:476–477. 407. Cowett RM, Hakanson DO, Kocon RW, Oh W. Untoward neonatal effect of intraamniotic administration of methylene blue. Obstet Gynecol 1976; 48:74S–75S. 408. Kirsch IR, Cohen HJ. Heintz body hemolytic anemia from the use of methylene blue in neonates. J Pediatr 1980; 96:276–278.

Direct Drug Toxicity to the Fetus

313

409. Crooks J. Haemolytic jaundice in a neonate after intra-amniotic injection of methylene blue. Arch Dis Child 1982; 57:872–873. 410. McEnerney JK, McEnerney LN. Unfavorable neonatal outcome after intraamniotic injection of methylene blue. Obstet Gynecol 1983; 61:35S–36S. 411. Serota FT, Bernbaum JC, Schwartz E. The methylene-blue baby. Lancet 1979; 2:1142–1143. 412. Vincer MJ, Allen AC, Evans JR, et al. Methylene-blue-induced hemolytic anemia in a neonate. Can Med Assoc J 1987; 136:503–504. 413. Spahr RC, Salbburey DJ, Krissberg A, Prin W. intraamniotic injection of methylene blue leading to methemoglobinemia in one of twins. Int J Gynaecol Obstet 1980; 17:477–478. 414. Poinsot J, Guillois B, Margis D, et al. Neonatal hemolytic anemia after intraamniotic injection of methylene blue. Arch Fr Pediatr 1988; 45:657–660. 415. Fish WH, Chazen EM. Toxic effects of methylene blue on the fetus. Am J Dis Child 1992; 146:1412–1413. 416. Troche BI. The methylene-blue baby. N Engl J Med 1989; 320:1756–1757. 417. Nicolini U, Monni G. Intestinal obstruction in babies exposed in utero to methylene blue. Lancet 1990; 336:1258–1259. 418. Van Der Pol JG, Wolf H, Boer K, et al. Jejunal atresia related to the use of methylene blue in genetic amniocentesis in twins. Br J Obstet Gynaecol 1992; 99:141–143. 419. Bland BAR, Lawes EG, Duncan PW, et al. Comparison of midazolam and thiopental for rapid sequence anesthetic induction for elective cesarean section. Anesth Analg 1987; 66: 1165–1168. 420. Ravlo O, Carl P, Crawford ME, et al. A randomized comparison between midazolam and thiopental for elective cesarean section anesthesia: II. Neonates. Anesth Analg 1989; 68: 234–237. 421. Kaler SG, Patrinos ME, Lambert GH, et al. Hypertrichosis and congenital anomalies associated with maternal use of minoxidil. Pedaitrics 1987; 79:434–436. 422. Rosa FW, Idanpaan-Heikkila J, Asanti R. Fetal minoxidil exposure. Pediatrics 1987; 80:120. 423. Bond GR, Zee AV. Ovedosage of misoprostol in pregnancy. Am J Obstet Gynecol 1994; 171:561–562. 424. Gilbert G, Dixon AB. Observations on Demerol as an obstetric analgesic. Am J Obstet Gynecol 1943; 45:320–326. 425. Way WL, Costley EC, Way EL. Respiratory sensitivity of the newborn infant to meperidine and morphine. Clin Pharmacol Ther 1965; 6:454–461. 426. Fox RE, Marx C, Stark AR. Neonatal effects of maternal nadolol therapy. Am J Obstet Gynecol 1985; 152:1045–1046. 427. Feinstein SJ, Lodeiro JG, Vintzileos AM, et al. Sinusoidal fetal heart rate pattern after administration of nalbuphine hydrochloride: a case report. Am J Obstet Gynecol 1986; 154:159– 160. 428. Miller RR. Evaluation of nalbuphine hydrochloride. Am J Hosp Pharm 1980; 37:942– 949. 429. Guillonneau M, Jacqz-Aigrain E, De Grepy A, Zeggout H. Perinatal adverse effects of nalbuphine given during parturition. Lancet 1990; 335:1588. 430. Sgro C, Escousse A, Tennenbaum D, Gouyon JB. Perinatal adverse effects of nalbuphine given during labour. Lancet 1990; 336:1070. 431. Wilson CM, McClean E, Moore J, Dundee JW. A double-blind comparison of intramuscular pethidine and nalbuphine in labour. Anaesthesia 1986; 41:1207–1213. 432. Frank M, McAteer EJ, Cattermole R, et al. Nalbuphine for obstetric analgesia. Anaesthesia 1987; 42:697–703. 433. Goodlin RC. Naloxone and its possible relationship to fetal endorphin levels and fetal distress. Am J Obstet Gynecol 1981; 139:16–19. 434. Wilkinson AR. Naproxen levels in preterm infants after maternal treatment. Lancet 1980; 2:591–592.

314

Diav-Citrin and Koren

435. Wilkinson AR, Aynsley-Green A, Mitchell MD. Persistent pulmonary hypertension and abnormal prostaglandin E levels in preterm infants after maternal treatment with naproxen. Arch Dis Child 1979; 54:942–945. 436. Cotton DB, Longmire S, Jones MM, et al. Cardiovascular alterations in severe pregnancyinduced hypertension: effects of intravenous nitroglycerin coupled with blood volume expansion. Am J Obstet Gynecol 1986; 154:1053–1059. 437. Longmire S, Leduc L, Jones MM, et al. The hemodynamic effects of intubation during nitroglycerin infusion in severe preeclampsia. Am J Obstet Gynecol 1991; 164:551–556. 438. Donchin Y, Amirav B, Sahar A, Yarkoni S. Sodium nitroprusside for aneurysm surgery in pregnancy. Br J Anaesth 1978; 50:849–851. 439. Hagler S, Schultz A, Hankin H, Kunstadter RN. Fetal effects of steroid therapy during pregnancy. Am J Dis Child 1963; 106:586–590. 440. Jacobson BD. Hazards of norethindrone therapy during pregnancy. Am J Obstet Gynecol 1962; 84:962–968. 441. Wilson JG, Brent RL. Are female sex hormones teratogenic? Am J Obstet Gynecol 1981; 141:567–580. 442. Bongiovanni AM, McFadden AJ. Steroids during pregnancy and possible fetal consequences. Fertil Steril 1960; 11:181–184. 443. Shearer WT, Schreiner RL, Marshall RE. Urinary retention in a neonate secondary to maternal ingestion of nortriptyline. J Pediatr 1972; 81:570–572. 444. McConell JB, Glasgow JF, McNair R. Effect on neonatal jaundice of oestrogens and progestogens taken before and after conception. BMJ 1973; 3:605–607. 445. Profuno R, Toce S, Kotagal S. Neonatal choreoathetosis following prenatal exposure to oral contraceptives (letter). Pediatrics 1990; 86:648–649. 446. Baxi LV, Gindoff PR, Pregenzer GJ, Parras MK. Fetal heart rate changes following maternal administration of a nasal decongestant. Am J Obstet Gynecol 1985; 153:799–800. 447. Freedman H, Tafeen CH, Harris H. Parenteral Win 20,228 as analgesic in labor. NY State J Med 1967; 2849–2851. 448. Refstad SO, Lindbaek E. Ventilatory depression of the newborn of women receiving pethidine or pentazocine. Br J Anaesth 1980; 52:165–270. 449. Sadove M, Balagot R, Branion J Jr, Kobak A. Report on the use of a new agent, phenazocine, in obstetric analgesia. Obstet Gynecol 1960; 16:448–453. 450. Corbit J, First S. Clinical comparison of phenazoxine and meperidine in obstetric analgesia. Obstet Gynecol 1961; 18:488–491. 451. Spiedel BD, Meadow SR. Maternal epilepsy and abnormalies of the fetus and the newborn. Lancet 1972; 2:839–843. 452. Bleyer WA, Skinner AL. Fatal neonatal hemorrhage after maternal anticonvulsant therapy. JAMA 1976; 235:826–827. 453. Lawrence A. Anti-epileptic drugs and the foetus. BMJ 1963; 2:1267–1273. 454. Kohler HG. Haemorrhage in the newborn of epileptic mothers. Lancet 1966; 1:267. 455. Mountain KR, Hirsh J, Gallus AS. Neonatal coagulation defect due to anticonvulsant drug treatment in pregnancy. Lancet 1970; 1:265–268. 456. Evans AR, Forrester RM, Discombe C. Neonatal haemorrhage during anticonvulsant therapy. Lancet 1970; 1:517–518. 457. Margolin FG, Kantor NM. Hemorrhagic disease of the newborn: an unusual case related to maternal ingestion of an anti-epileptic drug. Clin Pediatr (Phila) 1972; 11:59–60. 458. Srinivasan G, Seeler RA, Tiruvury A, Plides RS. Maternal anticonvulsant therapy and hemorrhagic disease of the newborn. Obstet Gynecol 1982; 59:250–252. 459. Payne NR, Hasegawa DK. Vitamin K deficiency in newborns: a case report in α-1-antitrypsin deficiency and a review of factors predisposing to hemorrhage. Pediatrics 1984; 73:712– 716. 460. Lane PA, Hathaway WE. Vitamin K in infancy. J Pediatr 1985; 106:351–359.

Direct Drug Toxicity to the Fetus

315

461. Allen RW Jr, Ogden B, Bentley FL, Jung AL. Fetal hydantoin syndrome, neuroblastoma, and hemorrhagic disease in a neonate. JAMA 1980; 244:1464–1465. 462. Douglas H. Haemorrhage in the newborn. Lancet 1966; 1:816–817. 463. Davis PP. Coagulation defect due to anticonvulsant drug treatment in pregnancy. Lancet 1970; 1:413. 464. Stevensom MM, Bilbert EF. Anticonvulsants and hemorrhagic diseases of the newborn infant. J Pediatr 1970; 77:516. 465. Truog WE, Feusner JH, Baker DL. Association of hemorrhagic disease and the syndrome of persistent fetal circulation with the fetal hydantoin syndrome. J Pediatr 1980; 96:112– 114. 466. Solomon GE, Hilgartner MW, Kutt H. Coagulation defects caused by diphenyhydantoin. Neurology 1972; 22:1165–1171. 467. Griffiths AD. Neonatal haemorrhage associated with maternal anticonvulsant therapy. Lancet 1981; 2:1296–1297. 468. Page TE, Hoyme HM, Markarian M, Jones KL. Neonatal hemorrhage secondary to thrombocytopenia: an occasional effect of prenatal hydantoin exposure. Birth Defects 1982; 18:47– 50. 469. Montan S, Ingemarsson I, Marsal K, Sjoberg N-O. Randomized controlled trial of atenolol and pindolol in human pregnancy: effects on fetal haemodynamics. BMJ 1992; 304:946– 949. 470. Chamberlain MJ, Reynolds AL, Yeoman WB. Toxic effects of podophyllum application in pregnancy. BMJ 1972; 3:391–392. 471. Warrell DW, Taylor R. Outcome for the foetus of mothers receiving prednisolone during pregnancy. Lancet 1968; 1:117–118. 472. Rudd NL, Freedom RM. A possible primidone embryopathy. J Pediatr 1979; 94:835–837. 473. Thomas P, Buchanan N. Teratogenic effect of anticonvulsants. J Pediatr 1981; 99:163. 474. John E. Promazine and neonatal hyperbilirubinemia. Med J Aust 1975; 2:342–344. 475. Crawford JS, as quoted by Moya F, Thorndike V. The effects of drugs used in labor on the fetus and newborn. Clin Pharmacol Ther 1963; 4:628–653. 476. Borgstedt AD, Rosen MG. Medication during labor correlated with behavior and EEG of the newborn. Am J Dis Child 1968; 115:21–24. 477. Corby DG, Shulman I. The effects of antenatal drug administration on aggregation of platelets of newborn infants. J Pediatr 1971; 79:307–313. 478. Whaun JM, Smith GR, Sochor VA. Effect of prenatal drug administration on maternal and neonatal platelet aggregation and PF4 release. Haemostasis 1980; 9:226–237. 479. Celleno D, Capogna G, Tomassetti M, et al. Neurobehavioural effects of propofol on the neonate following elective caesarean section. Br J Anaesth 1989; 62:649–654. 480. Jackson GL. Treatment of hyperthyroidism in pregnancy. Pa Med 1973; 76:56–57. 481. Langer A, Hung CT, McA’Nulty JA, et al. Adrenergic blockade: a new approach to hyperthyroidism during pregnancy. Obstet Gynecol 1974; 44:181–186. 482. Bullock JL, Harris RE, Young R. Treatment of thyrotoxicosis during pregnancy with propranolol. Am J Obstet Gynecol 1975; 121:242–245. 483. Lightner ES, Allen HD, Aoughlin G. Neonatal hyperthyroidism and heart failure: a different approach. Am J Dis Child 1977; 131:68–70. 484. Habib A, McCarthy JS. Effects on the neonate of propranolol administered during pregnancy. J Pediatr 1977; 91:808–811. 485. Pruyn SC, Phelan JP, Buchanan GC. Long-term propranolol therapy in pregnancy: maternal and fetal outcome. Am J Obstet Gynecol 1979; 135:485–489. 486. Turner GM, Aukley CM, Dixon HG. Management of pregnancy complicated by hypertrophic obstructive cardiomyopathy. BMJ 1968; 4:281–284. 487. Schroeder JS, Harrison DC. Repeated cardioversion during pregnancy. Am J Cardiol 1971; 27:445–446.

316

Diav-Citrin and Koren

488. Levitan AA, Manion JC. Propranolol therapy during pregnancy and lactation. Am J Cardiol 1973; 32:247. 489. Reed RL, Cheney CB, Fearon RE, et al. Propranolol therapy throughout pregnancy: a case report. Anesth Analg (Cleve) 1974; 53:214–218. 490. Fiddler GI. Propranolol in pregnancy. Lancet 1974; 2:722–723. 491. Bauer JH, Pape B, Zajicek J, Groshong T. Propranolol in human plasma and breast milk. Am J Cardiol 1979; 43:860–862. 492. Teuscher A, Boss E, Imhof P, et al. Effect of propranolol on fetal tachydardia in diabetic pregnancy. Am J Cardiol 1978; 42:304–307. 493. Gladstone GR, Hordof A, Gersony WM. Proranolol administration during pregnancy: effects on the fetus. J Pediatr 1975; 86:962–964. 494. Tcherdakoff PH, Colliard M, Berrard E, et al. Propranolol in hypertension during pregnancy. BMJ 1978; 2:670. 495. Eliahou HE, Silverberg DS, Reisin E, et al. Propranolol for the treatment of hypertension in pregnancy. Br J Obstet Gynaecol 1978; 85:431–436. 496. Bott-Kanner G, Schweitzer A, Schonfeld A, et al. Treatment with propranolol and hydralazine throughout pregnancy in a hypertensive patient: a case report. Isr J Med Sci 1978; 14: 466–468. 497. Bott-Kanner G, Reisner SH, Rosenfeld JB. Propranolol and hydralazine in the management of essential hypertension in pregnancy. Br J Obstet Gynaecol 1980; 87:110–114. 498. Taylor EA, Turner P. Anti-hypertensive therapy with propranolol during pregnancy and lactation. Postgrad Med 1981; 57:427–430. 499. O’Connor PC, Jick H, Hunter JR, et al. Propranolol and pregnancy outcome. Lancet 1981; 2:1168. 500. Caldroney RD. Beta-blockers in pregnancy. N Engl J Med 1982; 306:810. 501. Livingstone I, Craswell PW, Bevan EB, et al. Propranolol in pregnancy: three year prospective study. Clin Exp Hypertens (B) 1983; 2:341–350. 502. Tunstall ME. The effect of propranolol on the onset of breathing at birth. Br J Anaesth 1969; 41:792–796. 503. Goodlin RC. Beta blocker in pregnancy-induced hypertension. Am J Obstet Gynecol 1982; 143:237–241. 504. Mitrani A, Oettinger M, Abinader EG, et al. Use of propranolol in dysfunctional labour. Br J Obstet Gynaecol 1975; 82:651–655. 505. Cheron RG, Kaplan MM, Larsen PR, et al. Neonatal thyroid function after propylthiouracil therapy for maternal Graves’ disease. N Engl J Med 1981; 304:525–528. 506. Aaron HH, Schneierson SJ, Siegel E. Goiter in newborn infant due to mother’s ingestion of propylthiouracil. JAMA 1955; 159:848–850. 507. Krementz ET, Hooper RG, Kempson RL. The effects on the rabbit fetus of the maternal administration of propylthiouracil. Surgery 1957; 41:619–631. 508. Branch LK, Tuthill SW. Goiters in twins resulting from propylthiouracil given during pregnancy. Ann Intern Med 1957; 46:145–148. 509. Man EB, Shaver BA Jr, Cooke RE. Studies of children born to women with thyroid disease. Am J Obstet Gynecol 1958; 75:728–741. 510. Hayashida CY, Duarte AJS, Sato AE, Yamashiro-Kanashiro EH. Neonatal hepatitis and lymphocyte sensitization by placental transfer of propythiouracil. J Endocrinol Invest 1990; 13: 937–941. 511. Scriver CR. Vitamin B6 deficiency and dependency in man. Am J Dis Child 1967; 113:109– 114. 512. Hunt AD Jr, Stokes J Jr, McCrory WW, Stroud HH. Pyridoxine dependency: report of a case of intractable convulsions in an infant controlled by pyridoxine. Pediatrics 1954; 13: 140–145. 513. Bankier A, Turner M, Hopkins IJ. Pyridoxine dependent seizures—a wider clinical spectrum. Arch Dis Child 1983; 58:415–418.

Direct Drug Toxicity to the Fetus

317

514. Bejsovec MIR, Kulenda Z, Ponca E. Familial intrauterine convulsions in pyridoxine dependency. Arch Dis Child 1967; 42:201–207. 515. Domula VM, Weissach G, Lenk H. Uber die Auswirkung medicamentoser Behandlung in der Schwangerschaft auf das Gerennungspotential des Neugeborenen. Zentralbl Gynaekol 1977; 99:473–479. 516. Nishimura H, Tanimura T. Clinical Aspects of the Teratogenicity of Drugs. New York: American Elsevier, 1976, pp 140–143. 517. Robinson GC, Brummitt JR, Miller JR. Hearing loss in infants and preschool children: II. Etiological Considerations. Pediatrics 1963; 32:115–124. 518. West RA. Effect of quinine upon auditory nerve. Am J Obstet Gynecol 1938; 36:241– 248. 519. McKinna AJ. Quinine induced hupoplasia of the optic nerve. Can J Ophthalmol 1966; 1: 261–264. 520. Morgon A, Charachon D, Brinquier N. Disorders of the auditory apparatus caused by embryopathy or fetopathy: prophylaxis and treatment. Acta Otolaryngol (Stockh) 1971; 291(suppl):5. 521. Mauer MA, DeVaux W, Lahey ME. Neonatal and maternal thrombocytopenic purpura due to quinine. Pediatrics 1957; 19:84–87. 522. Glass L, Rajegowda BK, Bowne E, Evans HE. Exposure to quinine and jaundice in a glucose6-phosphate dehydrogenase-deficient newborn infant. Pediatrics 1973; 82:734–735. 523. Budnick IS, Leikin S, Hoeck LE. Effect in the newborn infant to reserpine administration ante partum. Am J Dis Child 1955; 90:286–289. 524. Barden TP, Peter JB, Merkatz IR. Ritodrine hydrochloride: a betamimetic agent for use in preterm labor: I. Pharmacology, clinical history, administration, side effects, and safety. Obstet Gynecol 1980; 56:1–6. 525. Ritodrine for inhibition of preterm labor. Med Lett Drugs Ther 1980; 22:89–90. 526. Finkelstein BW. Ritodrine (Yutopar, Merrell Dow Pharmaceuticals Inc.). Drug Intell Clin Pharm 1981; 15:425–433. 527. Brosset P, Ronayette D, Pierre MC, et al. Cardiac complications of ritodrine in mother and baby. Lancet 1982; 1:1468. 528. Hermansen MC, Johnson GL. Neonatal supraventricular tachycardia following prolonged maternal ritodrine administration. Am J Obstet Gynecol 1984; 149:798–799. 529. Beitzke A, Winter R, Zach M, Grubbauer HM. Kongenitales Vorhofflattern mit Hydrops fetalis durch mutterliche Tokolytikamedikation. Klin Paediatr 1979; 191:410–417. 530. Nuchpuckdee P, Brodsky N, Porat R, Hurt H. Ventricular septal thickness and cardiac function in neonates after in utero ritodrine exposure. J Pediatr 1986; 109:687–691. 531. Leake RD, Hobel CJ, Oh W, et al. A controlled, prospective study of the effects of ritodrine hydrochloride for premature labor (abstr). Clin Res 1980; 28:90A. 532. Kazzi NJ, Gross TL, Kazzi GM, Williams TG. Neonatal complications following in utero exposure to intravenous ritodrine. Acta Obstet Gynecol Scand 1987; 66:65–69. 533. Schilthuis MS, Aarnoudse JG. Fetal death associated with severe ritodrine induced ketoacidosis. Lancet 1980; 1:1145. 534. Huisjes HJ, Touwen BCL. Neonatal outcome after treatment with ritodrine: a controlled study. Am J Obstet Gynecol 1983; 147:250–253. 535. Hansen NB, Oh W, LaRochelle F, Stonestreet BS. Effects of maternal ritodrine administration on neonatal renal function. J Pediatr 1983; 103:774–780. 536. Shenker L. Clinical experience with fetal heart rate monitoring of one thousand patients in labor. Am J Obstet Gynecol 1973; 115:1111–1116. 537. Boehm FH, Growdon JH Jr. The effects of scopolamine on fetal heart rate baseline variability. Am J Obstet Gynecol 1974; 120:1099–1104. 538. Ayromlooi J, Tobias M, Berg P. The effects of scopolamine and ancillary analgesics upon the fetal heart rate recording. J Reprod Med 1980; 25:323–326. 539. Evens RP, Leopold JC. Scopolamine toxicity in a newborn. Pediatrics 1980; 56:245–248.

318

Diav-Citrin and Koren

540. O’Hare MF, Murnaghan GA, Russell CJ, et al. Sotalol as a hypotensive agent in pregnancy. Br J Obstet Gynaecol 1980; 87:814–820. 541. Leroux M. Existe-t-il une surdite congenitale acquise due a la streptomycine? Ann Otolaryngol 1950; 67:194–196. 542. Nishimura H, Tanimura T. Clinical Aspects of the Teratogenicity of Drugs. New York: Excerpta Medica, 1976, p 130. 543. Donald PR, Sellars SL. Streptomycin ototoxicity in the unborn child. S Afr Med J 1981; 60: 316–318. 544. Heckel GP. Chemotherapy during pregnancy: danger of fetal injury from sulfanilamide and its derivatives. JAMA 1941; 117:1314–1316. 545. Ginzler AM, Cherner C. Toxic manifestations in the newborn infant following placental transmission of sulfanilamide: with a report of 2 cases simulating erythroblastosis fetalis. Am J Obstet Gynecol 1942; 44:46–55. 546. Lucey JF, Driscoll TJ Jr. Hazard to newborn infants of administration of long-acting sulfonamides to pregnant women. Pediatrics 1959; 24:498–499. 547. Kantor HI, Sutherland DA, Leonard JT, et al. Effect on bilirubin metabolism in the newborn of sulfisoxazole administration to the mother. Obstet Gynecol 1961; 17:494–500. 548. Dunn PM. The possible relationship between the maternal administration of sulphamethoxypyridazine and hyperbilirubinaemia in the newborn. J Obstet Gynaecol Br Commonw 1964; 71:128–131. 549. Perkins RP. Hydrops fetalis and stillbirth in a male glucose-6-phosphate dehydrogenasedeficient fetus possibly due to maternal ingestion of sulfisoxazole. Am J Obstet Gynecol 1971; 111:379–381. 550. Kramer W, Saade G, Belfort M, et al. Randomized double-blind study comparing sulindac to terbutaline: fetal cardiovascular effects (abstr). Am J Obstet Gynecol 1996; 174:326. 551. Rasanen J, Jouppila P. Fetal cardiac function and ductus arteriosus during indomethacin and sulindac therapy for threatened preterm labor: a randomized study. Am J Obstet Gynecol 1995; 173:20–25. 552. Tewari K, Bonebrake RG, Asrat T, Shanberg AM. Ambiguous genitalia in infant exposed to tamoxifen in utero. Lancet 1997; 350:183. 553. Haller DL. The use of terbutaline for premature labor. Drug Intell Clin Pharm 1980; 14: 757–764. 554. Andersson KE, Bengtsson LP, Gustafson I, Ingermarsson I. The relaxing effect of terbutaline on the human uterus during term labor. Am J Obstet Gynecol 1975; 121:602–609. 555. Ingermarsson I. Effect of terbutaline on premature labor: a double-blind placebo-controlled study. Am J Obstet Gynecol 1976; 125:520–524. 556. Ravindran R, Viegas OJ, Padilla LM, LaBlonde P. Anesthetic considerations in pregnant patients receiving terbutaline therapy. Anesth Analg (Cleve) 1980; 59:391–392. 557. Epstein MF, Nicholls RN, Stubblefield PG. Neonatal hypoglycemia after beta-sympathomimetic tocolytic therapy. J Pediatr 1979; 94:449–453. 558. Westgren M, Carlsson C, Lindholm T, et al. Continuous maternal glucose measurements and fetal glucose and insulin levels after administration of terbutaline in term labor. Acta Obstet Gynecol Scand 1982; 108(suppl):63–65. 559. Fletcher SE, Fyfe DA, Case CL, et al. Myocardial necrosis in a newborn after long-term maternal subcutaneous terbutaline infusion for suppression of preterm labor. Am J Obstet Gynecol 1991; 165:1401–1404. 560. Thorkelsson T, Loughead JL. Long-term subcutaneous terbutaline tocolysis: report of possible neonatal toxicity. J Perinatol 1991; 11:235–238. 561. Cohlan SQ, Bevelander G, Bross S. Effect of tetracycline on bone growth in the ptremature infant. Antimicrob Agents Chemother 1961; 340–347. 562. Harcourt JK, Johnson NW, Storey E. In vivo incorporation of tetracycline in the teeth of man. Arch Oral Biol 1962; 7:431–437.

Direct Drug Toxicity to the Fetus

319

563. Rendle-Short TJ. Tetracycline in teeth and bone. Lancet 1962; 1:1188. 564. Douglas AC. The deposition of tetracycline in human nails and teeth: a complication of long term treatment. Br J Dis Chest 1963; 57:44–47. 565. Kutscher AH, Zegarelli EV, Tovell HM, Hochberg B. Discoloration of teeth induced by tetracycline. JAMA 1963; 184:586–587. 566. Kline AH, Blattner RJ, Lunin M. Transplacental effect of tetracyclines on teeth. JAMA 1964; 188:178–180. 567. Macaulay JC, Lestyna JA. Preliminary observations on the prenatal administration of demethylchlortetracycline HCl. Pediatrics 1964; 34:423–424. 568. Stewart DJ. The effects of tetracyclines upon the dentition. Br J Dermatol 1964; 76:374– 378. 569. Swallow JN. Discoloration of primary dentition after maternal tetracycline ingestion in pregnancy. Lancet 1964; 2:611–612. 570. Porter PJ, Sweeney EA, Golan H, Kass EH. Controlled study of the effect of prenatal tetracycline on primary dentition. Antimicrob Agents Chemother 1965; 5:668–671. 571. Toaff R, Ravid R. Tetracyclines and the teeth. Lancet 1966; 2:281–282. 572. Kutscher AH, Zegarelli EV, Tovell HM, et al. Discoloration of deciduous teeth induced by administrations of tetracycline antepartum. Am J Obstet Gynecol 1966; 96:291–292. 573. Brearley LJ, Stragis AA, Storey E. Tetracycline-induced tooth changes: Part 1. Prevalence in preschool children. Med J Aust 1968; 2:653–658. 574. Brearley LJ, Storey E. Tetracycline-induced tooth changes: Part 2. Prevalence, localization and nature of staining in extracted deciduous teeth. Med J Aust 1968; 2:714–719. 575. Baker KL, Storey E. Tetracycline-induced tooth changes: Part 3. Incidence in extracted first permanent molar teeth. Med J Aust 1970; 1:109–113. 576. Anthony JR. Effect on deciduous and permanent teeth of tetracycline deposition in utero. Postgrad Med 1970; 48:165–168. 577. Genot MT, Golan HP, Porter PJ, Kass EH. Effect of administration of tetracycline in pregnancy on the primary dentition of the offspring. J Oral Med 1970; 25:75–79. 578. Arwood LL, Dasta JF, Friedman C. Placental transfer of theophylline: two case reports. Pediatrics 1979; 63:844–846. 579. Yeh TF, Pildes RS. Transplacental aminophylline toxicity in a neonate. Lancet 1977; 1:910. 580. Labovitz E, Spector S. Placental theophylline transfer in pregnant asthmatics. JAMA 1982; 247:786–788. 581. O’Donnell R, Costigan C, O’Connell LG. Two cases of acute leukaemia in pregnancy. Acta Haematol 1979; 61:298–300. 582. Volkenandt M, Buchner T, Hiddemann W, Van De Loo J. Acute leukaemia during pregnancy. Lancet 1987; 2:1521–1522. 583. Schiff D, Aranda J, Stern L. Neonatal thrombocytopenia and congenital malformation associated with administration of tolbutamide to the mother. J Pediatr 1970; 77:457–458. 584. Nau H, Rating D, Koch S, et al. Valproic acid and its metabolites: placental transfer, neonatal pharmacokinetics, transfer via mother’s milk and clinical status in neonates of epileptic mothers. J Pharmacol Exp Ther 1981; 219:768–777. 585. Bantz EW. Valproic acid and congenital malformations: a case report. Clin Pediar 1984; 23: 353–354. 586. Legius E, Jaeken J, Eggermont E. Sodium valproate, pregnancy, and infantile fatal liver failure. Lancet 1987; 2:1518–1519. 587. Felding I, Rane A. Congenital liver damage after treatment of mother with valproic acid and phenytoin? Acta Paediatr Scand 1984; 73:656–658. 588. Majer RV, Green PJ. Neonatal afibrinogenaemia due to sodium valproate. Lancet 1987; 2: 740–741. 589. Simila S, von Wendt L, Hartikainen-Sorri A-L, et al. Sodium valproate, pregnancy, and neonatal hyperglycinaemia. Arch Dis Child 1979; 54:985–986.

320

Diav-Citrin and Koren

590. Jager-Roman E, Deichi A, Jakob S, et al. Fetal growth, major malformations, and minor anomalies in infants born to women receiving valproic acid. J Pediatr 1986; 108:997–1004. 591. Hill LM. Fetal distress secondary to vancomycin-induced maternal hypotension. Am J Obstet Gynecol 1985; 153:74–75. 592. Sperling RS, Stratton P, O’Sullivan MJ, et al. A survey of zidovudine use in pregnant women with human immunodeficiency virus infection. N Engl J Med 1992; 326:857–861. 593. Connor EM, Sperling RS, Gelber R, et al. for the Pediatric AIDS Clinical Trials Group Protocol 076 Study Group. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. N Engl J Med 1994; 331:1173–1180. 594. Pryde PG, Sedman AB, Nugent CE, Barr M. Angiotensin converting enzyme inhibitor fetopathy. J Am Soc Nephrol 1993; 3:1575–1582. 595. Barr M. Teratogen update: angiotensin converting enzyme inhibitors. Teratology 1994; 50: 399–409. 596. Mastrobattista JM. Angiotensin converting inhibitors in pregnancy. Semin Perinatol 1997; 21:124–134.

17 The Approach to the Mother on Nonmedicinal and Chemical Use in Pregnancy Joyce F. Schneiderman The Addiction Research Foundation and The University of Toronto, Toronto, Ontario, Canada

Clinical Case One of your pregnant patients admits to having used narcotic opioids for several years. She wishes very much to discontinue this addiction pattern but does not know how. How should she be treated?

INTRODUCTION Nonmedical drug use can be defined as the taking of any psychoactive drug in the absence of a clearly defined medical indication (1). People ingest many chemicals to modify mood: as part of a social or recreational activity, as part of a lifestyle, or to self-medicate. Drugs are widely available in modern society. Alcohol, caffeine, and nicotine are so commonly used that they are not always considered to be drugs. Prescription drugs, like narcotics and benzodiazepines, may be obtained through a physician or other sources. Illicit drugs— whose purity, dose, and sometimes even the substance itself are unknown—include narcotics, stimulants, cannabinoids, and hallucinogens. This chapter focuses on the management of women who use psychoactive drugs during pregnancy; it covers advice appropriate to the woman who plans to discontinue her drug use and discusses the care of the drug abuser who will require intensive treatment and support in a specialized setting. Neonatal withdrawal and drug use during breastfeeding are discussed in Chapters 8 and 13. Detailed discussion of drug abuse treatment and of the teratogenicity of psychoactive drugs is beyond the scope of this chapter. Women of childbearing age have been increasing their nonmedical drug use over the last two decades (2–5). This increase and the thalidomide tragedy have fueled a concern for the effects of drugs on pregnancy. Recreational drug users and drug-dependent women may appear unconcerned about health risks to themselves, but they, too, usually wish to spare their unborn children from risks associated with drug use.

321

322

Schneiderman

Our experience at the Motherisk Clinic in Toronto suggests that many women and their primary care physicians overestimate the risk of drugs to the fetus (6). Some women will consider therapeutic abortion, often on the recommendation of their doctor, fearing a major teratogenic effect. It is therefore extremely important to provide accurate information on the known risks of maternal drug taking. There is no evidence that the father’s drug and alcohol consumption directly influences the outcome of pregnancy. His use may nonetheless be a marker for other factors contributing to increased risk. Concerns that previous drug use in either parent, particularly of hallucinogens and cannabinoids, may lead to an adverse outcome based on permanent chromosomal damage are unwarranted. In the 1990s, it is rare to find a woman drug abuser who uses only one drug or even one drug class. The synergistic effects of drug interactions on the fetus remain largely unknown. An adverse outcome may be related to drug exposure at a critical period in gestation, to maximal drug level obtained, to cumulative exposure, or to a combination of these and other factors. The outcome of pregnancy in the alcohol- or drug-dependent woman depends on more than just her drug-taking behavior. The pharmacological effects are often compounded by lack of prenatal care, higher risk for other complications of pregnancy, cigarette smoking, and poor diet. The outcome in women of the inner-city drug subculture cannot be extrapolated to the woman who has used a substance one or several times socially or to the one who has received a short course of parenteral narcotics in the treatment of an acute medical or surgical illness prior to knowledge of the pregnancy. Neurobehavioral abnormalities have been reported in neonates of drug-using mothers (7); alcohol (8), marijuana (9), methadone (10), and cocaine (11) have been among the substances implicated. However, it is not possible to separate the contribution of prenatal drug exposure from the effects of persistent drug levels at birth, withdrawal symptoms, difficulties with mother-infant bonding, and continued drug exposure if the mother is breast-feeding. Data on long-term sequelae of prenatal drug use are limited, and results must be interpreted in light of the frequently suboptimal postnatal environment. General Guidelines for Clinical Management Because pregnancy is usually diagnosed at 6–8 weeks after the last menstrual period, it is important to consider the woman’s usual drug-taking behavior. Confirmation of pregnancy may be further delayed in the drug abuser because the menstrual cycle is often irregular, and prolonged amenorrhea is common in this population. If early symptoms of pregnancy—such as nausea, cramps, and fatigue—are attributed to drug withdrawal, they may prompt an increase in drug use. For the woman presenting early in pregnancy and motivated to discontinue her drug use, a single counseling session addressing the adverse effects on pregnancy outcome if drug use is not curtailed may be sufficient. Even the problem user with some social stability can be managed by the primary care physician as an outpatient with a tailored detoxification regimen (see below) and a short course of weekly or biweekly counseling. If the patient is unsuccessful in following through on this plan after a 2-week trial or if she is severely drug-dependent and in crisis, referral for specialized alcohol and drug treatment services is indicated. The initial approach to the woman with a significant alcohol or drug problem should

Nonmedicinal and Chemical Use in Pregnancy

323

include a complete medical and obstetrical assessment. The drug history is part of data collection and should be taken in a nonjudgmental manner. Use of alcohol, tobacco, caffeine, over-the-counter drugs, prescription medications, and illicit substances should be asked about specifically in each case, including drug names, dose, route, timing in relation to conception, adverse reactions, and symptoms of overdose and withdrawal. Familiarity with street names and local availability of illicit drugs is helpful. A detailed psychosocial history provides the information on the patient’s attitude about the pregnancy, the involvement of the baby’s father, the condition of other children, and the current housing, financial, employment, and legal situations necessary to plan treatment. The physical examination should screen for associated medical problems, such as trauma, hepatitis, abscesses, and sexually transmitted diseases. Urine toxicology for alcohol, benzodiazepines, barbiturates, cannabinoids, narcotics, and stimulants should be done at the first visit and repeated regularly throughout the pregnancy. An ultrasound examination is useful to establish gestational age and to assess fetal morphology. For women who are ambivalent about discontinuing drug use, this procedure may have therapeutic benefit (12). By making the pregnancy more ‘‘real,’’ it may be possible to increase the commitment of such a woman to ensuring the best possible outcome for her child. Women with a long history of alcohol- and drug-related problems will usually require inpatient detoxification or methadone maintenance substitution. Prenatal care should be integrated with drug treatment. A multidisciplinary staff team comprising a physician, nurse, social worker, addiction counselor, and dietitian is useful for case management. These patients can be difficult and time-consuming to treat. A consistent approach by all team members is crucial so that the patient receives a clear message. The individual patient’s situation will dictate the need for a residential treatment program or outpatient individual or group counseling. If a spouse or partner who also has a drug problem is involved, every attempt should be made to facilitate his treatment. Liaison with Alcoholics Anonymous, Narcotics Anonymous, and other voluntary self-help groups is valuable. The ideal situation assumes a motivated patient presenting early in pregnancy and the availability of resources to provide intensive treatment if necessary. In many cases, prenatal care is sporadic at best, and crisis intervention (medical or psychosocial) may be the only treatment accepted. Each contact with the patient should be taken as an opportunity to engage her in ongoing care. At present, there is little to offer the woman who refuses treatment, but with sensitive handling she may ultimately comply voluntarily with a treatment plan. In Ontario, the Children’s Aid Society will not intervene until after the birth of the child. This is also the case for child protection agencies in most other jurisdictions. If the woman presents before 12 weeks of gestation and wishes to consider a therapeutic abortion, this should be discussed if applicable laws permit, but this option should not be recommended as a standard practice. The clinician’s personal views on the use of illicit substances and alcohol should not be allowed to interfere with forming a therapeutic alliance with the patient. The pregnant substance-abusing patient is generally ambivalent about seeking treatment and responds best to a nonthreatening approach. She should be provided with specific objective information on the known effects of her drug use on the pregnancy, and she should be encouraged to stop all use in the context of ensuring the best possible fetal outcome. After the birth, mother and child should be followed closely. Postnatal care should include supportive counseling, contraception counseling, visits from a public health nurse, and training in parenting skills.

324

Schneiderman

ILLICIT SUBSTANCES AND INTRAVENOUS DRUG ABUSE The lifestyle of women using intravenous drugs plays a very important part in the formulation of the treatment plan. If they and their children remain in the drug-using community, they are not likely to abstain. Furthermore, pregnancy does not guarantee protection from the high incidence of physical violence these women experience (13). Exchange of sexual favors for drugs, in addition to prostitution, puts these women at high risk for sexually transmitted diseases. Sharing unsterilized needles puts them at risk both for hepatitis B and HIV infections. With prenatal screening and immunization as required, perinatal transmission of hepatitis B can now largely be prevented. Women account for 7% of AIDS cases in the United States (14–16). More than half these women are intravenous drug users (17), and over 80% are of childbearing age. The incidence of perinatal transmission of AIDS is still uncertain. Friedland and Klein estimate a rate of roughly 40–50% based on information available in 1987 (16). Infants retain maternal antibody to a median age of 10 months (18) and must be monitored for the development of symptoms of AIDS. Prevention must be based on education and counseling of all intravenous drug users and their sexual partners. Routine screening for HIV in this population is controversial. Instead, serological testing should be offered to any high-risk woman contemplating a pregnancy. Contraceptive counseling is mandatory in any woman known to be HIV-positive.

Individual Drugs Alcohol Most women are aware that heavy drinking in pregnancy can harm the fetus. The fetal alcohol syndrome (FAS), first described by Smith and Jones in 1973, is now generally accepted to occur in the offspring of alcoholic women—that is, women who drink at least six standard drinks per day throughout the first trimester [one standard drink ⫽ 12 oz beer ⫽ 5 oz wine ⫽ 1.5 oz liquor, or approximately 15 g (0.5 oz) absolute alcohol]. The signs of FAS are prenatal and postnatal growth retardation, central nervous system dysfunction (often including mental retardation), facial dysmorphology, and many other congenital abnormalities (19–24). A proposed minimum criterion for the diagnosis of FAS recommends that at least two of the following be present: microcephaly (head circumference less than the third percentile), microphthalmia and/or short palpebral fissures, poorly developed philtrum, thin upper lip, and flattening of the maxillary area in addition to growth and neurological abnormalities (25). The fetal alcohol syndrome has been estimated to occur in between one and two live births per 1000 in the general population and is a major preventable cause of mental retardation. An incidence of greater than 40% has been reported in alcoholic women. However, prospectively collected data suggest a true incidence closer to 2.5% (26). On long-term follow-up, FAS children display lack of catch-up growth, attentional deficits, mental retardation, and dysmorphic features. With time, some improvement in all these areas occurs owing to biological maturation, but the children manifest poor school performance, which is independent of environment (27). Other adverse fetal outcomes in heavy drinkers include increased spontaneous abortions (26), premature placental separation, stillbirth, low birth weight (28), and congenital malformations (29). Drinking by the father in the month prior to conception has been

Nonmedicinal and Chemical Use in Pregnancy

325

associated with decreased birth weight (30), but—as indicated above—this finding does not necessarily represent causality. The data on women drinking less than six drinks per day are more difficult to interpret. Studies may rely on retrospective self-reports, use averaged amounts throughout the pregnancy, and employ varying definitions for moderate drinking. At drinking levels greater than two drinks per day, a partial FAS not meeting the minimal criteria, termed fetal alcohol effects, has been described. No such effects have been reported when consumption is less than two drinks per day (31). A prospective study examining firsttrimester alcohol use evaluated congenital malformations by chart review, including 32,409 patients drinking two or fewer drinks per day. There was no significant increase in malformations between the group drinking up to two drinks and abstainers (32). Another study of more than 12,000 pregnancies showed an increase in abruptio placentae but no other adverse outcome associated with an intake of less than two drinks per day (33). There is no known safe lower limit for alcohol consumption in pregnancy, and many authorities recommend complete abstinence on this basis. Clearly, this is the most conservative approach. If all women of childbearing age were to abstain from alcohol, presumably FAS would disappear. The absence of risk is, of course, very difficult to demonstrate scientifically. Pregnancy should not be a time for undue restriction of lifestyle, and most women do drink. Even women who plan to abstain during pregnancy usually continue their customary drinking pattern until they confirm they are pregnant. The mother of an infant with an unrelated congenital abnormality should not have to bear the guilt of thinking that her minimal alcohol intake was responsible. Rosett and Weiner conclude that there is no measurable risk from consuming less than 1 oz absolute alcohol (two standard drinks) per day (34). A Canadian committee has recommended abstinence ‘‘or at least to limit consumption to less than 4 drinks per week’’ (35). Setting reasonable goals for moderation may enhance clinical credibility and compliance while fairly emphasizing the hazards of heavy drinking. Alcohol consumption in the month preceding recognition of pregnancy most accurately reflects early postconception drinking. Even among the heaviest drinkers, more than 50% of women spontaneously decrease their intake at the diagnosis of pregnancy (36). The social drinker also markedly decreases her drinking, often citing distaste, lack of appeal, or adverse physiological effects (37). In the heavy social drinker or physically dependent woman, detoxification should occur as early as possible, and complete abstinence should be the only treatment goal considered. For a woman who is drinking more than 80 g per day, a gradual reduction over 3–4 days is preferable (34). Intensive supportive counseling should be available during detoxification and throughout the pregnancy. If the woman uses other drugs, has a medical illness, or seems unlikely to be successful as an outpatient because of prior failure or current circumstances, inpatient admission should be arranged. We have successfully used a diazepam-loading regimen in women requiring hospitalization and pharmacological treatment of alcohol withdrawal (38). The long half-life of diazepam offers a ‘‘pharmacological taper,’’ so that the patient usually does not require further drug after the first day of admission. This regimen is well tolerated by the pregnant woman, allowing early initiation of a global treatment plan instead of focusing on withdrawal symptomatology. Disulfiram interferes with the intermediary metabolism of ethanol. The alcohol-disulfiram reaction is believed to be due to accumulation of acetaldehyde, and this pharmacological effect has been used as an adjunct in the treatment of alcohol dependence. Since

326

Schneiderman

the efficacy of disulfiram is unproven and the potential for a severe reaction and teratogenic effect must be considered, this drug is contraindicated in the woman who is pregnant or planning a pregnancy (39). The woman taking disulfiram who has an unplanned pregnancy should be counseled with respect to the reported associations of teratogenesis. One series of five cases reported on spontaneous abortion and two cases of clubfoot (40). Another group described two case reports of infants with limb reduction (41). Isolated case reports do not necessarily mean teratogenesis; therefore therapeutic abortion should not routinely be recommended. A detailed ultrasound examination may be helpful. Narcotics Heroin use is associated with decreased birth weight and a higher incidence of medical— primarily infectious—and obstetrical complications. Narcotics do not appear to be teratogenic in humans (42). Infants of narcotic-dependent women are at risk for neonatal withdrawal (43) and sudden infant death syndrome (44). Although total abstinence seems preferable, this is often not realistically feasible. Pregnant women recently detoxified from narcotics have the same high rate of recidivism as nonpregnant addict (45,46). Rapid detoxification has been associated with fetal distress and death (47,48). Methadone maintenance has been the treatment of choice for the heroin addict (49). Pharmacologically, methadone suppresses withdrawal symptoms; in high enough doses, it will block opioid effects. It provides a constant drug level rather than the wide fluctuations that occur with illicit intravenous use, thus avoiding fetal risks from overdose or withdrawal. In addition to decreasing illicit use, methadone maintenance provides a framework for ongoing, often daily contact with a treatment facility and institution of counseling and prenatal care. With methadone maintenance, fetal outcome improves and birth weights increase, albeit less than those in matched non-drug-using controls. Kandall et al. (50) report mean birth weights of 2490 g in heroin-abusing mothers, 2961 g in methadone-maintained mothers, and 3176 g in controls. Birth weight correlates to the methadone dose (50); however, it is likely that other factors are responsible. For example, there may be less illicit use in women receiving high doses. A good outcome also correlates with the number of prenatal visits. A short initial hospitalization may provide the opportunity for prenatal assessment, treatment of concurrent medical problems, crisis intervention, introduction of the members of the multidisciplinary team, and stabilization on methadone (46). Ideally, the methadone dose should be carefully titrated to withdrawal symptoms, with special attention to lower abdominal cramping (which may reflect uterine irritability), so that the woman is on the lowest possible dose. With a starting dose of 10–20 mg and additional doses every 6–8 hours, the maintenance dose can be arrived at within 24–48 hours. In some programs, this is not feasible, and higher doses of 50–80 mg daily are chosen empirically. If a woman can be maintained on 20–25 mg daily, significant neonatal withdrawal usually does not occur (51,52). Areas outside major urban centers often lack methadone maintenance clinics. For the prescription abuser, or other highly motivated women with good community supports in such areas, a gradual outpatient detoxification with methadone or another narcotic can be undertaken over 1 to several weeks. This is most safety accomplished in the second trimester, when the risk of spontaneous abortion or precipitation of premature labor is lowest.

Nonmedicinal and Chemical Use in Pregnancy

327

What are the long-term effects of prenatal methadone exposure? Longitudinal studies of infants to age 2 have found mental and physical development to be well within the normal range, although some differences from controls were noted. When evaluated at ages 4 and 5, no differences were found from controls, but both groups had low scores, most likely an effect of low socioeconomic status (53,54). Accepting the difficulties of attempting to control for confounding variables, it appears that methadone in utero is one of the least important factors in the developmental outcome of these children. Cocaine and Other Stimulants The use of the stimulant cocaine increased rapidly in the past decade but now appears to have passed its peak. By 1986, nearly 40% of the U.S. population in the age range of 25–30 years had tried cocaine (55,56). Women of childbearing age are of particular concern because of the potential risk to the fetus. The first report of cocaine effects on pregnancy outcome appeared in 1985 (11). Spontaneous abortions, prematurity, and intrauterine growth retardation are increased in cocaine-using mothers. Labor may be precipitated following a large intravenous bolus of cocaine, and there is an increase in the incidence of abruptio placentae (57–60). Several infants have been reported with perinatal cerebral infarction (59). These adverse effects can be explained by the physiological alterations of acute cocaine intoxication. Since the first description of pregnancy outcome after cocaine use, scores of studies have been published; these are analyzed in Chapters 18 and 21. A withdrawal syndrome and neurobehavioral abnormalities have been described in a small number of infants of cocaine-using mothers; however, other drugs could not be ruled out (11). There does not appear to be an increase in the rate of major congenital abnormalities. Because of the small sample size, difficulty in controlling for other drug use, and differing inclusion criteria in all the reports above, these findings must be considered to be preliminary. Management of the mother depends on the extent of her drug use as well as concurrent medical and social problems. Many women can be managed as outpatients. Cocaine use often follows a cyclical pattern. Periods of abstinence of one to several weeks do not imply a less severe dependence, and counseling aimed at preventing relapse is crucial. Residential treatment should be made available to patients unable to abstain. Cocaine withdrawal consists of the ‘‘crash’’: several hours to days of depression, hypersomnolence, and hyperphagia, followed by a period of anergia and intense cocaine craving (56). No specific pharmacological therapy is required, since this syndrome does not appear to have any serious consequences to mother or fetus unless depression is severe enough to cause suicidal ideation. Heavy cocaine users often take central nervous system depressants such as alcohol and benzodiazepines for symptom relief during the crash phase and may have significant cross-dependence on these agents. A complete drug and alcohol history should be obtained to identify patients at risk for depressant withdrawal, since this information is not always volunteered. Amphetamines and other stimulants have a pharmacological profile similar to that of cocaine. Owing to restrictions on prescription use and limited illicit availability, the abuse of stimulants with the exception of cocaine is now a rarity. The Collaborative Perinatal Project found no evidence of teratogenesis with this class of drugs (61). Schardein, who reviewed the reports citing increased malformations, concluded that the evidence is

328

Schneiderman

generally negative (42). Clinically, the amphetamine-using mother presents in much the same way as the cocaine user and should be managed similarly (60,62). Cannabinoids The prevalence of marijuana use at the time of conception, which has been reported as 9–20% (9,63), varies with the population studied. More than half these women become abstainers during the first trimester. Although marijuana is not usually thought to be associated with a physical withdrawal syndrome, many regular users report irritability, sleep disturbance, and decreased appetite upon abrupt cessation (64). We have had a number of pregnant patients complain of severe nausea and vomiting on discontinuing marijuana smoking in the first trimester. This is consistent with the antiemetic effect of the cannabinoids. Several studies have reported a decrease in birth weight and an increase in the prevalence of small-for-gestational-age babies and preterm infants among marijuana users (65,66). This has not been a consistent finding; it is likely related to a wide range of intake. Wu et al. reported a nearly fivefold greater increment in the blood carboxyhemoglobin level and a threefold increase of inhaled tar with smoking of marijuana compared with tobacco cigarettes (67). Thus, effects similar to those in tobacco users would be expected in heavy marijuana users in addition to any effect from its active ingredient, ∆-9-tetrahydrocannabinol. Marijuana has not been associated with congenital abnormalities in humans. Fried found a doserelated association between marijuana smoking and alterations in visual responsiveness as well as increased tremors and startle reflexes in the newborn (9). Hallucinogens, Phencyclidine, and Solvents Prenatal exposure to lysergic acid diethylamide (LSD) has been implicated in an increased incidence of limb defects (68) and central nervous system and ocular abnormalities (42), but the evidence for teratogenic effects in humans is not convincing. With the widespread use of LSD in the 1970s, it is unlikely that a major teratogenic effect would have been missed. An early ultrasound examination is nonetheless recommended to rule out serious dysmorphogenesis. Phencyclidine (PCP) was used by 0.8% of pregnant patients in one study; 7.3% gave a history of past use (69). A single case report describes an infant born with abnormal appearance and behavior who subsequently developed a spastic quadriparesis (70). Chasnoff et al. described seven infants with sudden outbursts of agitation and rapid changes in the level of consciousness at birth but no congenital abnormalities (71). Recreational solvent abusers choose toluene-containing products preferentially. Toluene can be obtained in relatively pure form in a variety of readily available and inexpensive products, such as lacquer thinner and contact cement cleaner. Acute solvent intoxication is similar to that of alcohol; there is no significant withdrawal syndrome (72). With chronic use, a persistent encephalopathy may develop, characterized by signs of cortical and cerebellar dysfunction. This is at least partially reversible with abstinence (73). The evidence for teratogenicity from occupational exposure is inconclusive. Hersh reported three infants whose mothers regularly inhaled a toluene product throughout pregnancy. All the infants had microcephaly, central nervous system dysfunction, attention deficits, developmental delay, growth deficiency, and facial dysmorphology—a pattern quite similar to that of fetal alcohol syndrome (74). Users tend to have deprived backgrounds and

Nonmedicinal and Chemical Use in Pregnancy

329

limited social supports; inpatient treatment is usually indicated to initiate abstinence, but prognosis is generally poor. Withdrawal from hallucinogens, PCP, and solvents can usually be managed supportively. When severe agitation in the intoxication phase is unresponsive to reassurance, benzodiazepines can be administered. The acutely psychotic patient may require a neuroleptic; haloperidol, a butyrophenone, is usually better tolerated than the phenothiazines. Benzodiazepines Benzodiazepines are in widespread use therapeutically and are frequently consumed by polydrug abusers. Although isolated abuse of high doses of benzodiazepines is rate, it is not uncommon to find women who have been maintained on chronic therapeutic doses for many years. In 1975 two groups reported an increase in risk of cleft lip and palate with first-trimester exposure to benzodiazepines (75,76). A subsequent case-control study did not confirm this effect (77). The newer benzodiazepines have not been associated with specific abnormalities. Seven cases of a fetal benzodiazepine syndrome have been reported, but without adequate controls it is difficult to attribute a causative role (78). Given the high prevalence of benzodiazepine use, a human teratogenic effect is not likely to be large if it exists at all (79). A woman planning a pregnancy should be encouraged to discontinue her benzodiazepine use prior to conception (80). In counseling the woman with a history of first-trimester use, the clinician should provide a balanced summary of the reported effects. Although benzodiazepines are probably not teratogenic in humans, an ultrasound examination can be reassuring even if it does not guarantee the absence of abnormalities. A neonatal benzodiazepine abstinence syndrome has been reported (81), so detoxification should be advised regardless of the time during pregnancy at which the woman presents. Attempts at withdrawal are often unsuccessful when abrupt cessation leads to uncomfortable symptoms. In addition to anxiety and insomnia, these patients characteristically complain of sensory misperceptions and illusions as well as depersonalization. A tapering regimen with diazepam obviates the fluctuations in drug levels that occur with the shorter-acting benzodiazepines. A rough guideline for converting to an equivalent dose of diazepam is to assume that the largest unit dose available is equivalent to 10 mg diazepam per week with weekly outpatient supportive counseling (82). Abstinence can usually be achieved in 6–8 weeks. If outpatient detoxification is unsuccessful, a rapid diazepam taper may be tried in an inpatient unit. If abstinence cannot be achieved, the patient should be maintained on the lowest possible dose. Barbiturates and Hypnosedatives Owing to their limited availability, the barbiturates and nonbarbiturate hypnosedatives (chloral hydrate, glutethimide, meprobamate, methaqualone) are no longer commonly abused substances. Butalbital, available only as a proprietary combination with aspirin and caffeine with or without codeine (Fiorinal-C), is the only barbiturate we see frequently (83). Heinonen et al., in the Boston Collaborative Study, found little evidence for a teratogenic effect in this group of drugs (61). Because of the danger of a major withdrawal syndrome (tonic-clonic seizures and delirium), the pregnant woman ingesting more than 400 mg of barbiturate or equivalent daily should be admitted to the hospital for detoxification. Phenobarbital is the drug of choice, given either on a tapering schedule (84) or by an oral loading technique (85). A neonatal abstinence syndrome has been described.

330

Schneiderman

Tobacco Annual per capita cigarette consumption has decreased steadily in North America since 1973. However, the decrease has been much less in women than in men and the percentage of heavy smokers has increased, perhaps suggesting that many less severely dependent individuals have successfully quit (86). For married white women age 20 and over, the prevalence of smoking during pregnancy decreased from 40 to 25% between 1967 and 1980. This drop was greatest in women with high educational levels (87). The smoker is 80% more likely than the nonsmoker to have a spontaneous abortion and is 100% more likely to have a low-birth-weight infant (⬍2500 g). Cigarette smoking in pregnancy is associated with decreased birth weight by an average of 200 g. This effect is dose-dependent, is independent of other factors, and can be reversed by cessation of smoking in early pregnancy. Smoking during pregnancy leads to increased perinatal mortality, prematurity, placenta previa, and abruption (88,89). Although some studies have shown an increase in congenital malformations, most have not, and summing the data, we can say that tobacco smoking does not appear to cause human teratogenicity. Sexton and Hebel performed a randomized controlled study of an intensive smokingcessation intervention during pregnancy. Prior to randomization, both groups had decreased their mean number of cigarettes smoked by half and about 15% had quit. The reported quitting rate at 8 months of gestation was 20% for the control group and 43% for the treatment group (91). Use of a self-help booklet has been shown to increase smoking cessation in public health maternity clinics (92). Education and support to decrease smoking can be effective in patients who do not spontaneously change their smoking behavior at the onset of pregnancy (93) and may result in improved fetal outcome. Nicotine chewing gum, although a useful adjunct to smoking cessation therapy, should not be used in pregnancy. Caffeine The average intake of caffeine in pregnancy is 144 mg/day, or about one cup of filtered coffee or two cups of tea (94). Several studies have demonstrated little or no fetal effect of caffeine (95,96). Animal studies using high doses and some human studies reporting adverse outcome have led most clinicians to recommend caution in beverage caffeine consumption. Since many of the adverse physiological effects of caffeine do not occur at doses below 400 mg/day (97), it is reasonable to limit consumption to this level. SUMMARY The mood-altering drugs with the exception of alcohol and possibly organic solvents have not been documented to be major teratogens. Most have been associated with intrauterine growth retardation and a neonatal withdrawal syndrome at birth in infants of chronic, heavy users. In many women, a brief intervention consisting of education and support in the primary care setting can limit the morbidity of nonmedical drug use. A comprehensive prenatal program for drug-dependent women can help improve outcome in this high-risk population. Clinical Case Answer It is important to link this patient without delay with a program that specializes in addiction. Therapy should include all domains of life, since addiction cannot be effectively

Nonmedicinal and Chemical Use in Pregnancy

331

treated without accounting for its psychosocial aspects. Pharmacologically, most favorable experience comes from protocols using the long-acting opioid methadone. FURTHER READING Finnegan LP ed. Drug Dependence in Pregnancy: Clinical Management of Mother and Child. Washington, DC: National Institute on Drug Abuse, 1979. Lester BM ed. Prenatal drug exposure and child outcome. Clin Perinatol 1999; 26:1–250.

REFERENCES 1. Jaffe JH. Drug addiction and drug abuse. In: Gilman AG, Goodman LS, Rall TW, Murad F, eds. The Pharmacological Basis of Therapeutics, 7th ed. New York: Macmillan, 1985, pp 532–581. 2. Nicholi AM. The nontherapeutic use of psychoactive drugs. N Engl J Med 1983; 308:925– 933. 3. Smart RG, Adlaf EM. Alcohol and other drug use among Ontario adults. Toronto, Ontario, Canada: Alcoholism and Drug Addiction Research Foundation, 1987, pp 1–57. 4. Rayburn W, Wible-Kant J, Bledsoe P. Changing trends in drug use during pregnancy. J. Reprod Med 1982; 27:569–575. 5. Fried PA, Watkinson B, Grant A, Knights RM. Changing patterns of soft drug use prior to and during pregnancy: a prospective study. Drug Alcohol Depend 1980; 6:323–343. 6. Koren G, Bologa-Campeneau M, Long D, et al. The perception of teratogenic risk by pregnant women exposed to drugs and chemicals during the first trimester. Am J Obstet Gynecol 1989; 160:1190–1194. 7. Chasnoff IJ, Schnoll SH, Burns WJ, Burns K. Maternal non-narcotic substance abuse during pregnancy: effects on infant development. Neurobehav Toxicol Teratol 1984; 6:277–280. 8. Landesman-Dwyer S, Keller S, Streissguth AP. Naturalistic observations of newborns: effects of maternal alcohol intake. Alcoholism Clin Exp Res 1978; 2:171–177. 9. Fried PA. Marijuana use by pregnant women: neurobehavioral effects in neonates. Drug Alcohol Depend 1980; 6:415–424. 10. Rosen TS, Johnson HL. Methadone exposure: effects on behavior in early infancy. Pediatr Pharmacol 1982; 2:192–196. 11. Chesnoff IJ, Burns WJ, Schnoll SH, Burns KA. Cocaine use in pregnancy. N Engl J Med 1985; 313:666–669. 12. Fletcher JC, Evans MI. Maternal bonding in early fetal ultrasound examinations. N Engl J Med 1983; 308:392–393. 13. Regan DO, Ehrlich SM, Finnegan LP. Infants of drug addicts: at risk for child abuse, neglect, and placement in foster care. Neurotoxicol Teratol 1987; 9:315–319. 14. Scott GB, Fischl MA, Klimas N, et al. Mothers of infants with the acquired immunodeficiency syndrome: evidence for both symptomatic and asymptomatic carriers. JAMA 1985; 253:363– 366. 15. Thomas PA, Lubin K, Milberg J, et al. Cohort comparison study of children whose mothers have acquired immunodeficiency syndrome and children of well inner city mothers. Pediatr Infect Dis 1987; 6:247–251. 16. Friedland GH, Klein RS. Transmission of the human immunodeficiency virus. N Engl J Med 1987; 317:1125–1135. 17. Guinan ME, Hardy A. Epidemiology of AIDS in women in the United States 1981 through 1986. JAMA 1987; 257:2039–2042.

332

Schneiderman

18. Mok JQ, Guaguinto C, De Rossi A, et al. Infants born to mothers seropositive for human immunodeficiency virus. Lancet 1987; 1:1164–1168. 19. Jones KL, Smith DW. Recognition of the fetal alcohol syndrome in early infancy. Lancet 1973; 2:999–1001. 20. Jones KL, Smith DW, Ulleland CN, et al. Pattern of malformation in offspring of chronic alcoholic mothers. Lancet 1973; 1:1267–1271. 21. Clarren SK, Smith DW. The fetal alcohol syndrome. N Engl J Med 1978; 298:1063–1067. 22. Jones KL, Smith DW, Streissguth AP, Myrianthopoulos NC. Outcome in offspring of chronic alcoholic women. Lancet 1974; 1:1076–1078. 23. Hanson JW, Jones KL, Smith DW. Fetal alcohol syndrome: experience with 41 patients. JAMA 1976; 235:1458–1460. 24. Streissguth AP, Herman CS, Smith DW. Intelligence, behavior, and dysmorphogenesis in the fetal alcohol syndrome: a report on 20 patients. J Pediatr 1978; 92:363–367. 25. Rosett HL. A clinical perspective of the fetal alcohol syndrome. Alcohol Clin Exp Res 1980; 4:119–122. 26. Sokol RJ, Miller SI. Reed G. Alcohol abuse during pregnancy: an epidemiologic study. Alcohol Clin Exp Res 1980; 4:135–145. 27. Spohr HL, Steinhausen HC. Follow-up studies of children with fetal alcohol syndrome. Neuropediatrics 1987; 18:13–17. 28. Kaminski M, Rumeau-Rouquette C, Schwartz D. Alcohol consumption in pregnant women and the outcome of pregnancy. Alcohol Clin Exp Res 1978; 2:155–163. 29. Ouellette EM, Rosett HL, Rosman NP, Weiner L. Adverse effects on offspring of maternal alcohol abuse during pregnancy. N Engl J Med 1977; 297:528–530. 30. Little RE, Sing CF. Association of father’s drinking and infant’s birth weight. N Engl J Med 1986; 314:1644–1645. 31. Hanson JW, Streissguth AP, Smith DW. The effects of moderate alcohol consumption during pregnancy on fetal growth and morphogenesis. J Pediatr 1978; 92:457–460. 32. Mills JL, Graubard BI. Is moderate drinking during pregnancy associated with an increased risk for malformations? Pediatrics 1987; 80:309–314. 33. Marbury MC, Linn S, Monson R, et al. The association of alcohol consumption with outcome of pregnancy. Am J Public Health 1983; 73:1165–1168. 34. Rosett HL, Weiner L. Alcohol and the Fetus: A Clinical Perspective. New York: Oxford University Press, 1984. 35. The effects of alcohol on the outcome of pregnancy. Bull Soc Obstet Gynaecol Can 1984; 6: 1–2. 36. Little RE, Streissguth AP. Drinking during pregnancy in alcoholic women. Alcohol Clin Exp Res 1978; 2:179–182. 37. Little RE, Schultz FA, Mandell W. Drinking during pregnancy. J Stud Alcohol 1976; 37:375– 379. 38. Sellers EM, Naranjo CA, Harrison M, et al. Diazepam loading: simplified treatment of alcohol withdrawal. Clin Pharmacol Ther 1983; 34:822–826. 39. Briggs GG, Freeman RK, Yaffe SJ. Drugs in Pregnancy and Lactation. Baltimore: Williams & Wilkins, 1986. 40. Favre-Tissot M, Delatour P. Psychopharmacologie et teratogenesis a propos du disulfirame: essai experimental. Ann Med Psychol 1965; 123:735–740. 41. Nora AH, Nora JJ, Blu J. Limb-reduction anomalies in infants born to disulfiram-treated alcoholic mothers. Lancet 1977; 2:664. 42. Schardein JL. Chemically Induced Birth Defects. New York: Marcel Dekker, 1985. 43. Zelson C, Rubio E, Wasserman E. Neonatal narcotic addition: 10 year observation. Pediatrics 1971; 48:178–189. 44. Chavez CJ, Ostrea EM, Stryker JC, et al. Sudden infant death syndrome among infants of drug-dependent mothers. J Pediatr 1979; 95:407–409.

Nonmedicinal and Chemical Use in Pregnancy

333

45. Blinick G, Wallach RC, Jerez E. Pregnancy in narcotics addicts treated by medical withdrawal. Am J Obstet Gynecol 1969; 105:997–1003. 46. Finnegan LP, ed. Drug Dependence in Pregnancy: Clinical Management of Mother and Child. Washington, DC: National Institute on Drug Abuse, 1979. 47. Zuspan FP, Gumpel JA, Mejia-Zelaya A, et al. Fetal stress from methadone withdrawal. Am J Obstet Gynecol 1975; 122:43–46. 48. Rementeria JL, Nunag NN. Narcotic withdrawal in pregnancy: stillbirth incidence with a case report. Am J Obstet Gynecol 1973; 116:1152–1156. 49. Connaughton JF, Reeser D, Schut J, Finnegan LP. Perinatal addiction: outcome and management. Am J Obstet Gynecol 1977; 129:679–686. 50. Kandall SR, Albin S, Lowinson J, et al. Differential effects of maternal heroin and methadone use on birthweight. Pediatrics 1976; 58:681–685. 51. Ostrea EM, Chavez CJ, Strauss ME. A study of factors that influence the severity of neonatal narcotic withdrawal. J Pediatr 1976; 88:642–645. 52. Strauss ME, Andresko M, Stryker JC, Wardell JN. Relationship of neonatal withdrawal to maternal methadone dose. Am J Drug Alcohol Abuse 1976; 3:339–345. 53. Kaltenbach K, Finnegan LP. Perinatal and developmental outcome of infants exposed to methadone in utero. Neurotoxicol Teratol 1987; 9:311–313. 54. Kaltenbach K, Finnegan LP. Developmental outcome of children born to methadone maintained women: a review of longitudinal studies. Neurobehav Toxicol Teratol 1984; 6:271– 275. 55. Abelson HI, Miller JD. A decade of trends in cocaine use in the household population. Natl Inst Drug Abuse Res Monogr Ser 1985; 61:35–49. 56. Gawin FH, Ellinwood EH. Cocaine and other stimulants: actions, abuse, and treatment. N Engl J Med 1988; 318:1173–1182. 57. Madden JD, Payne TF, Miller S. Maternal cocaine abuse and effect on the newborn. Pediatrics 1986; 77:209–211. 58. Bingol N, Fuchs M, Diaz V, et al. Teratogenicity of cocaine in humans. J Pediatr 1987; 110: 93–96. 59. Chasnoff IJ, Bussey ME, Savich R, Stack CM. Perinatal cerebral infarction and maternal cocaine use. J Pediatr 1986; 108:456–459. 60. Oro AS, Dixon SD. Perinatal cocaine and methamphetamine exposure: maternal and neonatal correlates. J Pediatr 1986; 108:456–459. 61. Heinonen OP, Slone D, Shapiro S. Birth Defects and Drugs in Pregnancy. Littleton, CO: PSG Publishing, 1977. 62. Eriksson M, Larsson G, Winbladh B, Zetterstrom R. The influence of amphetamine addiction on pregnancy and the newborn infant. Acta Paediatr Scand 1978; 67:95–99. 63. Hatch EE, Bracken MB. Effect of marijuana use in pregnancy on fetal growth. Am J Epidemiol 1986; 124:986–993. 64. Hollister LE. Health aspects of cannabis. Pharmacol Rev 1985; 38:1–20. 65. Linn S, Schoenbaum SC, Monson RR, et al. The association of marijuana use with outcome of pregnancy. Am J Public Health 1983; 73:1161–1164. 66. Hingson R, Alpert JJ, Day N, et al. Effects of maternal drinking and marijuana use on fetal growth and development. Pediatrics 1982; 70:539–546. 67. Wu TC, Tashkin DP, Djahed B, Rose JE. Pulmonary hazards of smoking marijuana as compared with tobacco. N Engl J Med 1988; 318:347–351. 68. Long S. Does LSD induce chromosomal damage and malformations? A review of the literature. Teratology 1972; 6:75–90. 69. Golden NL, Kuhnert BR, Sokol RJ, et al. Phencyclidine use during pregnancy. Am J Obstet Gynecol 1984; 148:254–259. 70. Golden NL, Sokol RJ, Rubin IL. Angel dust: possible effects on the fetus. Pediatrics 1980; 65:18–20.

334

Schneiderman

71. Chasnoff IJ, Burns WJ, Hatcher RP, Burns KA. Phencyclidine: effects on the fetus and neonate. Dev Pharmacol Ther 1983; 6:404–408. 72. Hayden JW, Comstock EG, Comstock BS. The clinical toxicology of solvent abuse. Clin Toxicol 1976; 9:169–184. 73. King MD, Day RE, Oliver JS, et al. Solvent encephalopathy. BMJ 1981; 283:663–665. 74. Hersh JH, Podruch PE, Rogers G, Weisskopf B. Toluene embryopathy. J Pediatr 1985; 106: 922–927. 75. Saxen I, Saxen L. Association between maternal intake of diazepam and oral clefts. Lancet 1975; 2:498. 76. Safra MJ, Oakley GP. Association between cleft lip with or without cleft palate and prenatal exposure to diazepam. Lancet 1975; 2:478–480. 77. Rosenberg L, Mitchell AA, Parsells JL, et al. Lack of relation of oral clefts to diazepam use during pregnancy. N Engl J Med 1983; 309:1282–1285. 78. Laegrid L, Olegard R, Wahlstrom J, Conradi N. Abnormalities in children exposed to benzodiazepines in utero. Lancet 1987; 1:108–109. 79. Weber LWD. Benzodiazepines in pregnancy—academic debate or teratogenic risk? Biol Res Pregnancy 1985; 6:151–167. 80. Loudon JB. Psychotropic drugs. BMJ 1987; 294:167–169. 81. Romenteria JL, Bhatt K. Withdrawal symptoms in neonates from intrauterine exposure to diazepam. Pediatr Pharmacol 1977; 90:123–126. 82. Busto U, Sellers EM, Naranjo CA, et al. Withdrawal reaction after long-term therapeutic use of benzodiazepines. N Engl J Med 1986; 315:854–859. 83. Devenyi P, Rideout J, Schneiderman J. Abuse of a commonly prescribed analgesic preparation. Can Med Assoc J 1985; 133:294–296. 84. Smith DE, Wesson DR. Phenobarbital technique for treatment of barbiturate dependence. Arch Gen Psychiatry 1971; 24:56–60. 85. Robinson GM, Sellers EM, Janecek E. Barbiturate and hypnosedative withdrawal by a multiple oral phenobarbital loading dose technique. Clin Pharmacol Ther 1981; 309:71–76. 86. Fielding JE. Smoking: health effects and control. N Engl J Med 1985; 313:491–498. 87. Kleinman JC, Kopstein A. Smoking during pregnancy, 1967–80. Am J Public Health 1987; 77:823–825. 88. Murphy JF, Mulcahy R. The effect of age, parity, and cigarette smoking on baby weight. Am J Obstet Gynecol 1971; 111:22–25. 89. Landesman-Dwyer S, Emanuel I. Smoking during pregnancy. Teratology 1979; 19:119–126. 90. Hebel JR, Nowicki P, Sexton M. The effect of antismoking intervention during pregnancy: an assessment of interactions with maternal characteristics. Am J Epidemiol 1985; 122:135– 148. 91. Sexton M, Hebel JR. A clinical trial of change in maternal smoking and its effect on birth weight. JAMA 1984; 251:911–915. 92. Windsor RA, Cutter G, Morris J, et al. The effectiveness of smoking cessation methods for smokers in public health maternity clinics: a randomized trial. Am J Public Health 1985; 75: 1389–1392. 93. Donovan JW. Randomized controlled trial of anti-smoking advice in pregnancy. Br J Prev Soc Med 1977; 31:6–12. 94. Morris MB, Weinstein L. Caffeine and the fetus—is trouble brewing? Am J Obstet Gynecol 1981; 140:607–610. 95. Rosenberg L, Mitchell AA, Shapiro S, Slone D. Selected birth defects in relation to caffeinecontaining beverages. JAMA 1982; 247:1429–1432. 96. Linn S, Schoenbaum SC, Monson RR, et al. No association between coffee consumption and adverse outcomes of pregnancy. N Engl J Med 1982; 306:141–145. 97. Curatolo PW, Robertson D. The health consequences of caffeine. Ann Intern Med 1983; 98: 641–653.

18 Maternal and Obstetrical Effects of Prenatal Drug Exposure Raafat Bishai and Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

INTRODUCTION Recent research on drug abuse during pregnancy has largely focused on fetal outcome, with very little attention to the biological and psychological effects on the mother. This review focuses on determinants of maternal health in drug-dependent women. These, directly or indirectly, are affecting fetal outcome, and careful attention to them is needed if one wishes to optimize perinatal outcome. Nonmedical drug use is defined as the use of any psychoactive drug in the absence of a clearly defined medical indication (23). While recreational drug users and drug-dependent women may appear unconvinced about health risk themselves, they, too, usually wish to spare their unborn children from risks associated with drug use (51). A Swedish study has shown that almost one-third of chemically dependent women succeeded in overcoming their addiction in early pregnancy once they realized that they were pregnant (12), indicating that pregnancy may, by itself, provide powerful motivation for women to overcome addiction. Another German study (50) documented that pregnancy and subsequent motherhood enhance and stabilize the motivation of the mother to quit drug consumption. The outcome of pregnancy in the drug-dependent woman depends on a complex of determinants far beyond the drug-taking behavior itself. The pharmacological effects of the abused substance are often compounded by suboptimal prenatal care, poor diet, higher risk for sexually transmitted diseases, and concomittant use of cigarettes and alcohol (51). Much of the research in this field comes from studies in women of the American innercity drug subculture, which may not be easily extrapolated to either women who have used a substance once or several times socially or to women who have received a short course of parenteral opioid in the treatment of an acute medical or surgical illness prior to the knowledge of the pregnancy (51). In this paper, we overview the published evidence on maternal complications in pregnancy as a result of drug dependence.

335

336

Bishai and Koren

A GENERAL VIEW ON MATERNAL COMPLICATIONS Silver and colleagues (53), have shown that drug-dependent women (DDW) have an increased incidence of medical and obstetrical complications; within the context of a family center, their study was undertaken to determine whether the DDW had normal patterns of labor and whether standard intrapartum management was appropriate. The study population included 336 women, of which 112 were DDW (72% of them receiving methadone maintenance). The comparison group of 224 non-drug-dependent women was matched for gravidity, parity, and socioeconomic background. The incidence of premature delivery, abruptio placentae, breech presentation, and intrauterine growth retardation, was significantly higher in the DDW. The average duration of the first, second, and third stages of labor were similar to the normal course of labor in the matched comparison group. Rates of complications during labor and cesarean section were not higher, but there were more than twice as many forceps deliveries, which correlated with a 40% increased use of epidural anesthesia. Of importance, the doses of analgesia and anesthesia needed for the DDW were in excess of those given to the control patients. Postpartum complications were common among the DDW, but most of them were secondary to the need to use subclavian intravenous lines due to the presence of sclerotic peripheral veins. These data suggest that high-risk prenatal management and careful monitoring in the intra- and postpartum periods—as well as the use of epidural anesthesia—usually prevents complications in DDW. Hawthorne and colleagues (20) have shown that patients tested positive for drugs of abuse weighed less, were older, were more prone not to seek prenatal care, and were more likely to deliver prematurely and have a growth-retarded infant than controls. Funkhouser et al. (16) have found that illicit drug use in pregnancy and socioeconomic status were negatively correlated with effective utilization of prenatal care. The following discussion focuses on the effects of specific drugs of abuse; however, it should be remembered that a woman rarely abuses one drug only, and it is very common for a cocaine-using mother, for example, also to consume alcohol, cigarettes, and marijuana. Any attempt to understand the clinical status of the mother must, therefore, address all exposures together.

Table 1 Maternal Complications That May Appear Among DrugDependent Women Poor antenatal care Preterm delivery Abruptio placentae Breech presentation Postpartum complications Sexually transmitted diseases Poor nutrition Psychiatric comorbidity Higher rates of prostitution

Effects of Prenatal Drug Exposure

337

DRUG-SPECIFIC COMPLICATIONS Cocaine The first report of cocaine effects on pregnancy outcome appeared in 1985 (6). The rates of spontaneous abortions, prematurity, and intrauterine growth retardation are increased among cocaine-using mothers. Labor may be precipitated following a large intravenous bolus of cocaine, a method often used by addicted mothers to induce labor, and there is an increase in the incidence of abruptio placentae (34,43). In a case-control study (38) of cocaine use in pregnancy, Miller and colleagues detected complications that occurred significantly more often among study patients—including vaginal bleeding, abruptio placentae, and premature rupture of the membranes— than among controls. Study patients were found to use prenatal care significantly less often (45 vs. 86%). Delaney and his group (8) have shown that women with recent cocaine use present with ruptured membranes at an earlier gestational age and may actually have a longer latency period than women who do not use cocaine. These results were corroborated by Kistin (28), who demonstrated higher risks than among smokers for prematurity, abruptio placentae, placenta previa, and prenatal death. Dinsmoor and colleagues suggested that preterm rupture of membranes associated with recent cocaine use is characterized by advanced cervical dilation at admission and a shorter latency period to labor and delivery (9). Singer et al. (54) concluded that maternal cocaine use predicts negative birth outcome directly as well as through obstetric risk factors of abortion history and less prenatal care. In a retrospective chart review (1), cocaine use (52% of the study group vs. 10% in the general population, p ⬍ 0.01) and HIV positivity (24% in the study group vs. 2% in the general population, p ⬍ 0.01), were significant risk factors for antepartum pneumonia. Moreover, Chan and colleagues (5) documented that cocaine use is a predisposing factor for pneumothorax during pregnancy, with spontaneous pneumothorax carrying a high risk of recurrence, possibly higher if associated with continued cocaine use. Jawahar et al. reported fatal complications in a pregnant woman who was an active cocaine abuser and developed an intestinal gangrene and multiorgan failure (24). Due to needle sharing as well as unplanned and unprotected sex with multiple partners, cocaine-using pregnant women have a higher risk of becoming infected with HIV. In an attempt to detect a biological cause for the high incidence of HIV and other sexually transmitted diseases among cocaine users during pregnancy, Hedstrom and colleagues tried to determine if cocaine suppresses mitogen-induced lymphocyte proliferation in pregnant women, but no such correlation was found in this study (22). Moen reported the case of a woman in her late pregnancy who experienced hepatic rupture associated with cocaine use. Liver damage may result from the potent vasoconstrictor property of cocaine, leading to vasospasm and ischemia (40). The use of cocaine in pregnancy can result in systemic and focal vasoconstriction and abnormal uterine contractions forceful enough to cause complete rupture of a gravid uterus along a previous vertical cesarean section scar, as reported by Mishra et al. (39). Adverse renal effects in pregnant cocaine-using women have been reported, characterized by hematuria, proteinuria, hemolytic anemia, thrombocytopenia, renal insufficiency, and pulmonary edema (4).

338

Bishai and Koren

Lampley and colleagues reported a case of a woman who developed abruptio placentae approximately 18 hours after using cocaine. Six hours later, a cesarean section was performed because of vaginal bleeding and fetal bradycardia. Oliguria was present from admission and persisted despite aggressive management. Serum CPK and urine myoglobin were both elevated. Hemodialysis was required for renal cortical necrosis. Rhabdomyolysis, as indicated by elevated CPK and the presence of myoglobin in the urine, suggested that myoglobinuria may contribute to the acute renal failure seen in cases of cocaineinduced abruptio placentae (30). As suggested by Towers and colleagues (60), cocaine exposure in pregnant women can present itself as preeclampsia. In attempting to better understand the pathophysiological changes induced by cocaine during gestation, Smith concluded that the drug may downregulate β-adrenergic receptors in human myometrium. This could result in a decreased capacity of the uterus to relax and, consequently, may predispose it to preterm labor (55). Erratic binge use of cocaine results in perinatal complications that may be as severe as those occurring with regular daily use. In women with erratic binging (variable in interval, duration, and amount) acute adverse effects (vaginal bleeding, abruptio placentae, still-births) were most prevalent, whereas in women who binge daily, chronic problems are more prominent (systemic infections, anemia, and low maternal weight) and significant (3). In an alerting message to family physicians who practice obstetrics, Fox summarized the potentially catastrophic maternal outcomes in cocaine-abusing pregnant women (15), including acute pulmonary edema, seizures, cardiac arrhythmia, and sudden death. In an attempt to understand the biological nature of the effects of cocaine in pregnancy, cardiovascular and neurological responses to cocaine during pregnancy are enhanced when compared with responses in nonpregnant subjects to the same dose per kilogram or to metabolites of crack cocaine (64). In addition, the placenta itself is a direct target organ for cocaine toxicity, and interaction of cocaine with specific placental proteins in the placenta may play an important role (17). Cocaine has been shown to inhibit specific transport of amino acids across the placenta; this is similar to the action of amphetamine. In a meta-analysis, we detected relatively few adverse reproductive effects associated with pregnancy exposure to cocaine, when the control groups were made up of polydrug users who did not use cocaine. This may indicate that other confounders, such as smoking and drinking, may contribute to adverse pregnancy outcome (33). For example, even placental abruption is not more common among cocaine users in this analysis as compared to polydrug (noncocaine) users.

Amphetamine Clinically, amphetamine-using mothers have been reported to present in much the same way as cocaine users and should probably be managed similarly (13,43). Elliott and Rees reported a case of a woman consuming amphetamine in late pregnancy who presented with confusion, agitation, and convulsions. Hypertension and proteinuria led to a diagnosis of eclampsia, for which a cesarean section was performed (10). Cardiovascular collapse during anesthesia for cesarean section has been reported in cases of chronic abuse (49,56), probably because chronic ingestion of amphetamine de-

Effects of Prenatal Drug Exposure

339

Table 2 Maternal Complications That May Appear Among Cocaine- and AmphetamineAbusing Pregnant Women Vaginal bleeding Abruptio placentae Placenta praevia Premature rupture of membranes Abortions Poor antenatal care Pneumothorax Pneumonia Malnutrition Seizures (25,41) Sexually transmitted diseases Psychiatric comorbidity History of sexual and psychological abuse

pletes catecholamine in adrenergic neurons, thus impairing the response to sympathomimetics and inhibiting the stress response. A cohort study of pregnant patients using prescribed or illicit amphetamine (52 and 237 patients, respectively) did not detect an increased incidence of placental abruption. It is possible that studies that detect an increased incidence of abruptio placentae include large numbers of patients who use multiple substances, including cocaine (25). It has been postulated that the human placenta is a direct target organ for amphetamine toxicity, as is the case with cocaine. Ramamoorthy and colleagues have showed that amphetamine and methamphetamine are potent competitive inhibitors of the placental norepinephrine transporter and, to a lesser extent, of the placental serotonin transporter. It is possible that increased levels of these vasoactive monoamines in the intervillous space may explain the clinical findings, which include maternal hypertension, fetal growth retardation, decreased uteroplacental blood flow, premature labor, and abruptio placentae (46). Opioids Among heroin-addicted women, the obstetric complications seen most frequently included premature labor, breech presentation, premature rupture of membranes, and toxemia (57). The more frequently noticed medical complications were anemia, syphilis, and infectious hepatitis, although maternal deaths are very rare. Many of these complications could be at least partially explained on the basis of no or inadequate prenatal care. Only one-quarter of the patients had registered in a prenatal clinic, and more than half of this small group did so only in the last trimester. The average number of clinic visits was less than one per patient. The average length of observed labor was considerably shortened, and many patients delivered their babies within 3 hours, with more than half having their babies in less than 7 hours. Most patients delayed coming to the hospital until they felt they were ready for delivery, partially in order to avoid a long labor without access to the opioids and thus also withdrawal. As a result of this delay, deliveries at home, in the ambulance, or on the

340

Bishai and Koren

stretcher were not uncommon. Appropriate counseling and reassurance that withdrawal will not happen with appropriate use of methadone, for example, may partially address this grim tendency. It has been suggested that pregnant patients on heroin may have increased uterine irritability (7), partially explaining the irregular progress of delivery. Ellwood and colleagues (11) conducted a retrospective analysis of the outcome of pregnancy among 174 women abusing opioids. The group was characterized by a high prevalence of previous obstetric and medical problems; the most common complications were preterm labor (24%) and anemia (12%). In another retrospective case-control study (31) carried out on 51 Chinese gravidas who had abused narcotics (mainly heroin) and who gave birth in a teaching hospital in Hong Kong, the major antenatal complications were late antenatal booking (average 28 weeks), prematurity (41%), antepartum hemorrhage (13.7%), and high prevalence of disease (23.5%). Of 51 pregnancies in addicted pregnant women (50 of them on heroin), 10 (20%) resulted in spontaneous abortion, 2 (4%) in extrauterine pregnancy, and 17 (36%) in preterm delivery (61). The apparently higher incidence of extrauterine pregnancy could be explained by the high incidence of previous salpingo-oophoritic disorders that frequently affect these subjects; moreover, the high incidence of miscarriages in the same subjects could have been due to their lifestyle and poor prenatal care or to an increased excitability of the myometrium. Preterm delivery and short labor seem to result from increased excitability of the myometrium, mainly during the period of drug craving, further emphasizing the need for appropriate replacement therapy. In a French study, lack of follow up, prematurity, short child stature, infections (notably with HIV) and the neonatal withdrawal syndrome (2) were commonly detected. The worse obstetric and prenatal outcomes have been encountered among HIV-seropositive drug addicts, and a positive correlation was found between the number of infectious episodes during pregnancy and the mean dose of opioid consumed daily, suggesting that regardless of the maternal serological status, opioid intake not only causes a further worsening of gestational and prenatal outcomes but also increases the mother’s susceptibility to infection (37). The study of Rodriguez et al. found that increased probability of a positive maternal HIV culture on delivery was significantly associated with prenatal use of hard drugs (48). In a study of a large obstetric service in Dallas (32), pregnancy outcome and health status of infants born to 24 heroin addicts were compared to those in a group of 100 unexposed women and their infants. Women who used heroin during pregnancy tended to use other substances (tobacco, alcohol, cocaine) more often than did controls. The frequency of preterm birth was increased significantly in women who abused heroin during pregnancy. Supraovulation and multiple births were found to be higher among opioidaddicted women, exposing them to the known hazards of this medical condition (47). In a retrospective Spanish study, the incidence of maternal syphilis, positivity to hepatitis B surface antigen, premature rupture of membranes, and abortions was significantly more common among heroin-addicted mothers (42). Data on the pre- and postnatal effects of pyribenzamine (‘‘Ts and blues’’) abuse in pregnancy were compared with those of a control group. Anemia, syphilis, hepatitis, and gonorrhea were more common among mothers abusing Ts and blues (62). Acute bacterial endocarditis, a fulminating disorder most often caused by Staphylococcus aureus, is uncommon in pregnancy. However, the frequency of this disease may

Effects of Prenatal Drug Exposure

341

Table 3 Maternal Complications That May Appear Among OpioidAbusing Pregnant Women Premature labor Premature rupture of membranes Breech presentation Antepartum hemorrhage Toxemia Poor antenatal care Anemia Uterine irritability Infections (HIV, hepatitis, syphilis)

be increasing due to increased intravenous drug abuse. Three cases were reported during pregnancy at Charity Hospital, New Orleans, and all three patients were intravenous drug abusers (44). Alcohol Although there is no known safe lower limit for alcohol consumption in pregnancy, two recent meta-analyses have failed to show increased risk of major malformations and perinatal complication in babies of social drinkers (less than one drink per day) (35,45). Abstinence is the most conservative approach. However, half of all pregnancies are unplanned and half of all pregnant women drink socially before being aware of their pregnancy. Hence, in about one-quarter of all pregnancies, the fetus is exposed to some alcohol (51). The results of our recent meta-analyses may be used to reassure such women. A study of more than 12,000 pregnancies showed an increase in rates of abruptio placentae but not in other adverse outcomes associated with an intake of less than two drinks per day (36), suggesting that only persistent and higher doses of ethanol increase obstetric risk in a clinically meaningful manner. Adverse outcomes in heavy drinkers include increased spontaneous abortions (59), premature placental separation (26), and, of course, the risks associated with alcoholrelated birth defects. It has been shown repeatedly that among alcoholic women, the risk of having a child with fetal alcohol syndrome (FAS) increases with gravidity, It appears that this risk is not due to increased drinking with age. Rather, it may be associated with increased maternal morbidity with longer drinking history, including hepatic, gastrointestinal, and neurological sequelae, to mention a few. Pregnancy complicated by pancreatitis may lead to substantial fetal and maternal morbidity and mortality. Among 30 women who developed pancreatitis during pregnancy, the etiology was gallstones in 22 patients, alcohol in 2 patients, and idiopathic in 6 patients (58). Organic Solvents Recreational abusers of solvents tend to choose toluene-containing products. Acute solvent effects resemble those of alcohol; however, there is no typical withdrawal syndrome (21).

342

Bishai and Koren

Table 4 Maternal Complications That May Appear Among SolventAbusing Pregnant Women Renal tubular acidosis Hypokalemia Cardiac dysrythmias Rhabdomyolosis Cortical and cerebellar dysfunction

With chronic use, a persistent encephalopathy, characterized by signs of cortical and cerebellar dysfunction may develop in the mother (27). In another study, abuse of toxic vapors during pregnancy was found to be associated with increased maternal and fetal morbidity. Toluene-induced renal tubular acidosis occurred in over half of these women and was clustered among long-duration abusers. The renal acidosis placed the mother at risk of hypokalemia, with associated cardiac dysrythmias and rhabdomyolosis (63). Cannabinoids Pregnant patients often complain of severe nausea and vomiting on discontinuing marijuana smoking in the first trimester. This is consistent with the antiemetic effect of the cannabinoids (51). We have interviewed women in the Motherisk Program who reported that one line of hash had a profound effect on their nausea and vomiting of pregnancy, better than that of prescription drugs. Wu et al. reported a nearly fivefold greater increment in the blood carboxyhemoglobin level and a threefold increase of inhaled tar with smoking marijuana when compared with regular cigarettes (65). Pregnant women with high serum levels of cannabinoids, phencyclidine and cocaine, exhibited significantly lower concentrations of folate and feritin than subjects with lower serum drug concentrations. High maternal serum concentrations of illicit drugs were associated with significant increase in leukocyte counts. In a study conducted by Shiono et al., to evaluate prospectively the effects of cocaine and marijuana use on pregnancy outcome, marijuana use was relatively common and was not associated with adverse pregnancy outcome (52). Similarly, in another prospective study of the effects of marijuana use in pregnancy, the rates of adverse effects were very infrequent, although infants born to cannabinoid users exhibited significantly more meconium staining (57 vs. 25% in nonusers). Significant differences in the duration of labor were also observed (18). Users of marijuana experienced slightly elevated rates of dysfunctional labor (43% vs. 35% in nonusers), precipitate labor (13% vs. 8%) and meconium staining (17% vs. 13%). Although these were not felt to be of clinical significance. (19).

REFERENCES 1. Berkowitz K, LaSala A. Risk factors associated with the increasing prevalence of pneumonia during pregnancy. Am J Obstet Gynecol 1990; 163(3):981–985.

Effects of Prenatal Drug Exposure

343

2. Bongain A, Huss M, Gillet JY. Drug addiction and pregnancy. Rev Pract 1992; 1542(8):1004– 1009. 3. Burkett G, Ysin SY, Palow D, et al. Patterns of cocaine binging: effect on pregnancy. Am J Obstet Gynecol 1994; 171(2):372–378. 4. Campbell D, Parr MJ, Shutt LE. Unrecognized (‘‘crack’’) cocaine abuse in pregnancy. Br J Anesth 1996; 77(4):553–555. 5. Chan L, Pham H, Reece EA. Pneumothorax in pregnancy associated with cocaine use. Am J Perinatol 1997; 14(7):385–388. 6. Chesnoff IJ, Burns WJ, Schnoll SH, et al. Cocaine use in pregnancy. N Engl J Med 1985; 313:666–669. 7. Claman AD, Strang RI. Am J Obstet Gynecol 1962; 83:252. 8. Delaney DB, Larrabee KD, Monga M. Preterm premature rupture of the membranes associated with recent cocaine use. Am J Perinatol 1997; 14(5):285–288. 9. Dinsmoor MJ, Irnos SJ, Christmas JT. Preterm rupture of the membranes associated with recent cocaine use. Am J Obstet Gynecol 1994; 171(2):305–308. 10. Elliot RH, Rees GB. Amphetamine ingestion presenting as eclampsia. Can J Anesth 1990; 37(1):130–133. 11. Ellwood DA, Sutherland P, Kent C, O’Connor M. Maternal narcotic addiction: pregnancy outcome in patients managed by a specialized drug-dependency antenatal clinic. Aust NZ J Obstet Gynecol 1987; 27(2):92–98. 12. Eriksson M, Larsson G, Zetterstrom R. Amphetamine addiction and pregnancy: II. Pregnancy, delivery and the neonatal period—socio-medical aspects. Acta Obstet Gynecol Scand 1981; 60(3):253–259. 13. Eriksson M, Larsson G, Winbladh B, Zetterstorm R. The influence of amphetamine addiction on pregnancy and the newborn infant. Acta Paediatr Scand 1978; 67:95–99. 14. Fletcher JC, Evans MI. Maternal bonding in early fetal ultrasound examinations. N Engl J Med 1983; 308:392–393. 15. Fox CH. Cocaine use in pregnancy. J Am Board Fam Pract 1994; 7(3):225–258. 16. Funkhouser AW, Butz AM, Feng TI, et al. Prenatal care and drug use in pregnant women. Drug Alcohol Depend 1993; 33(1):1–9. 17. Ganapathy V, Leibach FH. Human placentae: a direct target for cocaine action. Placenta 1994; 15(8):785–795. 18. Greenland S, Staisch KJ, Brown N, Gross SJ. The effects of marijuana use during pregnancy: I. A preliminary epidemiological study. Am J Obstet Gynecol 1982; 143(3):408–413. 19. Greenland S, Richwald GA, Honda GD. The effects of marijuana use during pregnancy: II. A study in a low-risk home-delivery population. Drug Alcohol Depend 1983; 11(3–4):359– 366. 20. Hawthorne JL, Maier RC. Drug abuse in an obstetric population of a midsized city. South Med J 1993; 86(12):1334–1338. 21. Hayden JW, Comstock EG, Comstock BS. The clinical toxicology of solvent abuse. Clin Toxicol 1976; 9:169–184. 22. Hedstorm SA, Monga M, Bishop K, Blanco JD. The effect of cocaine on mitogen-included lymphyocyte proliferation in pregnant women. Am J Perinatol 1997; 14(10):583–586. 23. Jaffe JH. Drug addiction and drug abuse. In: Gilman AG, Goodman LS, Rall TW, Murad F, eds. The Pharmacological Basis of Therapeutics, 7th ed. New York: Macmillian, 1985, pp 532–581. 24. Jawaher D, Leo PJ, Anandarao N, Pachter BR. Cocaine-associated intestinal gangrene in a pregnant woman. Am J Emerg Med 1997; 15(5):510–512. 25. Joffe GM, Kasnic T. Medical prescription of dextroamphetamine during pregnancy. J Perinatol 1994; 14(4):301–303. 26. Kaminiski M, Rumeau-Rouquette C, Schwartz D. Alcohol consumption in pregnant women and the outcome of pregnancy. Alcohol Clin Exp Res 1978; 2:155–163.

344

Bishai and Koren

27. King MD, Day RE, Oliver JS, et al. Solvent encephalopathy. BMJ 1981; 283:663–665. 28. Kistin N, Handler A, Davis F, Ferre C. Cocaine and cigarettes: a comparison of risks. Paediatr Perinat Epidemiol 1996; 10(3):269–278. 29. Knight EM, James H, Edwards CH, et al. Relationships of serum illicit drug concentrations during pregnancy to maternal nutritional status. J Nutr 1994; 124(6 suppl):973S–980S. 30. Lampley EC, Williams S, Myers SA. Cocaine-associated rhabdomyolysis causing renal failure in pregnancy. Obstet Gynecol 1996; 87(5 pt 2):804–806. 31. Lam SK, To WK, Duthie SJ, Ma HK. Narcotic addiction in pregnancy with adverse maternal and perinatal outcome. Aust NZ J Obstet Gynaecol 1992; 32(3):216–221. 32. Little BB, Snell LM, Klien VR, et al. Maternal and fetal effects of heroin addiction during pregnancy. J Reprod Med 1990; 35(2):159–162. 33. Lutiger B, Graham K, Einarson TR, Koren G. Relationship between gestational cocaine use and pregnancy outcome: a meta-analysis. Teratology 1991; 44(4):405–414. 34. Madden JD, Payne TF, Miller S. Maternal cocaine abuse and effect on the new born. Pediatrics 1986; 77:209–211. 35. Makarechian N, Agro K, Devlin J, et al. Association between moderate alcohol consumption during pregnancy and spontaneous abortion, stillbirth and premature birth: a meta-analysis. Can J Clin Pharmacol 1998; 5(3):169–176. 36. Marbury MC, Linn S, Monson R, et al. The association of alcohol consumption with outcome of pregnancy. Am J Public Health 1983, 73:1165–1168. 37. Mauri A, Piccione E, Deiana P, Volpe A. Obstetric and perinatal outcome in human immunodeficiency virus-infected pregnant women with and without opiate addiction. Eur J Obstet Gynecol Reprod Biol 1995; 58(2):135–140. 38. Miller JM Jr, Boudreaux MC, Regan FA. A case-control study of cocaine use in pregnancy. Am J Obstet Gynecol 1995; 172(1 pt 1):180–185. 39. Mishra A, Landzberg BR, Parente JT. Uterine rupture in association with alkaloidal (‘‘crack’’) cocaine abuse. Am J Obstet Gynecol 1995; 173(1):243–244. 40. Moen MD, Caliendo MJ, Marshall W, Uhler ML, Hepatic rupture in pregnancy associated with cocaine use. Obstet Gynecol 1993; 82(4 pt 2, suppl):687–89. 41. Murphy EH, Chon J, Darvish-Sefat F, et al. Effects of cocaine-induced seizures during pregnancy in the rabbit. Physiol Behav 1997; 62(3):597–604. 42. Navarro C, Botel F, Figueras J, et al. Perinatal aspects of the children of heroin addicts. (Spanish). An Espanoles Pediatria 1987; 2694:251–254. 43. Oro AS, Dixon SD. Perinatal cocaine and methamphetamine exposure: maternal and neonatal correlates. J Pediatr 1987; 111:571–578. 44. Pastorek JG II, Plauche WC, Faro S. Acute bacterial endocarditis in pregnancy: a report of three cases. J Reprod Med 1983; 28(9):611–614. 45. Polygenis D, Wharton S, Malmberg C, et al. Moderate alcohol consumption during pregnancy and the incidence of fetal malformations: a meta-analyses. Neurotoxicol Teratol 1998; 20:61– 68. 46. Ramamoorthy JD, Ramamoorthy S, Leibach FH, Ganapathy V. Human placental monoamine transporters as targets for amphetamines. Am J Obstet Gynecol 1995; 173(6):1782–1787. 47. Rementeria JL, Janakammal S, Hollander M. Multiple births in drug-addicted women. Am J Obstet Gynecol 1975; 122(8):958–960. 48. Rodrigeuz EM, Mofenson LM, Chang BH, et al. Association of maternal drug use during pregnancy with maternal HIV culture positive and perinatal HIV transmission. AIDS 1996; 10(3):273–282. 49. Samuels IS, Maze A, Albright G. Cardiac arrest during cesarean section in a chronic amphetamine abuser. Anesth Analg 1979; 58:528–530. 50. Schafer A, Eck M, Bell U, et al. Use of methadone in obstetric and gynecologic management of drug-dependent females with and without HIV infection. Geburtshilfe Frauenheilk 1991; 51(8):595–601.

Effects of Prenatal Drug Exposure

345

51. Schneiderman J. Nonmedical drug and chemical use in pregnancy. In: Koren G. ed. MaternalFetal Toxicology: A Clinician’s Guide, 2nd ed. New York: Marcel Dekker, 1994. 52. Shiono PH, Klebanoff MA, Nugent RP, et al. The impact of cocaine and marijuana use on low birth weight and preterm birth: a multicenter study. Am J Obstet Gynecol 1995; 172(1 pt 1):19–27. 53. Silver H, Wapner R, Loriz-Vega M, Finnegan LP. Addiction in pregnancy: high risk intrapartum management and outcome. J Perinatol 1987; 7(3):178–184. 54. Singer L, Arendt R, Song LY. Direct and indirect interactions of cocaine with childbirth outcomes. Arch Pediatr Adolesc Med 1994; 148(9):950–964. 55. Smith YR, Dombrowski MP, Leach KC, Hurd WW. Decrease in myometrial beta-adrenergic receptors with prenatal cocaine use. Obstet Gynecol 1995; 85(3):357–360. 56. Smith DS, Gutsche BB. Amphetamine abuse and obstetrical anesthesia. Anesth Analg 1980; 59:710–711. 57. Stone ML, Salerno LJ, Green M, Zelson C. Narcotic addiction in pregnancy. Am J Obstet Gynecol 1971; 109(5):716–723. 58. Swisher SG, Hunt KK, Schmit PJ, et al. Management of pancreatitis complicating pregnancy. Am Surg 1994; 60(10):759–762. 59. Sokol RJ, Miller SI, Reed G. Alcohol abuse during pregnancy: an epidemiologic study. Alcohol Clin. Exp Res 1980; 4:135–145. 60. Towers CV, Pircon RA, Nageotte MP, et al. Cocaine intoxication presenting as preeclampsia and eclampsia. Obstet Gynecol 1993; 81(4):545–547. 61. Volpe A, Correnti E, Grasso A, et al. Drug addiction during pregnancy. Biol Res Pregnancy Perinatol 1983; 493:137–138. 62. Von Almen WF II, Miller JM Jr. Ts and Blues in pregnancy. J Reprod Med 1986; 31(4):236– 239. 63. Wilkins-Haug L, Gabow PA. Toluene abuse during pregnancy: obstetric complications and perinatal outcomes. Obstet Gynecol 1991; 77(4):504–509. 64. Woods JR. Maternal and transplacental effects of cocaine. Ann NY Acad Sci 1998; 846:1– 11. 65. Wu TC, Tashkin DP, Djahed B, et al. Pulmonary hazards of smoking marijuana as compared with tobacco. N Engl J Med 1988; 318:347–351.

19 Neonatal Drug Withdrawal Syndromes James B. Besunder Metro Health Medical Center and Case Western Reserve University School of Medicine, Cleveland, Ohio

Jeffrey L. Blumer Rainbow Babies and Children’s Hospital and Case Western Reserve University School of Medicine, Cleveland, Ohio

Clinical Case You are consulted about a baby who is suspected of having an opioid withdrawal syndrome: restlessness, intractable crying, diarrhea, vomiting, sweating, tremor, and poor feeding. The mother claims that although she is a heroin user, she has not been able to get any drug for several days. Neonatal urine test is negative, and the house staff suspects that the neonatal presentation is something other than drug withdrawal.

INTRODUCTION Drug abuse is a global problem within our society. As a result, the fetus, by passive addiction, has become an unfortunate victim of increasing maternal substance abuse. Although most pediatricians are acquainted with the neonatal narcotic withdrawal syndrome, substance withdrawal syndromes have now been described for many other classes of drugs, including sedative-hypnotics, stimulants, antidepressants, neuroleptics, antihistamines, and alcohol (1,2). The pediatrician, in particular, must be cognizant of the potential that signs or symptoms observed in a young infant may be the consequence of drug withdrawal. This chapter reviews drugs known to precipitate withdrawal syndromes in newborns, their clinical manifestations and therapy, and, when known, the mechanism for induction of the substance withdrawal syndrome.

OPIOIDS Although addiction to heroin and methadone is encountered most frequently, the narcotic withdrawal syndrome has also been described with propoxyphene (3–5), codeine (6,7), and the opiate agonist/antagonist pentazocine (8–10). A great deal is known about the 347

348

Besunder and Bloomer

short-term clinical effects of maternal opiate addiction on the newborn; however, many confounding variables such as polydrug abuse, improper nutrition, and socioeconomic factors obscure a description of their potential long-term effects. Mechanism of Opiate Withdrawal Evidence exists that at least part of the clinical manifestation of narcotic withdrawal results from α2-adrenergic supersensitivity in the locus ceruleus (LC) (11). Increased firing of neurons located in the LC of monkeys produces many of the central nervous system symptoms associated with narcotic withdrawal, including behavioral changes, increased wakefulness, and tremors (12,13). α2-Adrenergic receptors (14) as well as opiate receptors (15– 17) are located in the LC, and stimulation of these receptors by clonidine and morphine, respectively, inhibits activation of the LC in primates and rats (18,19). Hamburg and Tallman (20) demonstrated an increased number of α2-adrenergic receptors in rat brains following chronic morphine administration. In 1978 Aghajanian (21) observed similar depressant effects on LC neuronal firing by morphine and clonidine in rats, although the two drugs appeared to act at independent receptors within the LC. Initially, the administration of morphine markedly depressed LC-neuronal activation. However, tolerance developed rapidly, with normal firing rates observed after 4–5 days of daily morphine administration. Using a microiontophore technique, a withdrawal response (⬎ 100% activation of LC-neuronal firing) in tolerant rats was then observed with the direct application of the competitive opioid antagonist naloxone. When clonidine was applied to the LC neurons, inhibition of firing comparable to that observed with morphine alone was demonstrated. Finally, Aghajanian in the same set of experiments demonstrated that piperoxane, an α-adrenergic receptor antagonist, blocked clonidine, but not morphine-induced LC neuronal activation. Thus, these data suggest that stimulation of opiate and α2-adrenergic receptors inhibits in parallel, but not independently, activation of neurons in the LC. Withdrawal of opiates, possibly resulting from or potentiated by an increased number of unbound α2-adrenergic receptor sites, causes activation of neurons in the LC, leading to manifestations of the narcotic withdrawal syndrome. The effect of naloxone alone on LC neurons was not determined, however, and as a result, a direct effect of naloxone rather than induction of a narcotic withdrawal response cannot be excluded as a possible reason for the heightened activation of LC neurons following naloxone administration. These observations of Aghajanian (21) also offer a cellular basis for the efficacy of clonidine in suppressing some of the symptoms of opiate withdrawal. A spinal cord effect of opioid withdrawal may account for other symptoms of the withdrawal syndrome, such as hyperreflexia and hyperalgesia. Substance P (SP) is thought to be the neurotransmitter responsible for nocioceptive input into the spinal cord (22,23), and its release from the spinal cord is inhibited by opioids in rats and cats (24,25). Vacca et al. (26) observed an increase in SP concentrations in the dorsal horn of rat spinal cords following chronic morphine administration. Bergstrom et al. (27) reported similar findings in rats and suggested that release of accumulated SP or supersensitivity of postsynaptic SP receptors following opioid withdrawal is responsible for the hyperreflexia seen during the narcotic withdrawal syndrome. Bell and Jaffe (28), in a study evaluating dorsal and ventral root depolarization responses in neonatal rat spinal cords, provide evidence supporting a presynaptic mechanism of morphine withdrawal (i.e., presynaptic accumulation of SP and then release) rather than a postsynaptic mechanism. Of interest, Gintzler and

Neonatal Withdrawal Syndromes

349

Scalisi (29) demonstrated that morphine inhibits and SP induces non-cholinergic-mediated contraction of isolated guinea pig ileum. Further studies in the isolated guinea pig ileum strongly suggested that naloxone reversed the effects of morphine by releasing SP (29). Previously, Schulz and Herz (30) had shown that naloxone-precipitated enteric withdrawal from opiates is mediated, at least in part, by excitation of post-ganglionic cholinergic neurons with release of acetylcholine and activation of smooth muscle muscarinic receptors. Although these preliminary data on the neurohumoral effects of narcotic withdrawal are encouraging, further research in primates is needed to elucidate the mechanism of opiate withdrawal in human infants. Clinical Manifestations of Opiate Withdrawal Characteristics at Birth The incidence of premature birth among heroin-addicted women ranges from 17 to 30% (31–34). Zelson et al. (31) reported more than a twofold increase in premature deliveries over a 10-year period in 384 infants born to heroin-addicted mothers compared with the overall premature birthrate at their institution over the same time period. In contrast to heroin-addicted mothers, women enrolled in methadone treatment programs appear to be at only slightly increased risk of premature delivery. Newman et al. (35) reported a 7% incidence of deliveries prior to 8 months gestation in a large series of methadone-maintained women. Of interest, Olofsson et al. (32) observed younger mean gestational ages in infants whose mothers had been acutely withdrawn from methadone within a month of delivery compared with mothers who were maintained on methadone until the time of delivery. Low birth weight (⬍ 2500 g) is a common sequela among newborns of heroin addicts as well as the infants of women who use both heroin and methadone and women who enter methadone treatment programs during their third trimester (36–39). Naeye et al. (36) reviewed postmortem data in infants born to heroin addicts and demonstrated that growth retardation had been due mainly to a decreased number of cells in various organs, including the heart, pancreas, adrenal glands, spleen, thymus, and kidneys. Brain specimens were not available. Not all cases of growth retardation could be attributed to poor nutrition during pregnancy. Kandall et al. (37) reported mean birth weights of 2490 and 2535 g (mean gestational age of 38 weeks for both groups) among infants born to heroinand combined heroin- and methadone-addicted mothers, respectively, compared with a mean birth weight of 3176 g in the control group. Finnegan (38) identified a strong relationship between prenatal care and birth weight among infants born to heroin addicts. Birth weight under 2500 g was observed in 47.6% of infants born to drug-dependent women with no prenatal care compared with an 18.8% incidence for women with good prenatal care. By comparison, Finnegan reported a 20% incidence of low birth weight in a control group of nonaddicts who received no prenatal care. In contrast to the infants mentioned above, several investigators have observed higher birth weights among infants born to women addicted to methadone than to infants of heroin addicts (32,37,39,40). Kandall et al. (37) reported a mean birth weight of 2961 g (mean gestational age of 39.4 weeks) for 108 infants born to women who used methadone as their sole or major nacrotic during pregnancy compared with a mean birth weight of 2490 g (mean gestational age 38 weeks) for 61 infants born to heroin addicts. The effect of methadone on birth weight was independent of maternal age, race, or prenatal care. The same investigators also observed a direct relationship between methadone dose

350

Besunder and Bloomer

during the first trimester and birth weight in a subgroup of infants born to mothers registered in a methadone maintenance program where complete records of drug dosages and urinalyses were available. The investigators suggest that methadone may promote fetal growth during the first trimester in a dose-dependent fashion. Although methadoneaddicted women have larger babies than women addicted to heroin, both groups have an increased risk of delivering infants that are small for gestational age (SGA). Zelson et al. (39) reported incidence of SGA infants of 35 and 22% among babies born to heroin- and methadone-addicted women, respectively. Klenka (34) confirmed the high incidence of SGA infants born to heroin-addicted mothers when he described this finding in eight of 32 infants. On the other hand, Ostrea and Chavez (41) reported a lower incidence of 16.5% in 830 infants born to drug-dependent women. Head circumference and birth length are also significantly reduced among infants born to narcotic-addicted mothers compared with controls (42,43), although the incidence of head circumference below the 10th percentile did not differ between the two groups (42). A high incidence of low Apgar scores in infants born to narcotic addicts has also been reported by Olofsson et al. (32) and Ostrea and Chavez (42). Both groups of investigators found a 20% incidence of Apgar scores less than 7 at 1 minute, double the incidence observed by Ostrea and Chavez in their control infants. Ostrea and Chavez (41) also reported more than a two-fold greater incidence of low 5-minute Apgar scores (ⱕ6) in infants born to drug-dependent women compared with controls (7 vs. 3.2%). Infants born to heroin addicts appear to be less at risk of developing severe hyperbilirubinemia. In a prospective study, Nathenson et al. (44) reported significantly higher mean total bilirubin concentrations in control infants compared with infants born to heroinaddicted women during each of the first 3 days of life. Upon excluding factors predisposing infants to hyperbilirubinemia, such as hemolytic disease, the elevated bilirubin concentrations were found to be due mainly to an increase in the indirect fraction. The lower bilirubin concentrations observed in the heroin-addicted group most likely resulted from increased activity of bilirubin glucuronyl transferase. This effect of heroin has been demonstrated in animals (43). Marked hypertrophy of the smooth endoplasmic reticulum has been described in liver biopsies from heroin-addicted adults (45). Zelson et al. (31) reported the occurrence of jaundice in only 24 of 384 infants born to heroin-addicted mothers. Of these 24 infants, 6 had an ABO incompatibility and 1 was septic. In contrast, limited data are available on the incidence of hyperbilirubinemia among infants born to women maintained on methadone during pregnancy. Harper et al. (46) reported ‘‘clinical jaundice’’ in approximately half of their group of infants born to women in a methadone treatment program. However, only five of these infants had serum bilirubin concentrations exceeding 10 mg/dL, and all these babies had mild ABO incompatibilities. More data are required to ascertain the effect of methadone on serum bilirubin concentrations in newborn infants. Whether maternal narcotic addiction is associated with congenital defects is unknown. Zelson et al. (31) reported congenital abnormalities in only 4 of 384 infants born to heroin-addicted mothers. In contrast, the data of Ostrea and Chavez (41) reveal a statistically significant increased incidence of congenital malformations in their cohort of 830 newborns. Among the 37 infants with congenital defects, 20 (2.4%) had major malformations, including hydrocephalus, heart disease, and genitourinary tract abnormalities. Only 0.5% of their control population had major malformations. Harper et al. (46) reported

Neonatal Withdrawal Syndromes

351

congenital defects in 3 of 51 infants born to women maintained on methadone; however, only one would be considered a major malformation. Hyaline membrane disease (HMD) appears to occur less frequently among infants born to heroin addicts. Glass et al. (47) reported no cases of HMD in 33 premature infants born to heroin addicts compared with 26 cases in 123 premature infants born to nonaddicted mothers. The findings of this study are supported by the data generated by Gluck and Kulovich (48) and Taeusch et al. (49). Gluck and Kulovich reported mature lecithin/ sphingomyelin rations in fetuses of narcotic addicts at an earlier gestational age than would be expected in fetuses of nonaddicted women, whereas Taeusch and coworkers demonstrated accelerated lung maturation in fetal rabbits injected with heroin compared with fetal rabbits injected with saline. However, Ostrea and Chavez (41) observed no difference in the incidence of HMD between narcotic-addicted and control infants. In summary, in utero narcotic exposure may have profound effects on the developing fetus, effects that may increase perinatal morbidity and mortality. The birth characteristics of infants born to narcotic-addicted women are summarized in Table 1. Signs and Symptoms of Withdrawal Overt symptoms of narcotic withdrawal occur in 67–90% of infants born to narcoticaddicted women whether the additive drug is heroin or methadone (31–33,35,39,40, 46,50). Although the incidence of individual symptoms is similar between the heroin and methadone groups, infants born to methadone-addicted mothers appear to experience more symptoms (39) and the withdrawal syndrome appears to be more severe (39,40). Rahbar (40) reported moderate and severe symptoms associated with withdrawal in 50 and 18% of symptomatic infants, respectively, born to methadone-addicted women compared with 14 and 7% of infants born to heroin-addicted mothers. The severity (50,51) and possibly the duration of the withdrawal syndrome (32) have been shown to correlate directly with the amount of methadone but not heroin taken

Table 1 Birth Characteristics of Infants Born to NarcoticAddicted Women Characteristic Birth weight Premature infants (frequency) Small for gestational age Birth length Head circumference at birth 1-minute Apgar score ⬍7 (frequency) Hyperbilirubinemia (frequency) Congenital malformations a

Addictiona H⬍M⬍N H⬎M⬎N H⬎M⬎N H⫽M⬍N H⫽M⬍N H, M ⬎ N b H ⬍ Nc ⫾ H, M ⬎ N

H, Heroin; M, methadone; N, nonaddicted. Data comparing Apgar scores in infants born to heroin- and methadoneaddicted mothers are not available. c No data available evaluating the incidence of hyperbilirubinemia among infants born to methadone-addicted mothers. Source: Modified from Ref. 53. b

352

Besunder and Bloomer

during pregnancy. Ostrea and colleagues (51) observed a milder course among infants whose mothers took less than 20 mg/day of methadone during pregnancy. Olofsson et al. (32), in a similar study, reported withdrawal symptoms lasting longer than 5 days in 20 of 37 infants whose mothers were taking 20 mg or more per day of methadone at the time of delivery compared with 10 of 41 infants whose mothers took less than 20 mg/day. The onset of narcotic withdrawal symptoms varies tremendously from minutes after delivery to 1–2 weeks of age. However, the majority of infants will develop withdrawal symptoms by 48 hours of life (33,39,40,46). Methadone withdrawal, owing to its longer elimination half-life (t1/2β ) of approximately 35 hours (52), may be delayed compared with heroin but usually manifests itself within 48 hours (46). Harper et al. (46) reported the onset of withdrawal symptoms within 24 hours of birth in 42%, 48 hours in 73%, and 72 hours in 87% of symptomatic infants born to mothers in a methadone treatment program. Late-onset symptoms of withdrawal developed between 7 and 14 days of life in only 2% of these infants. The timing of the last maternal dose of narcotic with respect to birth also appears to influence the time to onset of symptoms (38,53). The neonatal narcotic withdrawal syndrome is characterized by signs and symptoms of central nervous system excitation, altered gastrointestinal function, respiratory distress, and vague autonomic symptoms. The signs and symptoms of withdrawal have been represented by the mnemonic withdrawal (1): W—wakefulness I —irritability T —tremulousness, temperature variation, tachypnea H —hyperactivity, high-pitched cry, hyperacusia, hyperreflexia, hypertonia D —diarrhea, diaphoresis, disorganized suck R —rub marks (excoriations of knees and face), respiratory distress, rhinorrhea A —apneic spells, autonomic dysfunction W—weight loss or failure to gain weight A —alkalosis (respiratory) L —lacrimation Other symptoms include hiccups, vomiting, stuffy nose, sneezing, yawning, photophobia, twitching, myoclonic jerks, opisthotonos, and seizures (1,53). Irritability is the most common manifestation of narcotic withdrawal, appearing in 45 to as many as 100% of symptomatic infants (31,33,39,50). Other symptoms that have been reported in more than half of symptomatic infants include tremors and hypertonicity (31,33,39,50). Vomiting occurs in one-quarter to one-third of infants. The frequency, if reported, of withdrawal manifestations is summarized in Table 2. Signs and symptoms of central nervous system hyperexcitability, such as irritability, restlessness, and tremors, appear early in the clinical course. Primitive reflexes, such as the Moro reflex, are exaggerated, as are deep tendon reflexes. Fist sucking is common and is reported in up to 80% of narcotic-addicted infants (51). Poor oral intake, regurgitation of feeds, vomiting, and diarrhea are frequently observed upon initiation of oral feedings (33). Feeding difficulties are most likely the result of an uncoordinated and ineffective sucking reflex (54,55). Excessive weight loss or the failure to gain weight, which is commonly observed in narcotic-addicted infants, is due to both poor caloric intake and to an increased tissue oxygen consumption (56). Abnormalities in ventilation also exist in infants whose mothers abuse narcotics. Glass et al. (57) reported tachypnea with a concomitant primary respiratory alkalosis in

Neonatal Withdrawal Syndromes

353

Table 2 Incidence of Withdrawal Manifestations in Symptomatic Infants Born to Narcotic-Addicted Womena Signs and symptoms

Percent with manifestation

Ref.

Irritability Tremors Hypertonicity Vomiting High-pitched cry Respiratory distressb Diarrhea Fever Convulsions Diaphoresis Yawning Sneezing Hiccups

45–100 65–86 20–86 8–37 17–45 9–30 7–22 6–16 3–21 6 3 ⬍1–65 ⬍1

31,33,39,50,51 31,33,39,50,51 31,39,50,51 31,39,50,51 31,39,50,51 31,33,39,50 31,50,51 31,33,39,50 31,32,39,50,61 31 31 31,51 31

a

Incidence is reported as percentage of infants with clinical manifestations, not as percentage of all infants born to narcotic-addicted women. b Respiratory distress due to etiologies other than hyaline membrane disease.

22 infants born to heroin addicts compared with 19 normal infants. Olsen and Lees (58) demonstrated a blunted ventilatory response to increasing concentrations of inspired carbon dioxide in 9 infants born to methadone addicts. This depressed response lasted for an average of 15 days but persisted for 31 days in one infant. Davidson et al. (59) reported abnormal sleeping ventilatory patterns in 27 infants born to substance-abusing mothers, many of whom abused opiates, compared with 43 control infants. These abnormalities included more frequent apneic episodes (ⱖ6 s), longer duration of apneic events, and more periodic breathing. The significance of these findings is presently unknown. Followup data have not disclosed a relationship in these infants between abnormal ventilatory patterns and future problems, such as the sudden infant death syndrome (SIDS) (60). Seizures have been reported in up to 21% of infants exhibiting narcotic withdrawal. The onset of seizure activity is usually delayed. Rosen and Pippenger (50) observed seizures in 5 of 18 symptomatic infants born to narcotic addicts. These infants developing seizures between the sixth and eighth days of life. Infants receiving paregoric for withdrawal symptoms had a significantly lower incidence of developing seizures than infants being treated with other drug combinations. These seizures were difficult to control with phenobarbital, diazepam, and paraldehyde alone, but responded to paregoric alone or in combination with diazepam or phenobarbital. Herzlinger et al. (61) reported the occurrence of seizures at a mean age of 10 days (range 3–34 days) in 18 narcotic-addicted infants. Seven infants experienced generalized seizures, seven myoclonic seizures, three had seizure activity manifested as automatisms (staring, blinking, lip smacking or abnormal sucking), and one infant exhibited multifocal clonic seizures. Paregoric was more effective than diazepam in controlling seizures. Interictal electroencephalograms (EEGs) were interpreted as normal in 12 of 13 infants in whom an EEG was obtained. Olofsson et al. (32), who found that seizures were statistically more frequent in infants with 1-minute Apgar scores below 7, inferred that birth asphyxia may predispose infants born to narcotic-addicted women to convulsions. In summary, seizures should be anticipated in infants born

354

Besunder and Bloomer

to narcotic-addicted mothers and should be treated with either paregoric or a parenteral narcotic agonist, such as morphine, until controlled. Maintenance therapy with paregoric appears to be effective in preventing the recurrence of seizures. The clinical course of the narcotic-addicted newborn can be extremely variable, ranging from only mild symptoms of brief duration to a crescendo in severity to intermittent symptomatology. A protracted clinical course with symptoms lasting 4–6 months has also been described (62). Wilson et al. (63) observed a biphasic course in 82% of symptomatic infants born to heroin-addicted mothers. These infants had an exacerbation or recurrence of symptoms upon discharge to the home environment. Symptoms included restlessness, agitation, tremors, wakefulness, hyperphagia, colic, and vomiting. These symptoms persisted for 3–6 months, and in some infants mandated resumption of therapy. Therefore, close outpatient follow-up to evaluate these infants for a recurrence of signs or symptoms of drug withdrawal is essential. Factors affecting the timing and severity of the withdrawal syndrome are summarized in Table 3. Although mortality from neonatal narcotic withdrawal approaches 0 with appropriate treatment, these infants may be at increased risk of dying from SIDS. Chavez et al. (64) compared the incidence of SIDS in 688 infants born to narcotic-addicted women with a control group of 388 randomly selected infants born to nonaddicted mothers of similar socioeconomic backgrounds. A fivefold increased incidence of SIDS (2.5 vs. 0.5%) was observed in the group of infants born to narcotic-addicted mothers. Of interest, more deaths were observed in infants who exhibited moderate to severe withdrawal compared with infants who manifested only minor or mild symptoms. The difference achieved significance at the p ⬍ 0.01 level. The mean age at the time of death was 9.2 weeks. Pierson et al. (65) also reported two confirmed and one possible case of SIDS among 14 infants whose mothers received methadone during pregnancy. Signs and symptoms of narcotic withdrawal resulting from maternal use of propoxyphene (3–5) codeine (6,7), and pentazocine (8–10) are similar and may be as severe as those described in heroin or methadone withdrawal. In these cases, the majority of infants manifested symptoms within the first 24 hours of life but were asympatomatic by 10 days of age. One infant developed seizures at 36 hours of life (4), and one infant died of SIDS at 14 weeks of age (10). Treatment Some infants exhibiting signs or symptoms of narcotic withdrawal can be effectively managed utilizing conservative measures such as holding, swaddling, minimal stimulation, and demand feedings using a hypercaloric formula (24 cal/oz). However, the majority of symptomatic infants require pharmacological therapy. Finnegan et al. (66) developed a neonatal narcotic abstinence scoring system in order to assess the severity of withdrawal

Table 3 Factors Influencing the Onset and Severity of the Neonatal Narcotic Withdrawal Syndrome Addictive drug used Drug dose during pregnancy and at time of delivery Timing of last dose prior to delivery Type and amount of analgesia/anesthesia given during labor

Neonatal Withdrawal Syndromes

355

and to guide therapy. They ranked 20 common symptoms of narcotic withdrawal, giving each symptom 1–5 points based on its clinical significance. The least significant symptoms, such as yawning, receive 1 point, whereas seizures receive 5 points (Table 4). This scoring system assumes that the greater the number of symptoms and the greater their severity, the higher the associated morbidity and mortality. According to Finnegan’s guidelines, babies are scored at birth and then hourly for 24 hours, every 2 hours during the second day of life, and every 4 hours thereafter. Infants with abstinence scores of 8 or more received pharmacotherapy, whereas those with scores less than 8 were treated conservatively. Finnegan and her coinvestigators then compared data from a group of 37 infants born to narcotic-addicted women prior to the development of the scoring system with a group of 37 infants born after the institution of the scoring system. The groups were similar with respect to severity of symptoms. Utilization of the scoring system not only increased the number of infants managed conservatively (46 vs. 30%), but also reduced the duration of pharmacological therapy from an average of 8–6 days. The mean length of hospitalization was also shortened from 21–15 days. Lipsitz (67) has developed a similar scoring system evaluating 11 signs and symptoms of narcotic withdrawal. This scoring system—while concentrating on the more common symptoms of withdrawal such as tremors, irritability, muscle tone, hyperreflexia, and tachypnea—does not emphasize several clinically important symptoms that may increase morbidity from narcotic withdrawal, such as vomiting, dehydration, seizures, and postprandial sleep. Although we feel the Lipsitz scoring system reliably identifies infants with a drug withdrawal syndrome, Finnegan’s scoring system is more detailed, has a greater emphasis on symptoms likely to correlate with morbidity and mortality, and therefore is better suited to quantify the severity of withdrawal and to guide pharmacological therapy. Also, to the best of our knowledge, the Lipsitz scoring system has not been rigorously evaluated as a means for identifying infants requiring pharmacotherapy or to guide ongoing therapy. Many drugs—including paregoric, tincture of opium, morphine, methadone, diazepam, chlorpromazine, phenobarbital, and clonidine—have been used to treat neonatal narcotic withdrawal (1,2,68). We discuss therapy with paregoric and phenobarbital in detail, since these drugs are the most popular agents used and represent the two important classes of compounds for treatment. Treatment with other drugs has been reviewed (1,2). Paregoric, a camphorated opium tincture, contains 0.4 mg/mL anhydrous morphine. Since the turn of the century it has been used successfully for the treatment of the neonatal narcotic withdrawal syndrome (69). The recommended dose for full-term infants is 0.2– 0.5 mL (0.08–0.20 mg anhydrous morphine) administered orally every 3–4 hours until symptoms are controlled (1). In general, we select our initial dose depending on the severity of withdrawal symptoms (Table 5). If no clinical improvement is observed within 4 hours after the initial dose, we increase the paregoric dose by 0.05 mL q4h until symptoms are controlled (Finnegan score ⱕ4). When the withdrawal score has remained stable for 48 hours the total daily dose may be tapered by 10% each day while maintaining the dosage interval constant. If symptoms recur, the dose should be increased to the previous dose which effectively controlled symptoms. Paregoric has the advantage of being a narcotic, so treatment should be more physiological than with nonnarcotic agents; and indeed, treated neonates have a more physiological sucking pattern, higher caloric intake, and more weight gain than infants treated with phenobarbital (55). Also, dosing of paregoric is easily titratable. Disadvantages of paregoric are mainly attributable to the other constituents present in the prepara-

356

Besunder and Bloomer

Table 4 Neonatal Narcotic Withdrawal Score Sign or symptom 1. High-pitched cry Intermittent Continuous 2. Postprandial sleep ⬍3 h ⬍2 h ⬍1 h 3. Moro reflex Hyperactive Markedly hyperactive 4. Tremors Mild when disturbed Marked when disturbed Mild when undisturbed Marked when undisturbed 5. Tone Increased Seizures 6. Frantic sucking of fists 7. Poor feeding 8. Vomiting Regurgitation Projectile 9. Stools Loose Watery 10. Dehydration 11. Frequent yawning 12. Sneezing 13. Nasal stuffiness 14. Sweating 15. Mottling 16. Fever 38–38°C ⬎38°C 17. Respiratory rate ⬎60 ⬎60 with retractions 18. Excoriation of nose 19. Excoriation of knees 20. Excoriation of toes Total maximum Source: From Ref. 66.

Possible score 2 3 1 2 3 2 3 1 2 3 4 2 5 1 2 2 3 2 3 2 1 1 1 1 1 1 2 1 2 1 1 1 40

Neonatal Withdrawal Syndromes

357

Table 5 Initial Paregoric Dose in Full-Term Infants Based on Withdrawal Score Withdrawal scorea ⬍ 8 8–10 11–13 14–16 ⱖ17 a b

Dose (mL) b None 0.2 0.3 0.4 0.5

Based on Finnegan Scoring System (66). See text for dosing interval and guidelines for chronic therapy.

tion. Camphor, a central nervous system (CNS) stimulant, is eliminated slowly owing to its high lipid solubility and dependence on glucoronidation for elimination. Paregoric also contains alcohol and anise oil, which may cause dependency (1). Benzoic acid, an oxidative metabolite of benzyl alcohol, is also present in paregoric: circumstantial, but not conclusive, evidence has linked benzyl alcohol, when administered in excessive doses, to the ‘‘gasping syndrome’’ described in premature infants (70–72). Our initial dosing recommendation for paregoric in premature infants is 0.05 mL/kg q4h with increments of 0.02 mL/kg q4h until symptoms are controlled. Although paregoric may be more physiological, phenobarbital and its analogs have also been employed successfully in the treatment of the neonatal narcotic withdrawal syndrome. Most physicians feel comfortable with using phenobarbital, which has the added advantage of mitigating symptoms referable to the central nervous system. Kundstadter et al., in 1958 (73), commented that when sedation is required, phenobarbital is a safer drug than an opiate. This belief is still held by many physicians today. However, phenobarbital will not relieve gastrointestinal symptoms, such as vomiting or diarrhea (69), and may cause significant CNS depression. Also, phenobarbital is not a suitable drug for dose titration owing to its prolonged elimination half-life. Other disadvantages of phenobarbital relate to other chemicals present in the formulation; the elixir contains alcohol, whereas the parenteral formulation contains propylene glycol, ethyl alcohol, and benzyl alcohol (1). Finally, there are no standard dosing guidelines for phenobarbital therapy, nor is the therapeutic serum concentration necessary to control withdrawal known. Two studies have compared the efficacy of treatment with paregoric and phenobarbital (74,75). Kandall et al. (74) prospectively randomized infants to receive paregoric or phenobarbital if their withdrawal score (modified Lipsitz scoring system) was 7 or greater or if any infant developed excessive severity of any sign. Serum phenobarbital concentrations were not monitored. The drugs were considered by the investigators to be comparable in their ability to control gastrointestinal, CNS, and autonomic symptoms. However, 7 of 62 infants receiving phenobarbital developed seizures, compared with none of 49 infants maintained on paregoric ( p ⬍ 0.025). Carin et al. (75) failed to demonstrate a difference in weight gain between infants treated with paregoric and those treated with phenobarbital. Paregoric-treated newborns received therapy for a longer period of time (median duration 22 days vs. 17 days) compared with phenobarbital-treated infants. This is most likely due to two factors: paregoric therapy was not discontinued until the daily dose had been weaned to 0.1 mL/kg, whereas once phenobarbital administration was stopped, effective

358

Besunder and Bloomer

serum concentrations were probably maintained for an extended period of time owing to the markedly prolonged t1/2β of phenobarbital. Carin’s study has several major flaws, which cast concerns on the validity of the data and conclusions. Among the mothers of infants treated with paregoric, 4 of 16 admitted to using heroin in addition to methadone, and 3 of 16 mothers used cocaine concomitantly. This is in contrast to the mothers whose infants were treated with phenobarbital, who had not used heroin or cocaine. Also, control of withdrawal symptoms and initial and subsequent withdrawal scores were not reported. Therefore, the severity of withdrawal was not controlled for in the treatment randomization process, nor was the efficacy of therapy in controlling withdrawal symptoms reported. In a more critical evaluation, Kaltenbach and Finnegan (76) demonstrated the superiority of paregoric in a study designed to determine whether there was a relation between pharmacotherapy and development outcome. Of 23 infants treated initially with paregoric, treatment was effective with this drug alone in 21. By comparison, only 17 of 36 infants were treated effectively with phenobarbital alone. The remaining 19 infants required the addition of paregoric to their phenobarbital regimen. Finnegan and Ehrlich (77) also reported that paregoric was more effective than phenobarbital or diazepam in controlling symptoms of narcotic withdrawal, whereas phenobarbital was more effective in controlling symptoms associated with nonopiate withdrawal. In summary, clinical manifestations occur in most infants of narcotic-addicted women. The Finnegan scoring system is a useful tool for assessing the severity of withdrawal and in guiding therapy. However, we would initiate pharmacological therapy with paregoric regardless of the infant’s withdrawal score if presenting symptoms included seizures, vomiting, or severe and persistent diarrhea resulting in weight loss or dehydration, or inability to sleep. Figure 1 outlines our treatment approach to infants of narcoticaddicted mothers. Outcome Interpreting follow-up data in infants born to narcotic-addicted mothers is very difficult owing to the paucity of studies, lack of adequate control populations, and the chaotic home environment into which these children are returned. Several studies have revealed marked behavioral abnormalities in narcotic-addicted infants (55,78–80). Strauss et al. (78) used the Brazelton Neonatal Behavioral Assessment Scale to evaluate 44 infants born to narcotic-addicted mothers. Testing was done during the first 48 hours of life. Addicted infants spend less time in an alert state and became less responsive to social and nonsocial stimuli over the first 2 days of life. The investigators suggested that this behavior may adversely affect development of infant-caregiver interaction patterns. This speculation is supported by a subsequent study by Kaltenbach et al. (79), who investigated the ability of the infants born to a narcotic-dependent mother to interact with their environment. These infants were evaluated on days 1 and 30 of life. All infants who would eventually require pharmacological therapy had a deficient interaction score on day 1. By day 30, only a subset of infants still receiving treatment demonstrated poor interaction skills. Interestingly, these were the infants whose mothers rarely visited them while hospitalized. The investigators theorized that poor interaction skills on day 1 may have alienated mothers, or conversely, maternal attention may have improved the infant’s interaction. Olofsson et al. (80) corroborated the data from these two studies. Of 72 infants born to narcotic addicts who were reinvestigated between 1 and 10 years

Neonatal Withdrawal Syndromes

359

Figure 1 Management approach for the infant born to a narcotic-addicted mother.

of life, 21% had moderate or severe psychomotor retardation primarily attributable to a deprivation syndrome. In addition, both Olofsson et al. (80) and Wilson et al. (63) observed hyperactive and aggressive behavior with a decreased attention span and lack of social inhibition in more than half of infants followed through infancy and early childhood. In contrast to the behavioral abnormalities described above, conflicting data regarding motor and mental development in infants of narcotic-addicted mothers have been reported. Rosen and Johnson (81) observed significantly lower mental and psychomotor developmental indices by the Bayley Scales of Infant Development at 12 and 18 months of age, though all scores were within normal limits. Lifschitz et al. (42) reported a higher incidence of low average and mildly retarded intellectual performance in children between 3 and 6 years of age who were born to narcotic-addicted mothers. Analysis of the data suggested that environmental and prenatal factors, not intrauterine exposure to narcotics, were responsible for the lower scores. By comparison, Wilson et al. (82) found no difference in mental or psychomotor developmental indices by the Bayley Scale at 9 months of age. Fine motor coordination was poorer, however, in infants born to narcotic addicts, supporting a similar observation by Rosen and Johnson (81). Finally, Kaltenbach and Finnegan found no evidence of impaired cognitive function in infants at 6 months of age (83), or in children between 3.5 and 4.5 years of age (84) who were born to narcoticaddicted women.

360

Besunder and Bloomer

Kaltenbach and Finnegan (76) reported no differences in mental development at 6 months of age between infants who demonstrated severe withdrawal symptoms at birth and those with mild symptoms not requiring pharmacological therapy. They suggest that appropriate therapeutic intervention may obviate adverse cognitive sequelae. However, prenatal or environmental factors were not analyzed. As stated previously, infants born to narcotic-addicted women have low measurements on mean birth weight, length, and head circumference. The growth curve in early infancy follows the percentiles noted at birth for these three parameters. However, after 4–6 months of age, some infants appear to have an accelerated growth pattern (62). Chasnoff et al. (62) followed the growth patterns of 15 infants whose mothers were maintained on a low dose of methadone (14 of 15 mothers received ⬍ 20 mg/day) during their third trimester of pregnancy. At birth and 4 months of age, the mean birth weight, length, and head circumference lay at the 10th, 10th, and 5th percentiles respectively. However, at 7.5 months, these infants lay at the 40th, 25–50th, and 25th percentiles, respectively. The excellent outcome for growth noted by Chasnoff et al. may be due to the low dose of methadone administered to the pregnant women. Olofsson et al. (80) demonstrated a correlation between maternal methadone dose and weight, height, and head circumference at a mean age of 3.5 years. Of mothers receiving less than 20 mg/day at birth, none of 33 infants were below the 10th percentile for weight, height, or head circumference, compared with 6 of 31 children whose mothers received 20 or more mg/day of methadone. Unfortunately, growth parameters at birth were not compared between the two groups. Overall, the heights of these children were statistically below the mean for age, a finding similar to those of Wilson et al. (85). While Lifschitz et al. (42) did not show impaired head growth, Rosen and Johnson (81) reported that head circumferences remained below the third percentile in 63% of infants followed up at 18 months of age whose head circumference at birth was below the third percentile. Wilson et al. (85) reported that 14% of heroin-exposed children had a head circumference below the third percentile when examined between 3 and 6 years of age. Again, the head circumference at birth was not reported. In summary, brain growth may be blunted long term subsequent to in utero narcotic exposure. Infants born to narcotic-addicted women may also be at increased risk of developing strabismus. Nelson et al. (86) and Rosen and Johnson (81) observed strabismus in 24 and 21% of narcotic-exposed infants, respectively, compared with a reported incidence of 2.8– 5.3% in the general population. Therefore, these infants should be evaluated for the presence of strabismus as part of their outpatient follow-up care. In summary, follow-up studies of narcotic-exposed infants have revealed significant behavioral abnormalities, but not long-term cognitive deficits.

STIMULANTS Adverse effects on the fetus and newborn attributable to maternal cocaine use during pregnancy are now being appreciated. It is not clear, however, whether infants of cocaineabusing mothers manifest symptoms from drug withdrawal. Madden et al. (87) reported no overt symptoms of withdrawal in eight infants whose mothers abused cocaine during pregnancy. Infants whose urine screens were positive for other abusive drugs were excluded from their study. Ryan et al. (88) recently compared three groups of infants; one group was born to mothers who used cocaine in addition to methadone and/or other drugs,

Neonatal Withdrawal Syndromes

361

another group of infants was born to mothers whose drug-use patterns were similar to those of the first group except they did not use cocaine, and the third group was a control population of non-drug-dependent women. The frequency of drug withdrawal symptoms was comparable between the cocaine/methadone- and methadone-exposed infants, implying that cocaine does not increase the risk of neonatal drug withdrawal. In contrast, other investigators have observed mild ‘‘withdrawal’’ syndromes consisting of abnormal sleep/wake patterns, poor feeding, irritability, tremulousness, hypertonia, and hyperreflexia (89–93). However, these symptoms are similar to those reported by Chasnoff et al. (94) in a 2-week-old cocaine-intoxicated infant. This child manifested symptoms for 48 to 72 hours. Cocaine should readily cross the placenta owing to its high lipid solubility. Since cocaine is primarily eliminated by hepatic metabolism and by plasma cholinesterases (95), which have low activity in the fetus and newborn (96), accumulation of cocaine in the fetus and subsequent toxic manifestations are not surprising. Owing to an expected slow elimination phase for cocaine by the fetus and newborn, we would anticipate withdrawal signs and symptoms to be delayed from birth if they occur at all. These pharmacokinetic/ pharmacodynamic predictions are supported by the data presented by Chasnoff et al. (97). These investigators reported the presence of cocaine or its metabolites in a newborn’s urine for at least 4 days after delivery. Therefore, we feel symptoms previously ascribed to ‘‘cocaine withdrawal’’ may in fact represent cocaine intoxication. Oro and Dixon (92) reported similar symptoms of CNS hyperexcitability among infants born to amphetamine-abusing mothers. However, some of these infants exhibited marked lethargy and required tube feedings when their hyperexcitable state subsided. This state of pronounced lethargy most likely represents a true withdrawal syndrome comparable to that observed during cocaine and methamphetamine withdrawal in adults (98). Other investigators have reported similar observations in infants born to mothers abusing amphetamines (99,100). Ramer (99), in 1974, reported the case of a full-term infant born to a mother using parenterally administered amphetamines during pregnancy. She was admitted to the maternity hospital 4 days prior to delivery. Episodes of diaphoresis, restlessness, and miotic pupils were observed in the newborn during the first day of life. By the third day of life, the infant was listless, with decreased muscle tone, and was observed to have a peculiar glassy-eyed stare. These symptoms abated by the ninth day of life. In summary, infants born to mothers abusing CNS stimulants may exhibit signs or symptoms of intoxication or withdrawal. As in the case of narcotic-addicted infants, the temporal relationship between the last maternal dose and delivery appears to be a major factor in the timing of neonatal symptoms and in determining whether the newborn manifests symptoms attributable to intoxication or withdrawal. Marked symptoms of CNS hyperexcitability can be controlled with the administration of benzodiazepines as needed (101), whereas withdrawal symptoms should be managed supportively. Although infants born to cocaine users may not manifest symptoms of drug withdrawal, cocaine clearly affects pregnancy and neonatal outcome adversely. Women who use cocaine during pregnancy are at increased risk for spontaneous abortion (88,89,102), premature labor (89), precipitous delivery (89), abruptio placentae (89,92,102), and meconium staining (89). Cocaine may also be teratogenic. Bingol et al. (103) reported a significantly higher congenital malformation rate in cocaine-using mothers compared with a non-drug-using control group. Affected infants presented with congenital heart disease or skeletal defects. Chasnoff et al. (104) observed genitourinary tract malformations in 3 of 23 infants born to women who only used cocaine during the first trimester of pregnancy,

362

Besunder and Bloomer

and in 6 of 52 infants whose mothers used cocaine throughout pregnancy. The genitourinary tract malformations included prune-belly syndrome (2), hydronephrosis (4), secondary hypospadias (2), and female pseudohermaphroditism (1). Two additional infants who were exposed to cocaine throughout pregnancy had ileal atresia. These clinical reports of the teratogenic potential of cocaine are supported by the work of Mahalik and coworkers (105), who described skeletal and genitourinary tract malformations in the offspring of gravid mice administered cocaine early in pregnancy. However, in none of these reports did a homogeneous pattern of malformations emerge, and the data should be interpreted cautiously. For example, during embryonic life the genitourinary system does not develop from one origin, and it is impossible to trace the malformations described above to one etiological factor. Although conflicting data initially appeared in the literature regarding the effect of maternal cocaine use on intrauterine growth (87,89,102), more recent investigations have observed a significant effect on growth (90,103,106). Bingol et al. (103) reported mean birth weight, length, and head circumference of 2276 g, 46.2 cm, and 32 cm in 50 fullterm infants born to women who abused only cocaine during pregnancy, whereas Fulroth et al. (90) discovered a 20% incidence of growth retardation and a 29% incidence of microcephaly in prenatally exposed infants. Most recently, Chasnoff et al. (104) reported lower mean birth weights, lengths, and head circumferences among term infants born to mothers who used cocaine throughout pregnancy compared with control infants. In contrast, no differences in intrauterine growth parameters existed between infants born to mothers who used cocaine only during the first trimester and control infants. Another adverse neurological outcome described in infants exposed in utero to cocaine is cerebral infarction. Chasnoff et al. (97) reported an infant with a right hemiparesis and right-sided focal seizures on the first day of life born to a mother who used excessive amounts of cocaine for 3 days prior to delivery. Computed tomography of the brain demonstrated a cerebral infarction within the distribution of the left middle cerebral artery, and a lumbar puncture revealed 1800 red blood cells and 5 white cells. The cerebrovascular accident may have resulted from intrauterine hypertension or hypotension due to the vasoconstrictor effects of cocaine on placental blood vessels. Furthermore, neurobehavioral effects similar to those observed in narcotic-exposed infants have been reported in infants born to cocaine users (90,91,102,104). Besides being more irritable, these infants demonstrate depressed interactive behavior and impaired responses to environmental stimuli. Chasnoff et al. (104), utilizing the Neonatal Behavioral Assessment Scale between 12 and 72 hours of life, observed significant deficiencies in orientation, motor ability, and state regulation among cocaine-exposed infants. In addition, Dixon et al. (91) observed very poor visual attention and abnormal flash-evoked visual potentials in 11 of 12 infants studied. Of concern are persistent visual disturbances at 4– 6 months of age, as revealed by preliminary data. Such neurobehavioral abnormalities may impede the establishment of appropriate caretaker-infant relationships. Abnormal (EEGs) featuring bursts of sharp wave-and-spike patterns and bursts of theta rhythm have also been seen in newborns of cocaine users (91,93). Doberczak et al. (93) reported an abnormal EEG during the first week of life in 17 of 38 infants studied. Although the EEG was still abnormal in 9 of 14 infants in whom the test was repeated between days 9 and 14 of life, all but one EEG had normalized by 3–12 months of age. These infants may also be at increased risk of SIDS (88,89). Chasnoff et al. (89) reviewed the histories of 66 cocaine-exposed infants born to women enrolled in perinatal drug programs in San Francisco or Chicago. Ten of these infants died of SIDS. This 15%

Neonatal Withdrawal Syndromes

363

incidence of SIDS is significantly greater than the risk previously reported for narcoticexposed infants and for the general population (64). In contrast, however, Bauchner et al. (106) did not detect an increased risk of SIDS among 175 infants exposed to cocaine in utero. In summary, cocaine intoxication should be anticipated in infants born to mothers abusing cocaine. Signs and symptoms of CNS hyperexcitability may persist for several days. Distinct neurobehavioral and EEG abnormalities have been attributed to in utero cocaine exposure. Long-term studies are required to determine whether these abnormalities will increase the infant’s risk for future developmental, cognitive, or behavioral sequelae.

SEDATIVE-HYPNOTICS Alcohol Jones et al. (107) first recognized malformations among eight infants born to alcoholic mothers and termed this pattern the fetal alcohol syndrome (FAS). Characteristic features of this syndrome include intrauterine growth retardation, slow postnatal growth, microcephaly, developmental delay, impaired intellectual performance, and craniofacial and musculoskeletal anomalies (107–109). Other reported abnormalities include cardiac defects (mostly septal defects), ear and external genitalia anomalies, and cutaneous anomalies such as hemangiomas, pigmented nevi, and hirsutism (108). Several of the functional and anatomical derangements observed in these infants most likely result from brain dysmorphogenesis. Jones and Smith (109) reported the results of an autopsy performed in one patient with FAS. Histopathology of the brain revealed small size, incomplete development of the cerebral cortex, agenesis of the corpus callosum, and disorganization of neural and glial elements. Although FAS is well recognized, symptoms of alcohol withdrawal in newborns have not been fully appreciated. Schaefer (110) published the first reported case of alcohol withdrawal in an infant born to an intoxicated American (Yukon) Indian woman. Symptoms appeared by 18–24 hours of life and included irritability, coarse tremors of the hands and feet, restlessness, sleepiness, and excessive crying. An ‘‘alcoholic fetor’’ was apparent up to 12 h of life. By the sixth day of life, the infant was still irritable and easily startled; however, other clinical manifestations had resolved. Since the report by Schaefer, other cases of alcohol withdrawal in neonates have been published (111,112). Central nervous system manifestations of alcohol withdrawal develop within 24 hour of birth and may include tremors, irritability, hypertonicity, muscle twitching, hyperventilation, hyperacusia, opisthotonos, and seizures (111,112). Gastrointestinal symptoms such as abdominal distention and vomiting are less frequent (111). Symptoms of alcohol withdrawal may be severe but appear to be of brief duration. Infants born to alcoholic mothers may be at a higher risk of seizures than narcotic-exposed infants; three of six newborns reported by Pierog et al. (111) and the neonate described by Nichols (112) developed seizures. Infants manifesting signs or symptoms of alcohol withdrawal require pharmacological intervention. The infants described above were treated with chlorpromazine or phenobarbital. Such interventions do not appear to be mechanistically based, and these drugs may not prove to be optimal therapy. Phenothiazines reduce the seizure threshold and thus may enhance the risk of seizures in these infants. Phenobarbital has been effective in adult cases of alcohol withdrawal, but it is not easily titrated. Our drug of choice is

364

Besunder and Bloomer

diazepam. Benzodiazepines demonstrate a cross-tolerance with alcohol and have a wider therapeutic index than phenothiazines or barbiturates (114). In adults, benzodiazepines are as effective as barbiturates in alleviating symptoms of alcohol withdrawal, and they can prevent withdrawal seizures (113). Several dosing regimens of diazepam have been used in adults (114). One such regimen, the diazepam-loading technique, appears to be both safe and effective (115). This technique relies on the slow clearance of diazepam and its active metabolites from the body. After effective loading with diazepam, the therapeutic effect usually persists longer than the duration of the withdrawal syndrome, obviating the need for additional pharmacological therapy. We would recommend administering 0.25 mg/kg IV of diazepam q2h until symptoms are controlled. Sellers et al. (115) reported successful responses to the diazepam-loading technique in a median time of 7.6 hour in 36 adults with acute alcohol withdrawal. Only three patients required therapy beyond 24 hours. In summary, alcohol withdrawal should be considered in the differential diagnosis of the irritable newborn, particularly when characteristic features of FAS are present. Signs and symptoms of neonatal alcohol withdrawal are remarkably similar to those observed in the neonatal narcotic withdrawal syndrome except that seizures may be more frequent in the former and gastrointestinal symptoms are more prevalent in the latter. Barbiturates Signs and symptoms of drug withdrawal have also been described in infants born to mothers receiving barbiturates (116–118). Desmond et al. (116) reported their observations on 15 infants born to mothers who either were addicted to barbiturates or were receiving phenobarbital for sedation, hypertension, or as part of an anticonvulsant regimen. Phenobarbital dosages ranged from 60 to 180 mg qd. Symptoms were similar to those observed in narcotic-exposed infants: all displayed hyperactivity, restlessness, disturbed sleep, and excessive crying, with the majority of infants also noted to be tremulous, hyperreflexic, and hyperphagic. Clinical manifestations of drug withdrawal appeared later in these infants compared with infants born to narcotic addicts. Most of the infants reported by Desmond et al. (116) became symptomatic toward the end of the first week of life, although the age at onset of barbiturate withdrawal was delayed up to 2 weeks in some infants. This delay should not be unexpected, considering the long t1/2β of barbiturates. Most infants were symptomatic for 2–6 weeks. Bleyer and Marshall (117) and Ostrea (118) also reported barbiturate withdrawal in infants born to mothers receiving secobarbital and butalbital, respectively. both these infants developed generalized seizures in addition to other symptoms of neuromuscular excitability. Infants demonstrating signs or symptoms of barbiturate withdrawal may be effectively treated with phenobarbital. Table 6 lists our indications for initiating therapy. We

Table 6 Indications for Initiating Pharmacological Therapy in Barbiturate Withdrawal Seizures Signs or symptoms interfering with normal caretaker–infant interactions, such as moderate to severe hyperactivity or irritability Poor feeding or failure to gain weight

Neonatal Withdrawal Syndromes

365

would administer a loading dose of 20 mg/kg IV or IM, and start maintenance therapy with 4 mg/kg/day IV or PO. Higher maintenance doses may be required to control symptoms (117). Serum phenobarbital concentrations should be monitored to avoid toxicity. When symptoms have been controlled for a week, we decrease the daily dose by 25% of the original dose every week. Benzodiazepines Symptoms indistinguishable from narcotic or barbiturate withdrawal, including seizures, have also been observed in infants born to mothers receiving therapeutic doses of chlordiazepoxide (119,120) and diazepam (121,122). The onset of symptoms may appear shortly after birth or be significantly delayed; Athinarayanan et al. (119) reported the onset of withdrawal at 21 days of age in twin infants born to a mother receiving 30 mg/day of chlordiazepoxide during pregnancy. Withdrawal symptoms persisted for long periods of time in several of these infants. We recommend initiating treatment of benzodiazepine withdrawal with 0.1 mg/kg diazepam q12h IV. More frequent dosing or larger doses may be required to control symptoms. When symptoms have been well controlled for approximately a week, we recommend weaning diazepam over 3–4 weeks by gradually reducing a given dose each day. Indications for initiating therapy are similar to those for barbiturate withdrawal (see Table 6). Case reports describing neonatal withdrawal from hydroxyzine (123), ethchlorvynol (124), and glutethimide (125) have also appeared in the literature. Although an abstinence syndrome from these agents would not be surprising, withdrawal from the concomitant use of phenobarbital, diazepam, and heroin, respectively, in these cases cannot be excluded. In summary, neonatal withdrawal from sedative hypnotics may occur and should be anticipated in infants born to mothers chronically receiving this class of drugs toward the end of pregnancy. Signs and symptoms of neuromuscular excitability predominate and may be delayed from birth in onset and may follow a protracted course. Treatment should be initiated when symptoms are life threatening or interfere with caretaker–infant interactions or when a potentially adverse effect on the growth of the infant is perceived.

MISCELLANEOUS DRUGS Other classes of drugs reported to be associated with a withdrawal syndrome in the newborn are tricyclic antidepressants (TCAs) (124–126) and phenothiazines (127). Webster (126) observed breathlessness, cyanosis, tachypnea, tachycardia, and irritability in a 1day-old infant born to a mother receiving desmethylimipramine during pregnancy. Gradual improvement in the infant’s condition was noted by 10 days of life. Eggermont (127) noted similar symptoms in three infants whose mothers took imipramine during pregnancy. Ben Musa and Smith (128) described an infant with hypothermia and jitteriness born to a mother receiving clomipramine. Symptoms were present by 12 h of life and persisted for 4 days. More data are required to permit us to appreciate the effect of maternally administered TCAs on the newborn and to decide whether these symptoms result from drug intoxication or drug withdrawal. O’Conner et al. (129), in 1981, reported an interesting case of late-onset neurological dysfunction in a 3-week-old infant born to a mother who had received intramuscular injec-

366

Besunder and Bloomer

tions of fluphenazine decanoate daily throughout her last two trimesters of pregnancy. The infant developed choreiform and dystonic movements, primarily of the upper extremities, associated with irritability, hypertonicity, and sleep disturbance. Diphenhydramine hydrochloride administration did little to alleviate the symptoms. The infant gradually improved and by 9 months of age was asymptomatic. This report illustrates a case of drug withdrawal following material administration of a high potency, long-acting phenothiazine. We feel that these symptoms may be analogous to a phenomenon known as withdrawal-emergent dyskinesias, a variant of tardive dyskinesias (130). This phenomenon most likely results from disuse supersensitivity of dopaminergic pathways in the brain. Treatment with phenothiazines may mask the dyskinesias. However, administration of phenothiazines or antimuscarinic agents may also worsen the underlying condition. Therefore, the aim of management should be supportive care with avoidance of neuroleptic and anticholinergic agents.

SUMMARY Neonatal drug withdrawal syndromes have been reported following chronic maternal administration of many classes of agents. Withdrawal from narcotics has been studied most intensely in newborns and is associated with both short- and long-term morbidity. Although infants born to mothers abusing stimulants may manifest symptoms of drug withdrawal, perinatal complications such as abruptio placentae and cerebrovascular accidents appear to be more worrisome. Neonatal withdrawal from sedative-hypnotics may also occur, with signs and symptoms of neuromuscular hyperexcitability predominating. Finally, the timing of the last maternal dose of a drug with respect to delivery as well as a given agent’s pharmacokinetic properties may influence the time to onset of a neonatal abstinence syndrome following delivery. Symptoms may be delayed from birth; they may be severe and life-threatening, and the clinical course may be protracted. Clinical Case Answer Drug withdrawal occurs when there is little or no drug in the system, therefore the negative neonatal urine test is consistent with the diagnosis. Although other diagnostic options must be carefully ruled out, the baby should benefit from oral paregoric acid treatment or other forms of sedation (e.g., phenobarbital). REFERENCES 1. Committee on Drugs. Neonatal drug withdrawal. Pediatrics 1983; 72:895–902. 2. Marx CM, Cloherty JP. Drug withdrawal. In: Manual of Neonatal Care, 2nd ed. Cloherty JP, Stark AR, eds. Boston: Little Brown, 1985, pp 17–28. 3. Tyson HK. Neonatal withdrawal symptoms associated with maternal abuse of propoxyphene hydrochloride. J Pediatr 1974; 85:684–685. 4. Klein RB, Blatman S, Little GA. Probable neonatal propoxyphene withdrawal: a case report. Pediatrics 1975; 55:882–884. 5. Quillan WW, Dunn CA. Neonatal drug withdrawal from propoxyphene. JAMA 1976; 235: 2128. 6. Mangurten HH. Neonatal codeine withdrawal in infants of nonaddicted mothers. Pediatrics 1980; 65:159–160.

Neonatal Withdrawal Syndromes

367

7. Van Leeuwen G, Guthrie R, Stange F. Narcotic withdrawal reaction in a newborn infant due to codeine. Pediatrics 1965; 36:635–636. 8. Scanlon JW. Pentazocine and neonatal withdrawal symptoms. J Pediatr 1974; 85:735–736. 9. Kopelman AE. Fetal addiction to pentazocine. Pediatrics 1975; 55:888–889. 10. Goetz RL, Bain RV. Neonatal withdrawal symptoms associated with maternal use of pentazocine. J Pediatr 1974; 84:887–888. 11. Rivers RPA. Neonatal opiate withdrawal. Arch Dis Child 1986; 61: 1236–1239. 12. Redmond DE Jr, Huang YH, Snyder DR, Maas JW. Behavioral effects of stimulation of the nucleus locus coeruleus in the stump-tailed monkey Macaca arctoides. Brain Res 1976; 116: 502–510. 13. Redmond DE Jr. Alterations in the function of the nucleus locus coeruleus: a possible model for studies of anxiety. In: Hanin I, Usdin E, eds. Animal Models in Psychiatry and Neurology. New York: Pergamon Press, 1977; pp 293–304. 14. Cedarbaum JM, Aghajanian GK. Catecholamine receptors on locus coeruleus neurons: Pharmacological characterization. Eur J Pharmacol 1977; 44:375–385. 15. Pert CB, Kuhar MJ, Snyder SH. Opiate receptor: autoradiographic localization in rat brain. Proc Natl Acad Sci USA 1976; 73:3729–3733. 16. Atweh SF, Kuhar MJ. Autoradiographic localization of opiate receptors in rat brain II. The brain stem. Brain Res 1977; 129: 1–12. 17. Bird SJ, Kuhar MJ. Iontophoretic application of opiates to the locus coeruleus. Brain Res 1977; 122:523–533. 18. Cedarbaum JM, Aghajanian GK. Noradrenergic neurons of the locus coeruleus: Inhibition by epinephrine and activation by the α-antagonist piperoxane. Brain Res 1976; 112:413– 419. 19. Korf J, Bunney BS, Aghajanian GK. Noradrenergic neurons: morphine inhibition of spontaneous activity. Eur J Pharmacol 1974; 25:165–169. 20. Hamburg M, Tallman JF. Chronic morphine administration increases the apparent number of α2-adrenergic receptors in rat brain. Nature 1981; 291:493–495. 21. Aghajanian GK. Tolerance of locus coeruleus neurons to morphine and suppression of withdrawal response by clonidine. Nature 1978; 276:186–188. 22. Henry JL. Effects of substance P on functionally identified units in cat spinal cord. Brain Res 1976; 114:439–451. 23. Jessell TM. Substance P in nocioceptive sensory neurons. In: Substance P in the Nervous System, Ciba Foundation Symposium 1982; 91:225–248. 24. Yakish TL, Jessell TM, Gamse R, Mudge AW, Leeman SE. Intrathecal morphine inhibits substance P release from mammalian spinal cord in vivo. Nature 1980; 286:155–157. 25. Lembeck F, Donnerer J. Opioid control of the function of primary afferent substance P fibres. Eur J Pharmacol 1985; 114:241–246. 26. Vacca LL, Abrahams SJ, Naftchi NE. Effect of morphine on substance P neurons in rat spinal cord: a preliminary study. Brain Res 1980; 182:229–236. 27. Bergstrom L, Sakurada T, Terenius L. Substance P levels in various regions of the rat central nervous system after acute and chronic morphine treatment. Life Sci 1984; 35:2375–2382. 28. Bell JA, Jaffe JH. Electrophysiological evidence for a presynaptic mechanism of morphine withdrawal in the neonatal rat spinal cord. Brain Res 1986; 382: 299–304. 29. Gintzler AR, Scalisi JA. Effects of opioids on noncholinergic excitatory responses of the guinea-pig isolated ileum: Inhibition of release of enteric substance P. Br J Pharmacol 1982; 75:199–205. 30. Schulz R, Herz A. Aspects of opiate dependence in the myenteric plexus of the guinea pig. Life Sci 1976; 19:1117–1128. 31. Zelson C, Rubio E, Wasserman E. Neonatal narcotic addiction: 10 year observation. Pediatrics 1971; 48:178–189. 32. Olofsson M, Buckley W, Andersen GE, Friis-Hansen B. Investigation of 89 children born by drug-dependent mothers: I. Neonatal course. Acta Paediatr Scand 1983; 72:403–406.

368

Besunder and Bloomer

33. Reddy AM, Harper RG, Stern G. Observations on heroin and methadone withdrawal in the newborn. Pediatrics 1971; 48:353–358. 34. Klenka HM. Babies born in a distinct general hospital to mothers taking heroin. Br Med J 1986; 293: 745–746. 35. Newman RG, Bashkow S, Calko D. Results of 313 consecutive live births of infants delivered to patients in the New York City methadone maintenance treatment program. Am J Obstet Gynecol 1974; 121:233–237. 36. Naeye RL, Blanc W. Leblanc W, Khatamee MA. Fetal complications of maternal heroin addiction: abnormal growth, infections and episodes of stress. J Pediatr 1973; 83:1055–1061. 37. Kandall SR, Albin S, Lowinson J, Berle B, Eidelman AI, Gartner LM. Differential effects of maternal heroin and methadone use on birthweight. Pediatrics 1976; 58:681–685. 38. Finnegan LP. Effects of maternal opiate abuse on the newborn. Fed Proc 1985; 44:2314– 2317. 39. Zelson C, Lee SJ, Casalino M. Comparative effects of maternal intake of heroin and methadone. N Engl J Med 1973; 289:1216–1220. 40. Rahbar F. Observations on methadone withdrawal in 16 neonates. Clin Pediatr 1975; 14: 369–371. 41. Ostrea EM Jr, Chavez CJ. Perinatal problems (excluding neonatal withdrawal) in maternal drug addiction: A study of 830 cases. J Pediatr 1979; 94:292–295. 42. Lifschitz MH, Wilson GS, O’Brian-Smith E, Desmond MM. Factors affecting head growth and intellectual function in children of drug addicts. Pediatrics 1985; 75:269–274. 43. Lifschitz MH, Wilson GS, O’Brian-Smith E, Desmond MM. Fetal and postnatal growth of children born to narcotic-dependent women. J Pediatr 1983; 102:686–691. 44. Nathenson G, Cohen MI, Litt IF, McNamara H. The effect of maternal heroin addiction on neonatal jaundice. J Pediatr 1972; 81:899–903. 45. Holmes AW, Rosenblate H, Einstein R, Baldwin D. The liver disease of heroin addiction (abstr). Gastroenterology 1970; 58:310. 46. Harper RG, Solish GI, Purow HM, et al. The effect of a methadone treatment program upon pregnant heroin addicts and their newborn infants. Pediatrics 1974; 54:300–305. 47. Glass L, Rajegowda BK, Evens HE. Absence of respiratory distress syndrome in premature infants of heroin-addicted mothers. Lancet 1971; 2:685–686. 48. Gluck R, Kulovich MV. Lecithin/sphingomyelin ratios in amniotic fluid in normal and abnormal pregnancy. Am J Obstet Gynecol 1973; 115:539–546. 49. Taeusch HW Jr, Carson SH, Wang NS, Avery ML. Heroin induction of lung maturation and growth retardation in fetal rabbits. J Pediatr 1973; 82:869–875. 50. Rosen TS, Pippenger CE. Pharmacologic observations on the neonatal withdrawal syndrome. J Pediatr 1976; 88:1044–1048. 51. Ostrea EM, Chavez CJ, Strauss ME. A study of factors that influence the severity of neonatal narcotic withdrawal. J Pediatr 1976; 88:642–645. 52. Gilman AG, Goodman LS, Rall TW, Murad F, eds. The Pharmacologic Basis of Therapeutics, 7th ed, Macmillan, New York, 1985, Appendix II, p 1695. 53. Sweet AY. Narcotic withdrawal syndrome in the newborn. Pediatr Rev 1982; 3:285–291. 54. Kron RE, Litt M, Finnegan LP. Effect of maternal narcotic addiction on sucking behavior of neonates (abstr). Pediatr Res 1974; 8:364. 55. Kron RE, Litt M, Eng D, et al. Neonatal narcotic abstinence: effects of pharmacotherapeutic agents and maternal drug usage on nutritive sucking behavior. J Pediatr 1976; 88:637–641. 56. Hyde WH, Scharnberg JT, Rudolph AJ. Oxygen consumption in infants of narcotic addicts (abstr). Pediatr Res 1980; 14:467. 57. Glass L, Rajegowda BK, Kahn EJ, Floyd MV. Effects of heroin withdrawal on respiratory rate and acid-base status in the newborn. N Engl J Med 1972; 286:746–748. 58. Olsen GD, Lees MH. Ventilatory response to carbon dioxide of infants following chronic prenatal methadone exposure. J Pediatr 1980; 96:983–989.

Neonatal Withdrawal Syndromes

369

59. Davidson SL, Schuetz S, Krishna V, et al. Abnormal sleeping ventilatory pattern in infants of substance-abusing mothers. Am J Dis Child 1986; 140:1015–1020. 60. Checola RT, Prybylski D, Senie R, Kandall SR. Cardiorespiratory patterns in passively addicted (PA) neonates (abstr). Pediatr Res 1986; 20:345A. 61. Herzlinger RA, Kandall SR, Vaughan HG Jr. Neonatal seizures associated with narcotic withdrawal. J Pediatr 1977; 91:638–641. 62. Chasnoff IJ, Hatcher R, Burns WJ. Early growth patterns of methadone-addicted infants. Am J Dis Child 1980; 134:1049–1051. 63. Wilson GS, Desmond MM, Verniaud WM. Early development of infants of heroin-addicted mothers. Am J Dis Child 1973; 126:457–462. 64. Chavez CJ, Ostrea EM, STryker JC, Smialek Z. Sudden infant death syndrome among infants of drug dependent mothers. J Pediatr 1979; 95:407–409. 65. Pierson PS, Howard P, Kleber HD. Sudden deaths in infants born to methadone-maintained addicts. JAMA 1972; 220:1733–1734. 66. Finnegan LP, Kron RE, Connaughton JF Jr, Emich JP. A scoring system for evaluation and treatment of the neonatal abstinence syndrome: A new clinical and research tool. In: Morselli PL, Garattini S, Serini F, eds. Basic and Therapeutic Aspects of Perinatal Pharmacology New York: Raven Press, 1975, pp 139–153. 67. Lipsitz PJ. A proposed narcotic withdrawal score for use with newborn infants. Clin Pediatr 1975; 14:592–594. 68. Hoder EL, Leckman JF, Ehrenkranz R, et al. Clonidine in neonatal narcotic-abstinence syndrome (letter). N Engl J Med 1981; 305:1284. 69. Cobrink RW, Hood T Jr, Chusid E. The effect of maternal narcotic addiction on the newborn infant: review of literature and report of 22 cases. Pediatrics 1959; 24:288–304. 70. Brown WJ, Buist NRM. Cory Gipson HT, et al. Fatal benzyl alcohol poisoning in a neonatal intensive care unit (letter). Lancet 1982; 1:1250. 71. Gershanik J, Boecler B, Ensley H, et al. The gasping syndrome and benzyl alcohol poisoning. N Engl J Med 1982; 307:1384–1387. 72. Committee on Fetus and Newborn, Committee on Drugs. Benzyl alcohol: toxic agent in neonatal units. Pediatrics 1983; 72:356–358. 73. Kunstadter RH, Klein RI, Lundeen EC, et al. Narcotic withdrawal symptoms in newborn infants. JAMA 1958; 168:1008–1010. 74. Kandall SR, Koberczak TM, Mauer KR, et al. Opiate v. CNS depressant therapy in neonatal drug abstinence syndrome. Am J Dis Child 1983; 137:378–382. 75. Carin I, Glass L. Parekh A, et al. Neonatal methadone withdrawal: Effect of two treatment regimens. Am J Dis Child 1983; 137:1166–1169. 76. Kaltenbach K, Finnegan LP. Neonatal abstinence syndrome, pharmacotherapy and developmental outcome. Neurobehav Toxicol Teratol 1986; 8:353–355. 77. Finnegan LP, Ehrlich S. Maternal drug abuse during pregnancy and pharmacotherapy for neonatal abstinence syndrome (NAS). Pediatr Res (abstr). 1987; 21:234A. 78. Strauss ME, Lessen-Firestone JK, Starr RH, Ostrea EM. Behavior of narcotics-addicted newborns. Child Dev 1975; 46:887–893. 79. Kaltenbach K, Finnegan LP, Frankenfield M. Neonatal abstinence syndrome (NAS) and interactive behaviors at 1 and 30 days of life. Pediatr Res (abstr). 1981; 15:450. 80. Olofsson M, Buckley W, Andersen GE, Friis-Hansen B. Investigations of 89 children born by drug-dependent mothers: follow-up 1–10 years after birth. Acta Paediatr Scand 1983; 72: 407–410. 81. Rosen TS, Johnson HL. Children of methadone-maintained mothers: follow-up to 18 months of age. J Pediatr 1982; 101:192–196. 82. Wilson GS, Desmond MM, Wait RB. Follow-up of methadone-treated and untreated narcotic-dependent women and their infants: health, developmental and social implications. J Pediatr 1981; 98:716–722.

370

Besunder and Bloomer

83. Kaltenbach K, Finnegan LP. Perinatal and developmental outcome of infants exposed to methadone in utero. Neotoxicol Teratol 1987; 9:311–313. 84. Kaltenbach K, Finnegan LP. Children exposed to methadone in utero: cognitive ability in the preschool years (abstr). Pediatr Res 1987; 21:181A. 85. Wilson GS, McCreary R, Kean J, Baxter JC. The development of preschool children of heroin-addicted mothers: a controlled study. Pediatrics 1979; 63:135–141. 86. Nelson LB, Ehrlich S, Calhoun JH, et al. Occurrence of strabismus in infants born to drugdependent women. Am J Dis Child 1987; 141:175–178. 87. Madden JD, Payne TF, Miller S. Maternal cocaine abuse and effect on the newborn. Pediatrics 1986; 77:209–211. 88. Ryan L, Ehrlich S, Finnegan L. Cocaine abuse in pregnancy: effects on the fetus and newborn. Neurotoxicol Teratol 1987; 9:295–299. 89. Chasnoff IJ, Burns KA, Burns WJ. Cocaine use in pregnancy: perinatal morbidity and mortality. Neurotoxicol Teratol 1987; 9:291–293. 90. Fulroth RF, Phillips BL, Trueax RE, Durand DJ. Description of 72 infants exposed to cocaine prenatally (abstr). Pediatr Res 1987; 21:361A. 91. Dixon SD, Coen RW, Crutchfield S. Visual dysfunction in cocaine-exposed infants (abstr). Pediatr Res 1987; 21:359A. 92. Oro AS, Dixon SD. Perinatal cocaine and methamphetamine exposure: maternal and neonatal correlates. J Pediatr 1987; 111:571–578. 93. Doberczak TM, Shanzer S, Senie RT, Randall SR. Neonatal neurologic and electroencephalographic effects of intrauterine cocaine exposure. J Pediatr 1988; 113:354–358. 94. Chasnoff IJ, Lewis DE, Squires L. Cocaine intoxication in a breast-fed infant. Pediatrics 1987; 80:836–838. 95. Stewart DJ, Inaba T, Lucassen M, Kalow W. Cocaine metabolism: cocaine and norcocaine hydrolysis by liver and serum esterases. Clin Pharmacol Ther 1979; 25:464–468. 96. Echobichon DJ, Stephens DS. Perinatal development of human blood esterases. Clin Pharmacol Ther 1973; 14:41–47. 97. Chasnoff IJ, Bussey ME Savich R, Stack CM. Perinatal cerebral infarction and maternal cocaine use. J Pediatr 1986; 108:456–459. 98. Weiner N. Norepinephrine, epinephrine and the sympathomimetic amines. In: Gilman AG, Goodman LS, Rall TW, Murad F, eds. The Pharmacological Basis of Therapeutics, 7th ed. New York: Macmillan, 1985, pp 145–180. 99. Ramer CM. The case history of an infant born to an amphetamine-addicted mother. Clin Pediatr 1974; 13:596–597. 100. Eriksson M, Larsson G, Windlbadh B, Zetterstrom R. The influence of amphetamine addiction on pregnancy and the newborn infant. Acta Paediatr Scand 1978; 67:95–99. 101. Gay GR. You’ve come a long way baby! Coke time for the new American lady of the eighties. J Psychoactive Drugs 1981; 13:297–318. 102. Chasnoff IJ, Burns WJ, Schnoll SH, Burns KA. Cocaine use in pregnancy. N Engl J Med 1985; 313:666–669. 103. Bingol N, Fuchs M, Diaz V, et al. Teratogenicity of cocaine in humans. J Pediatr 1987; 110: 93–96. 104. Chasnoff IJ, Griffith DR, MacGregor S, et al. Temporal patterns of cocaine use in pregnancy: perinatal outcome. JAMA 1989; 261:1741–1744. 105. Mahalik MP, Gautieri RF, Mann DE Jr. Teratogenic potential of cocaine hydrochloride in CF-1 mice. J Pharmacol Sci 1980; 69:703–706. 106. Bauchner H, Zuckerman B, McClain M, et al. Risk of sudden infant death syndrome among infants with in utero exposure to cocaine. J Pediatr 1988; 113:831–834. 107. Jones KL, Smith DW, Ulleland CN, Streissguth AP. Pattern of malformation in offspring of chronic alcoholic mothers. Lancet 1973; 1:1267–1271.

Neonatal Withdrawal Syndromes

371

108. Hanson JW, Jones KL, Smith DW. Fetal alcohol syndrome: experience with 41 patients. JAMA 1976; 235:1458–1460. 109. Jones KL, Smith DW. Recognition of the fetal alcohol syndrome in early infancy. Lancet 1973; 2:999–1001. 110. Schaefer O. Alcohol withdrawal syndrome in a newborn infant of a Yukon Indian mother. Can Med Assoc J 1962; 87:1333–1334. 111. Pierog S, Chandavasu O, Wexler I. Withdrawal symptoms in infants with the fetal alcohol syndrome. J Pediatr 1977; 90:630–633. 112. Nichols MM. Acute alcohol withdrawal syndrome in a newborn. Am J Dis Child 1967; 113: 714–715. 113. Devenyi P, Harrison ML. Prevention of alcohol withdrawal seizures with oral diazepam loading. Can Med Assoc J 1985; 132:798–800. 114. Sullivan JT, Sellers EM. Treating alcohol, barbiturate and benzodiazepine withdrawal. Rational Drug Therapy 1986; 20:1–8. 115. Sellers EM, Naranjo CA, Harrison M, et al. Diazepam loading: simplified treatment of alcohol withdrawal. Clin Pharmacol Ther 1983; 34:822–826. 116. Desmond MM, Schwanecke RP, Wilson GS, et al. Maternal barbiturate utilization and neonatal withdrawal symptomatology. J Pediatr 1972; 80:190–197. 117. Bleyer WA, Marshall RE. Barbiturate withdrawal syndrome in a passively addicted infant. JAMA 1972; 221:185–186. 118. Ostrea EM Jr. Neonatal withdrawal from intrauterine exposure to butalbital. Am J Obstet Gynecol 1982; 143:597–599. 119. Athinarayanan P, Pierog S, Nigam SK, Glass L. Chlordiazepoxide withdrawal in the neonate. Am J Obstet Gynecol 1976; 124:212–213. 120. Bitnum S. Possible effect of chlordiazepoxide on the fetus (letter). Can Med Assoc J 1969; 100:351. 121. Mazzi E. Possible neonatal diazepam withdrawal: a case report. Am J Obstet Gynecol 1977; 129:586–587. 122. Rementeria JL, Bhatt K. Withdrawal symptoms in neonates from intrauterine exposure to diazepam. J Pediatr 1977; 90:123–126. 123. Prenner BM. Neonatal withdrawal syndrome associated with hydroxyzine hydrochloride. Am J Dis Child 1977; 131:529–530. 124. Rumack BH, Walravens PA. Neonatal withdrawal following maternal ingestion of ethchlorvynol (Placidyl). Pediatrics 1973; 52:714–716. 125. Reveri M, Pyati SP, Pildes RS. Neonatal withdrawal symptoms associated with glutethimide (Doriden) addiction in the mother during pregnancy. Clin Pediatr 1977; 16:424–425. 126. Webster PAC. Withdrawal symptoms in neonates associated with maternal antidepressant therapy (letter). Lancet 1973; 2:318–319. 127. Eggermont E. Neonatal effects of maternal therapy with tricyclic antidepressant drugs (letter). Arch Dis Child 1980; 55:81. 128. Ben Musa A, Smith CS. Neonatal effects of maternal clomipramine therapy (letter). Arch Dis Child 1979; 54:405. 129. O’Conner M, Johnson GH, James DI. Intrauterine effects of phenothiazines. Med J Aust 1981; 1:416–417. 130. Baldessarini RJ. Drugs and the treatment of psychiatric disorders. Gilman AG, Goodman LS, Rall TW, Murad F, eds. In: The Pharmacologic Basis of Therapeutics, 7th ed. New York: Macmillan, 1985, pp 387–445.

20 Current Management of the Neonatal Abstinence Syndrome: A Critical Analysis of the Evidence Jochen G. W. Theis and Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

Peter L. Selby St. Joseph’s Health Centre, Toronto, Ontario, Canada

Yazemir Ikizler The Addiction Research Foundation, Toronto, Ontario, Canada

INTRODUCTION Continuous or repeated fetal exposure to addicting drugs may lead to fetal drug dependence and after birth to the neonatal withdrawal or neonatal abstinence syndrome (NAS). While it is, of course, preferable to prevent NAS from occurring by reducing the incidence of maternal drug abuse, it needs to be acknowledged that prevention is not possible in many cases. Continuous fetal exposure to drugs of abuse is not uncommon. While a fraction of drug-using women discontinue the consumption after pregnancy has been diagnosed, others continue their use or consume replacements throughout pregnancy. A recent analysis estimated that during the period from 1988 to 1990 about 90,000 American women were discharged after parturation every year with a diagnosis of drug abuse, representing a rate of 2.2% of all women giving birth (1). NAS is most commonly observed after maternal use of opioids such as heroin (2,3), methadone (2,4,5), and even codeine (6,7), but it may also be observed after continuous fetal exposure to barbiturates (8), alcohol (9,10), or other psychoactive drugs with the potential to cause physical addiction (11,12). It is generally agreed that maternal heroin addiction during pregnancy should be treated by replacement with methadone which may be gradually reduced in dose (13,14). Abrupt withdrawal of opioids during pregnancy is not recommended because this may lead to major maternal withdrawal symptoms endangering both the mother and the fetus. Moreover, unplanned opiate withdrawal may lead to resumption of street opiate use or consumption of replacement drugs and may further disrupt prenatal care (15). The major drawback of methadone replacement therapy is that continuous fetal opioid exposure may often lead to NAS (16). From Biol Neonate 1997; 71:345–356. 373

374

Theis et al.

Table 1 Drugs Used to Treat NAS with Examples of Published Dosages Drug Phenobarbital Diazepam Chlorpromazine Paregoric Morphine Methadone Clonidine

Dosage 5–16 mg/kg/day q 8 h or loading dose: 16 mg/kg followed by 2–6 mg/kg/day 0.5–2.5 mg/kg/day q 8 h 2.8 mg/kg/day q 6 h 0.4–2.1 ml/kg/day q 3 h 1.6–6.4 ml/kg/day q 3 h 0.4–1.0 mg/kg/day 2–4 mg/day q 6 h 3–4 µg/kg/day

Citations Kandall et al. (29), 1983 Finnegan et al. (30), 1979 Herzlinger et al. (31), 1977 Kahn et al. (35), 1969 Kandall et al. (29), 1983 Herzlinger et al. (31), 1977 Pacifico et al. (27), 1989 Madden et al. (32), 1977 Hoder et al. (28), 1984

NAS is characterized by tremor, irritability, hypertonicity, high-pitched cry, vomiting and diarrhea, respiratory distress, sneezing, diaphoresis and fever, poor sucking and rarely convulsions (17,18). The onset of symptoms after continuous fetal exposure to opioids is usually observed within 48–72 hours after birth (16,19), and may be delayed after fetal exposure to barbiturates (20,21), or long-acting benzodiazepines (22). A more elaborate description of the signs and symptoms of NAS has been the topic of a number of reviews (12,18,23,24) and will thus not be discussed here in further details. Conservative comfort measures such as holding, swaddling, and minimal stimulation may be sufficient treatment if symptoms are mild and not progressing (12,18,21). However, it is generally agreed that more severe symptoms should be treated by adequate pharmacotherapy (12,18,24). It is important to realize that the aim of this pharmacotherapy should be to stabilize the neonate as far as physical dependence is concerned so that he or she can handle the other stresses of the newborn period (25). In addition to the suppression of tremors, hyperactivity, seizure activity, and other apparent symptoms, therapy should aim at restoring normal neonatal neurological patterns such as sleep and sucking. A number of different agents have been employed in order to achieve this aim, the ones most frequently cited are phenobarbital, opioids (mainly paregoric), chlorpromazine, diazepam, and more recently clonidine (26–35) (Table 1). Because these substances have very different modes of action, it should be expected that the desired effects and adverse effects may show marked differences depending on the agent employed. Nevertheless, in reviewing the literature, it appears that all of these agents are currently in use for the treatment of NAS. One of the most important changes in medicine during the late 1980s and the early 1990s has been the push toward evidence-based practice. It is more and more agreed that patient management and especially pharmacological therapy should be based on the available evidence with emphasis on data collected in well-controlled clinical trials (36). The aim of this article is to review and evaluate the evidence available in the present literature for the selection of agents effective in the treatment of NAS.

METHODS A literature review was performed for the years 1966 to April 1996 utilizing MEDLINE. To identify the articles dealing with the topic of NAS and its treatment, the following

Current Management of the Neonatal Abstinence Syndrome

375

search pattern using medical subject headings (MESH) was employed: ‘‘neonatal abstinence syndrome’’ or ‘‘substance withdrawal syndrome’’ [limited to ‘‘newborn infant (birth to 1 month)’’]. This search pattern was necessary since the MESH ‘‘neonatal abstinence syndrome’’ was not used by MEDLINE before 1988 (37). Citations dealing with NAS and published before 1988 are identifiable by limiting the MESH ‘‘substance withdrawal syndrome’’ which includes the earlier MESH ‘‘drug withdrawal symptoms’’ to ‘‘newborn infant (birth to 1 month).’’ The resulting set of citations was combined using the operator ‘‘and’’ with the set of citations identified by the MESH combination ‘‘comparative study’’ or ‘‘evaluation studies.’’ The resulting set of citations and additionally the citations identified by the pattern ‘‘explode opioids’’ [limited to ‘‘newborn infant (birth to 1 month)’’] were screened by title MESH and, whenever available, abstract. Every citation that was thought to have any chance of containing data or original information about the treatment of NAS was obtained for evaluation. The references given in review articles about NAS written during the last 5 years and the references given in the articles obtained for information were further screened for any additional citations which may possibly identify subsequent studies comparing or evaluating the effectiveness of pharmacological treatment. The reports identified by this search were analysed regarding study design and outcome to identify a group of well-conducted studies which can serve as a basis for a metaanalysis. The analysis of study design addressed the following items: assignment criteria (random/nonrandom), number of neonates treated for NAS, blinding of observers regarding treatment or fetal exposure, methods to obtain and confirm fetal drug exposure, and measurements of outcome.

RESULTS ‘‘Neonatal abstinence syndrome’’ or ‘‘substance withdrawal syndrome’’ [limited to ‘‘newborn infant (birth to 1 month)’’] was a topic of 638 citations. When these citations were combined with the set of citations labeled with ‘‘evaluation studies’’ or ‘‘comparative study,’’ this number was reduced to 42 citations. Only 5 of these contained comparisons between medical treatments of the infant suffering from NAS. Of the 638 citations, 134 were labeled to deal with the therapy of NAS by having the abbreviations dt or th (dt meaning drug therapy, the standing for therapy) attached to the MESH headings ‘‘neonatal abstinence syndrome’’ or ‘‘substance withdrawal syndrome.’’ The vast majority of these articles, 129, do not contain original data about comparative trials of the treatment of NAS, 124 of these indicate this by not being labeled with the MESH ‘‘evaluation studies’’ or ‘‘comparative study.’’ This search strategy would have resulted in the same five citations found after the initial search. A total of 14 references containing clinical comparisons between drug treatments for NAS were identified by screening all initial 638 citations, the citations identified by the MESH search pattern ‘‘explode opioids’’ [limited to ‘‘newborn infant (birth to 1 month)’’] and hand searching the references stated in the obtained articles and recent review papers and chapters. The number of 14 identified citations may overestimate the number of independent studies since four studies published by the group of Finnegan et al. (38,39) in 1975 and Kron et al. (40,41) in 1975 and 1976 appear to be successive publications of results during validation and implication of their study design. It is not clear but likely that the results reported in these publications involve some of the same

376

Theis et al.

patients. Similarly, the report of Kaltenbach and Finnegan (23) (1986) mainly evaluating the developmental effects of NAS and its pharmacological treatment also contains data about the immediate effectiveness of the pharmacological treatment of NAS. These data are similar to the data presented by Finnegan and Michael (42) 2 years earlier. It is not clear whether the data dealing with the effectiveness of the pharmacological treatment differ between the studies or whether the results are based on the same patient population. Of these up to 14 comparative studies only 8 stated that treatments were assigned at random (29,32,35,38,39,42–44). Moreover, the methods of randomization employed in the majority of these 8 studies are not accepted as valid today. Observers and raters were not blinded in any of the studies regarding the type of treatment. Seven reports did not report how fetal exposure was verified (23,27,32,38–41), two studies relied on maternal history only (35,43), and five studies specified that maternal or fetal urine toxicology was performed to confirm fetal drug exposure (29,31,42,44,45). Main outcome measures varied widely. Incidence, prevalence, and control of a variety of symptoms were assessed clinically or with the help of neonatal abstinence scales. Objective nutritional sucking measures were the main outcome measure in four studies which appear to be related (38,39), while one study compared developmental progress at 6 months age (23). The studies are summarized in Table 2. Due to the many different therapy regimens employed, the large differences in outcome measures, and flaws in study design in the majority of the studies, no set of studies qualified for a combined statistical analysis in the form of a meta-analysis. Therefore, the studies were critically reviewed to highlight the methodological strengths and weaknesses and to extract and summarize the pertinent data. The drugs leading to NAS were most commonly opioids, although fetal exposure to multiple drugs was not excluded. The drugs evaluated in these 14 comparative studies were phenobarbital (in twelve of the studies), paregoric [eleven], diazepam [nine], chlorpromazine [one], morphine [one] and methadone [one]. In contrast to the other drugs, paregoric is a preparation containing a number of substances, most of them pharmacologically active: opium, benzoic acid, camphor, glycerin, and 44–46% alcohol. The preparation of paregoric was originated in 1715 (46) and has more recently been described as ‘a needlessly complex therapeutic survival of a former day (47). The composition of the formula has changed numerous times since its introduction into medicine (46) and still slightly differs between the United States and the British Pharmacopoeiae. None of the reports on its efficacy in NAS state the exact contents of the paregoric formulation used. When compared to pure opioids, it is questionable whether a cocktail of substances which includes a substantive amount of alcohol and the central stimulant camphor may in fact be of superior or equal value in the treatment of NAS. Nevertheless, paregoric was the opioid of choice in 11 of the 14 comparative studies. Only two comparative studies evaluated pure opioids in the treatment of NAS. Pacifico et al. (27) concluded that morphine alone was superior to the combination of phenobarbital and diazepam or to the combination of morphine, phenobarbital, and diazepam. In contrast, Madden et al. (32) found no statistical differences between treatments consisting of methadone, phenobarbital or diazepam. Unfortunately, the report by Pacifico et al. (27) contains too few details on its study design to be interpretable. In the study by Madden et al. (32) the decision to continue or to discontinue treatment was made purly on clinical grounds without the help of a standardized abstinence score. Nevertheless, length of treatment and hospital stay were the main outcome measures. Because these

Current Management of the Neonatal Abstinence Syndrome

377

outcome measures may have been the result of decisions based on personal biases and other confounders, this report also has to be interpreted with caution. Phenobarbital was compared in different studies to chlorpromazine, paregoric, diazepam, and methadone. The study by Kahn et al. (35) comparing phenobarbital to chlorpromazine was the first after 1966 to comparatively evaluate pharmacological treatments for NAS. This study reported no significant differences in outcome after the treatment with either of the two drugs used in a short or long course. Outcome measure was assessment of irritability and tremors. Since phenobarbital and chlorpromazine both have sedating effects, the control of tremors and irritability may be achieved in a comparable way by either drug. Whether there were differences in the control of other symptoms of NAS such as sleep pattern or sucking and feeding performance was not evaluated. Moreover, fetal drug exposure was assessed by maternal history only. Almost one quarter of the neonates were born to mothers who were also fairly heavy drinkers, more than half of the neonates were born to heavy smokers, and only about one third of the mothers received antenatal care. These and other potential cofactors were not accounted for in the analysis of the data. It thus remains questionable whether the outcome would have been similar in a well-controlled study using objective or quantitative outcome measures. Phenobarbital was also compared to paregoric. Carin et al. (43) reported that phenobarbital treatment was necessary for an average of only 17 days compared to 22 days of treatment with paregoric. The authors acknowledged that this difference might have been due to the long elimination half-life of phenobarbital which prolongs its pharmacological action after the termination of the dose. There were also differences between the study groups which were not accounted for in the analysis: more than half of the infants receiving paregoric compared to only one-fifth of the infants treated with phenobarbital were exposed to multiple drugs of abuse in utero. In another study by Kandall et al. (29) comparing phenobarbital with paregoric, seizures were observed in 11% of the neonates treated with phenobarbital while no seizures were observed in the group of neonates treated with paregoric, this was statistically significant ( p ⬍ 0.025). Neonatal abstinence scores were not different between the two groups during treatment. In a study performed by Finnegan and Michael (42) paregoric successfully controlled withdrawal symptoms in 93% of the neonates exposed to narcotics in utero while phenobarbital and diazepam given at the highest dose were only successful in 50 and 0% of the treated neonates, respectively. In neonates exposed to multiple drugs in utero, phenobarbital was efficacious in the control of NAS symptoms in 89% of the infants while paregoric and diazepam were efficacious in 61 and 40%. The differences between paregoric and phenobarbital were significant in both comparisons. In a subsequent study by Finnegan and Ehrlich (44) utilizing the same study design, similar results were reported on a larger number of patients. In another study by Kaltenbach and Finnegan (23) focusing on the development after 6 months of life, no difference was seen among treatments with phenobarbital and paregoric. In the same study, it was reported that phenobarbital had been less efficacious in the control of NAS symptoms than paregoric but more efficient than diazepam. Important studies were performed by Finnegan et al. (38,39) and Kron et al. (40,41), who compared the rate and effectiveness of the nutritional sucking behavior of neonates suffering from NAS after fetal methadone exposure. While paregoric almost restored sucking measures to normal, phenobarbital had little effect compared to no pharmacological treatment. Diazepam worsened nutritive sucking behavior compared to no treatment or

Table 2

Clinical Studies Evaluating Drug Treatment for NAS Main outcome measurement

Phenobarbital, chlorpromazine Phenobarbital paregoric

Tremors, irritability

Phenobarbital, paregoric

Sucking measures

Phenobarbital, paregoric, diazepam, no therapy, normal control

Sucking measures

Phenobarbital, paregoric, diazepam, no therapy, normal control

Sucking measures

Paregoric, diazepam

Seizures

Paregoric, diazepam

Comparison of therapy not major aim of study

Sucking measures

No significant difference Paregoric ⬎ no treatment ⫽ phenobarbital Group 3: paregoric ⬎ no treatment ⬎ phenobarbital Paregoric ⬎ no treatment ⬎ phenobarbital ⬎ diazepam in restoring sucking pattern towards normal Normal control ⬎ paregoric ⬎ no treatment ⬎ phenobarbital ⬎ diazepam in restoring sucking pattern towards normal Incidence (p ⬍ 0.01) and control (p ⬍ 0.05) of seizures better with paregoric Convulsive activity occurred more often in the diazepam group (p ⬍ 0.01) No significant difference Paregoric 22 days, phenobarbital 17 days, p ⬍ 0.01

Assessment of main outcome

Maternal drug of abuse

Fetal exposure confirmed by. . .

Type of study

Neonates treated

Citation

Clinical observation

Heroin

History

Randomized

38

Objective measures with sucking instrument Objective measures with sucking instrument Objective measures with sucking instrument

Primarily methadone or heroin

Not stated

Randomized

38

Methadone or heroin

Not stated

Not randomized ⫹ randomized

26 ⫹ 38

Kron et al. (40), 1975

Methadone or heroin

Not stated

Assignment criteria not stated

39

Kron et al. (41), 1975

Objective measures with sucking instrument

Primarily methadone or heroin

Not stated

Assignment criteria not stated

26

Finnegan et al. (39), 1976

Clinical; in 72% of the patients with clinical seizures: EEG

Methadone, heroin and others

History, urine assays

Nonrandomized

60

Herzlinger et al. (31), 1977

Not mentioned

Heroin or methadone

Assignment criteria not stated

Clinical assessment not well standardized Neonatal abstinence scale developed by Finnegan et al. (38)

Mainly methadone and/or heroin

History, maternal and neonatal urine toxicology ‘when available’ Not stated

Maternal history

Methadone, 20–63% multidrug abusers

Kahn et al. (35), 1969 Finnegan et al. (38), 1975

132

Kendall et al (45), 1977

Randomized

51

Madden et al. (32), 1977

Randomized

31

Carin et al. (143), 1983

Theis et al.

Methadone, phenobar- Length of therapy rebital, diazepam quired, length of hospital stay Paregoric, phenobarDuration of necessary bital treatment

Result

378

Therapy regimen evaluated

Incidence of seizures; neonatal abstinence scores on days 3– 28

Paregoric, phenobarbital, diazepam,

Control of symptoms with highest dose of the agent employed

Paregoric, phenobarbital, diazepam, combination of above

Development after 6 months of life and control of symptoms

Morphine, phenobarbital ⫹ diazepam, morphine ⫹ phenobarbital ⫹ diazepam Paregoric, phenobarbital, diazepam

Daily average score on neonatal abstinence scale

Neonatal abstinence scale developed by Lipsitz (57) (modified)

Methadone, 52–61% multidrug abusers

History, urine toxicologic screening of mother and neonate

Randomized

111

Kandall et al. (29), 1983

Neonatal abstinence scale developed by Finnegan et al. (38)

Narcotic agents or multidrug abusers (analyzed separately)

Medical records, social data forms, urine toxicologies

Randomized unless clinically indicated

139

Finnegan et al. (42), 1984

Control of symptoms: not mentioned; development: Bayley Scale of Mental Development

Methadone polydrug abuse?

Not stated

Assignment criteria not stated

69

Kaltenbach and Finnegan (23), 1986

Neonatal abstinence scale developed by Finnegan et al. (38)

Heroin

Not stated

Assignment criteria not stated

25

Pacifico et al. (27), 1989

Medical records, social data forms, urine toxicologies Randomized

Randomized

Efficacy in opiate-ex- Days of treatment Narcotic agents or and days until synposed neonates: multidrug abusers dromes controlled; paregoric ⬎ pheno(analyzed sepatreatment and outbarbital ⬎ diazerately) pam efficiency in come dependent on scores on neonatal nonopiate-exposed abstinence scale deneonates: phenobarveloped by Finbital ⬎ diazepam ⬎ paregoric negan et al. (38)

176

Finnegan and Ehrlich (44), 1990

379

Days until abstinence symptoms controlled and days of treatment needed separated into neonates antenatally exposed to opiates, nonopiates, or both

Seizures occurring in 7 patients on phenobarbital and no patients on paregoric (p ⬍ 0.25); no difference in abstinence scores Success of treatment in: infants exposed in utero to opioids, paregoric 93%, phenobarbital 50%, diazepam 0%; infants exposed to multiple drugs: paregoric 61%, phenobarbital 89%, diazepam 40% Development: no difference between paregoric, phenobarbital and polytherapy; control of symptoms: paregoric 91%, phenobarbital 47%, diazepam 0% Morphine monotherapy better than polytherapy

Current Management of the Neonatal Abstinence Syndrome

Paregoric, phenobarbital

380

Theis et al.

treatment with the other two agents. These studies are the only studies using a measure which is clearly objective (neonatal abstinence scores have a subjective component). However, the interpretation of the data regarding diazepam is hampered by the small numbers (five or six) of neonates assigned to this particular treatment. Diazepam was compared to paregoric in one study focusing on seizure activity as part of NAS. Herzlinger et al. (31) reported that paregoric was more effective in controlling and preventing seizures once they had occurred. However, it has to be noted that the patients were not randomized to different treatments and that the use of diazepam was discouraged during the second half of the study when concomitant treatment might have been improving. In a report by Kandall et al. (45), convulsive activity occurred more frequently when diazepam rather than camphorated opium tincture was used to treat NAS. However, neither methods of treatment assignments nor numbers regarding this outcome are given. A comparison between treatments was clearly not the major aim of this report.

DISCUSSION The available studies performed to date do not allow meaningful comparisons in the efficacy of paregoric, morphine, methadone, phenobarbital, chlorpromazine, and diazepam when used for the treatment of NAS. It may be possible to single out diazepam as an undesirable agent, since it may worsen nutritive sucking behavior and may provide less seizure control than opioids. Paregoric appears to be superior in restoring sucking behavior, but similar superiority was not generally demonstrated when the main outcome was the clinical withdrawal symptoms as assessed by neonatal abstinence scores. Whether this may also be representative for opioid preparations that do not contain camphor and alcohol remains unknown. Phenobarbital was shown in one study to be superior to paregoric in controlling NAS in a subgroup of neonates who had been exposed to multiple drugs in utero. However, when not discriminating subgroups of neonates with NAS or when evaluating the efficacy of phenobarbital in neonates exposed only to opiates, the results were controversial with a tendency towards achieving better control of NAS symptoms in neonates treated with paregoric. The efficacy of chlorpromazine has only been addressed in a single study dating back to 1969. This study, being by design not very discriminative, resulted in no differences between the treatment with chlorpromazine and phenobarbital. In spite of the nearly complete lack of proof of efficacy, chlorpromazine has remained in use for the treatment of NAS as documented by a number of reports published in the 1990s (48–50). The efficacy of clonidine in controlling symptoms of NAS has not been evaluated in comparative studies. In general, it has to be acknowledged that the data regarding the relative efficacy of the different pharmacological treatment options of NAS are weak. An attempt to use adult data as an additional source of information is of limited value. Adult opiate withdrawal differs from NAS because of the comparative lack of seizures and other life-threatening complications. Therefore, it is regarded safe for adults to be detoxified without pharmacotherapy. Sedating agents such as phenobarbital, diazepam and chlorpromazine are mainly used as adjacent symptomatic therapy to reduce the severe discomfort. Cocktails such as paregoric or tincture of opium do not even appear in the recent adult literature. However, the impact of substance withdrawal on the developing brain at its greatest adjustment to extrauterine life may not allow to compare it to adult substance withdrawal.

Current Management of the Neonatal Abstinence Syndrome

381

Our analysis of available studies documents lack of strong evidence on the relative efficacy of different pharmacological regimens for the treatment of NAS. NAS, although often treated as one entity, cannot be regarded as a single pathological condition. It has been shown that the symptoms of NAS differ depending on the substances the fetus was exposed to. Neonates continuously exposed to alcohol in utero tend to have more seizures, opistotonos, and abdominal distention, while neonates exposed to opioids tend to have a high-pitched cry and suffer from sleeplessness and excessive but ineffective sucking (10). In contrast to opioids and alcohol, cocaine probably does not cause typical neonatal withdrawal symptoms (51). However, the drug or its metabolites are often still present in the neonate shortly after birth and may continue to have a toxic effect (52). These adverse effects may be confused with clinical signs of NAS. A study by Fulroth and Phillips (53) showed that neonates required treatment in less than half of the cases after fetal cocaine exposure compared to after fetal heroin exposure. When management was deemed necessary, the mean duration of treatment in cocaine-exposed neonates was only 2 days compared to 14 days in the heroin-exposed group. Clinical symptomatology thus may be different but is not pathognomonic; without a detailed and correct maternal history, the different etiologies cannot be distinguished with certainty. Unfortunately, maternal history is all too often an inaccurate source of information on drug abuse (54). In addition, polydrug abuse is common in the population of women taking illicit drugs during pregnancy. Utilization of modern analytical techniques such as meconium or neonatal hair analysis may allow better estimation of fetal drug exposure. Such knowledge, when built into the design of treatment trials of NAS, may allow for greater discrimination among treatment effects. Many confounders are often associated with pregnancies complicated by drug abuse. These include poor antenatal care, smoking, poverty, personal neglect, and poor health status. None of the existing studies evaluating the treatment of NAS have corrected for these confounders. Yet, during the last 10 years significant progress has been made in the quality of study design directed toward the evaluation of fetal exposure and outcome. Applying these standards to future studies on NAS is likely to allow clearer analysis of outcome measures. Careful assessment of outcome measures is a crucial element in controlled studies. NAS is a complex combination of clinical signs, which may respond differently to various pharmacological agents. While early studies assessed only tremor and irritability, a number of scoring scales have been developed, which range from simple indiscriminatory scales to very complex scales which assess a wide variety of symptoms (38,40,55–57). When selecting or developing a scale to monitor the pathology of NAS and the effectiveness of its treatment, it is important to understand the strengths and shortcomings of each of the scales. The most widely used and best validated scale today was developed by Finnegan et al. (38) and Kron et al. (40). However, even in this scale, sedation of a neonate suffering from NAS may lead to an improvement in scores. It is important to realize that the aim of treatment of NAS should also be to allow the neonate to adapt to and function in the extrauterine world. Utilizing other objective measures such as nutritive sucking measures or sleep pattern as additional independent endpoints may take this aim into account and may further discriminate between the effects of various modes of treatment. Neonates suffering from NAS deserve optimal care including the best possible pharmacological approach. Presently, good studies providing the evidence for the selection of the best pharmacotherapy for neonates with NAS are clearly missing. Centers caring for large numbers of neonates exposed to addicting drugs in utero should consider utilizing modern clinical epidemiological tools and up-to-date study design to produce the evidence

382

Theis et al.

needed to allow a rational choice between treatment modalities, and thus to lead to better future care of these neonates. REFERENCES 1. Dicker M, Leighton EA. Trends in the US prevalence of drug-using parturient women and drug-affected newborns, 1979 through 1990. Am J Publ Health 1994; 84:1433–1438. 2. Stimmel B, Goldberg J, Reisman A, et al. Fetal outcome in narcotic-dependent women: the importance of the type of maternal narcotic used. Am J Drug Alcohol Abuse 1982–83; 9: 383–395. 3. Zelson C, Kahn EJ, Neumann L, Polk G. Heroin withdrawal syndrome. J Pediatr 1970; 76: 483–484. 4. Doberczak TM, Kandall SR, Friedmann P. Relationship between maternal methadone dosage, maternal-neonatal methadone levels, and neonatal withdrawal. Obstet Gynecol 1993; 81:936– 940. 5. Newman RG, Bashkow S, Calko D. Results of 313 consecutive live births of infants delivered to patients in the New York City Methadone Maintenance Treatment Program. Am J Obstet Gynecol 1975; 121:233–237. 6. Mangurten HH, Benawra R. Neonatal codeine withdrawal in infants of nonaddicted mothers. Pediatrics 1980; 65:159–160. 7. Van Leeuwen G, Guthrie R, Stange F. Narcotic withdrawal reaction in a newborn infant due to codeine. Pediatrics 1965; 36:635–636. 8. Blumenthal I, Lindsay S. Neonatal barbiturate withdrawal. Postgrad Med J 1977; 53:157– 158. 9. Coles CD, Smith IE: Neonatal ethanol withdrawal: characteristics in clinically normal, nondysmorphic neonates. J Pediatr 1984; 105:445–451. 10. Robe LB, Gromisch DS, Iosub S. Symptoms of neonatal ethanol withdrawal. Curr Alcohol 1981; 8:485–493. 11. Auerbach JG, Hans SL, Marcus J, Maeir S. Maternal psychotropic medication and neonatal behavior. Neurotoxicol Teratol 1992; 14:399–406. 12. Levy M, Spino M. Neonatal withdrawal syndrome: Associated drugs and pharamacologic management. Pharmacotherapy 1993; 13:202–211. 13. Finnegan LP. Treatment issues for opioid-dependent women during the perinatal period. J Psychoactive Drugs 1991; 23:191–201. 14. Report of the Expert Advisory Committee on the use of drugs in the treatment of abuse and dependence to narcotic and controlled drugs. Can Med Assoc J 1990; 143:861–865. 15. Connaughton JF, Reeser D, Schut J, Finnegan LP. Perinatal addiction: Outcome and management. Am J Obstet Gynecol 1977; 129:679–686. 16. Zelson C, Lee SJ, Casalino M. Neonatal narcotic addiction. Comparative effects of maternal intake of heroin and methadone. N Engl J Med 1973; 289:1216–1220. 17. Harper RG, Solish G, Feingold E, et al. Maternal ingested methadone, body fluid methadone, and the neonatal withdrawal syndrome. Am J Obstet Gynecol 1977; 129:417–424. 18. Finnegan LP, Kaltenbach K. Neonatal abstinence syndrome. In: Hoekelman RA, Friedman SB, Nelson NM, Seidel HM, eds. Primary Pediatric Care, 2nd ed. St. Louis: Mosby–Year Book, 1992, pp 1367–1378. 19. Desmond MM, Wilson GS. Neonatal abstinence syndrome. Recognition and diagnosis. Addict Dis 1975; 2:113–121. 20. Desmond M, Schwanecke R, Wilson G. et al. Maternal barbiturate utilization and neonatal withdrawal symptomatology. J Paediatr 1972; 80:190–197. 21. Blinick G, Wallach R, Jerez E, et al Drug addiction in pregnancy and in the neonate. Am J Obstet Gynecol 1976; 125:135–142.

Current Management of the Neonatal Abstinence Syndrome

383

22. Sutton LR, Hinderliter SA. Diazepam abuse in pregnant women on methadone maintenance: implications for the neonate. Clin Pediatr 1990; 29:108–111. 23. Kaltenbach K, Finnegan LP. Neonatal abstinence syndrome, pharmacotherapy and developmental outcome. Neurobehav Toxicol Teratol 1986; 8:353–355. 24. Franck L, Vilardi J. Assessment and management of opioid withdrawal in ill neonates. Neonatal Network 1995; 14:39–48. 25. Kron RE, in round table discussion of Zelson C. Acute management of neonatal addiction. Addict Dis 1975; 2:159–168. 26. Wijburg FA, deKleine MJ, Fleury P, Soepatmi S. Morphine as an antiepileptic drug in neonatal abstinence syndrome. Acta Paediatr Scand 1991; 80:875–877. 27. Pacifico P, Nardelli E, Pantarotto MF. Neonatal heroin withdrawal syndrome: Evaluation of different pharmacological treatments. Pharmacol Res 1989; 21(suppl 1):63–64. 28. Hoder EL, Leckman JF. Clonidine treatment of neonatal narcotic abstinence syndrome. Psychiatry Res 1984; 13:243–251. 29. Kandall SR, Doberczak TM, Mauer KR, et al. Opiate v CNS depressant therapy in neonatal drug abstinence syndrome. Am J Dis Child 1983; 137:378–382. 30. Finnegan LP, Mitros TF, Hopkins LE. Management of neonatal narcotic abstinence utilizing a phenobarbital loading dose method. NIDA Res Monogr 1979; 27:247–253. 31. Herzlinger RA, Kandall SR, Vaughan HG Jr. Neonatal seizures associated with narcotic withdrawal. J Pediatr 1977; 91:638–641. 32. Madden JD, Chappel JN, Zuspan F, et al. Observation and treatment of neonatal narcotic withdrawal. Am J Obstet Gynecol 1977; 127:199–201. 33. Rosen M. Use of diazepam in neonatal narcotic withdrawal syndrome. Pediatrics 1972; 49: 314. 34. Nathenson G, Golden GS, Litt IF. Diazepam in the management of the neonatal narcotic withdrawal syndrome. Pediatrics 1971; 48:523–527. 35. Kahn EJ, Neumann LL, Polk GA. The course of the heroin withdrawal syndrome in newborn infants treated with phenobarbital or chlorpromazine. J Pediatr 1969; 75:495–500. 36. Evidence-Based Medicine Working Group. Evidence-based medicine. JAMA 1992; 268: 2420–2425. 37. Medical Subject Headings-Annotated Alphabetic List 1995. Bethesda, MD: National Library of Medicine, 1994. 38. Finnegan LP, Connaughton JF Jr, Kron RE, Emich JP. Neonatal abstinence syndrome: Assessment and management. Addict Dis 1975; 2:141–158. 39. Finnegan LP, Kron RE, Connaughton JF Jr, Emich JP: Assessment and treatment of abstinence in the infant of the drug-dependent mother. Int J Clin Pharmacol 1975; 12:19–32. 40. Kron RE, Litt M, Finnegan LP. Narcotic addiction in the newborn: differences in behavior generated by methadone and heroin. Int J Clin Pharmacol 1975; 12:63–69. 41. Kron RE, Litt M, Eng D, et al. Neonatal narcotic abstinence: Effects of pharmacotherapeutic agents and maternal drug usage on nutritive sucking behavior. J Pediatr 1976; 88:637–641. 42. Finnegan LP, Michael H. An evaluation of neonatal abstinence treatment modalities. NIDA Res Monogr 1984; 49:282–288. 43. Carin I, Glass L, Parekh A, et al. Neonatal methadone withdrawal. Effect of two treatment regimens. Am J Dis Child 1983; 137:1166–1169. 44. Finnegan LP, Ehrlich SM. Maternal drug abuse during pregnancy: Evaluation and pharmacotherapy for neonatal abstinence. In: Adler MW, Cowan A, eds. Modern Methods in Pharmacology. vol 6. Testing and Evaluation of Drugs of Abuse. New York: Wiley-Liss, 1990, pp 255– 263. 45. Kandall SR, Albin S, Gartner LM, et al. The narcotic-dependent mother: fetal and neonatal consequences. Early Hum Dev 1977; 1:159–169. 46. Osol A, Hoover E, eds. Remington’s Pharmaceutical Sciences, 15th ed. Easton, Mack, 1975, p 1037.

384

Theis et al.

47. Jaffe JH, Martin WR: Opioid analgesics and antagonists. In: Gilman AG, Goodman LS, Rall TW, Murad F, eds. Goodman and Gilman’s The Pharmacological Basis of Therapeutics. 7th ed. New York: Macmillan, 1985, pp 491–531. 48. Lam SK, To WK. Neonatal narcotic withdrawal in Hong Kong Chinese. Asia-Oceania J Obstet Gynecol 1994; 20:231–236. 49. Dahlem P, Bucher HU, Ursprung T, et al. Nachweis von Drogen im Mekonium. Monatsschr Kinderheilkd 1992; 140:354–356. 50. Arcas Cruz R, Figueras Aloy J, et al. Recien nacido de madre adicta a las drogas: aspectos materno perinatologicos, neonatales y sindrome de abstinencia. An Esp Pediatr 1991; 34:123– 127. 51. Kaltenbach KA. Effects of in-utero opiate exposure: New paradigms for old questions. Drug Alcohol Depend 1994; 36:83–87. 52. Konkol RJ, Murphey LJ, Ferriero DM, et al. Cocaine metabolities in the neonate: potential for toxicity. J Child Neurol 1994; 9:242–248. 53. Fulroth R, Philips B. Perinatal outcome of infants exposed to cocaine and/or heroin in utero. Am J Dis Child 1989; 143:905–910. 54. Koren G, Klein J, Forman R, et al. Biological markers of intrauterine exposure to cocaine and cigarette smoking. Dev Pharmacol Ther 1992; 18:228–236. 55. Chasnoff IJ, Burns WJ. The Moro reaction: a scoring system for neonatal narcotic withdrawal. Dev Med Child Neurol 1984; 26:484–489. 56. Green M, Suffet F. The Neonatal Narcotic Withdrawal Index: a device for the improvement of care in the abstinence syndrome. Am J Drug Alcohol Abuse 1981; 8:203–213. 57. Lipsitz PJ. A proposed narcotic withdrawal score for use with newborn infants: a pragmatic evaluation of its efficacy. Clin Pediatr 1975; 14:592–594.

21 Fetal Effects of Cocaine: An Updated Meta-Analysis Antonio Addis Istituto di Ricerche Farmacologiche Mario Negri, Milan, Italy

Myla E. Moretti, Fayyazuddin Ahmed Syed, and Gideon Koren The Hospital for Sick Children, Toronto, Ontario, Canada

Thomas R. Einarson The University of Toronto, Toronto, Ontario, Canada

INTRODUCTION The high prevalence of cocaine use has become a major health concern during pregnancy. Cocaine crosses the human placenta with varying proportions absorbed by the placenta, suggesting that the placenta may offer a degree of protection to some fetuses after bolus administration (1). Cocaine is a CNS stimulant with effects thought to be due to its sypathomimetic-driven fetal, uterine, or maternal vasoconstriction and hypertension, leading to infarcts or hemorrhages at any time during gestation and in any structure. This may explain the variability of clinical effects attributed to cocaine use. A typical well-defined ‘‘fetal cocaine syndrome’’ has not been identified (2); however, exposure to cocaine during pregnancy has been associated with shorter gestation, premature delivery, abruptio placentae, and other maternal and neonatal adverse effects. In the past, congenital malformations of almost every system have been reported leading many clinicians to believe that the drug is teratogenic (3). Reports of fetal cocaine effects have been controversial, and it has been difficult to elucidate these effects, because interpretation of the results is hampered by the fact that cocaine use is commonly accompanied by confounding factors such as concomitant use of cigarettes and other recreational drugs, including heroin, cannabis, methadone, and others (4), all of which may affect pregnancy outcome by themselves. Our original systematic review and meta-analysis in 1991 (5) suggested that cocaine exposure during pregnancy is not a major risk for malformation (except for the genitourinary tract). When cocaine users were compared to women not consuming any drugs of abuse, a variety of risks emerged. Yet when we compared pregnancy and neonatal outcomes in children of cocaine consumers with those of polydrug consumers, the increased risk for most of the adverse outcomes was nullified. A major issue at that stage was the relatively small number of studies for each adverse endpoint. Since 1989 (the last year 385

386

Addis et al.

included in our previous meta-analysis), however, scores of studies on cocaine exposure during pregnancy have been published. Hence, the present meta-analysis has allowed us to separate the exposure data into several comparisons: (1) between all combinations of exposures that involve cocaine use during pregnancy (cocaine alone, polydrugs including cocaine, polydrugs not including cocaine) and (2) between the different methodologies used to determine exposure (urine analysis, maternal interview, chart review). The aim of this updated meta-analysis was to arrive at an overall quantitative estimate of the effect of cocaine use on pregnancy outcome and potential pregnancy complications, with a substantially increased power.

METHOD Data Sources The medical literature published between January 1989 and December 1997 was searched for papers dealing with the outcome of pregnancy following gestational cocaine exposure. The literature search was performed in the MEDLINE and EMBASE bibliographic databases. A search strategy using a combination of pregnancy or abnormalities drug induced and cocaine keywords (MeSH) were used. All references in the retrieved articles were screened for further papers. From all the references we excluded the non-English papers, comments, letters, editorials and reviews (Table 1). However, the references from these excluded publications were also scanned for other studies not quoted by the databases. All studies considered in the previous meta-analysis were retrieved and resubmitted to the inclusion or exclusion process. Study Selection Searches were reviewed or completed independently and in duplicate. Investigators were blinded as to the name of the journal, the authors of the paper, the hospital or site of the study, and the funding agency supporting the study. This blinding was accomplished by removing all identifying statements and titles prior to photocopying the papers for review

Table 1 Results of Search on Cocaine Use During Pregnancy Type of study

No.

Case reports/case series Letters Editorials Reviews and commentaries Studies not reporting fetal or pregnancy outcomes In vitro studies/placental perfusion Outcomes beyond the scope of this meta-analysis Studies without control group Cocaine users not separated from other drugs Studies included in the meta-analysis Total

80 43 22 87 127 40 54 19 12 33 517

Fetal Effects of Cocaine

387

by another investigator. Without this identification, the prestige, reputation, or standards of the journal or author cannot interfere with the selection or analysis of articles. Although, theoretically, each journal has a type style that could identify it, the junior investigator who conducted blinding of the papers was not familiar with the different styles. The criteria for inclusion of papers in this analysis were human exposure to any amount of cocaine during any or all trimesters of pregnancy as evidenced by drug history or urine test and report of outcome of pregnancy or fetal development. Only case-control or cohort studies with at least one control group were included. Use of other drugs is common among cocaine users and was not an exclusion criterion. Case reports, reviews, editorials, and letters were excluded from the analysis but were retrieved and followed up for further papers. The ‘‘methods’’ section of each paper was evaluated using a scoring sheet, prepared a priori, which listed these inclusion criteria. Studies fulfilling all inclusion criteria were accepted for analysis. Successfully included studies were read and information was extracted using a standardized data extraction form. The information collected included the size and selection of cocaine and control groups; the pregnancy and fetal outcome measurements were studied. Studies were analyzed separately according to various exposure groups, namely: Pregnant women exposed to cocaine alone compared to drug-free controls Pregnant women exposed to cocaine plus other drugs compared to drug-free controls Pregnant women exposed to cocaine alone compared to cocaine plus other drugs Pregnant women exposed to cocaine alone versus other drug additions (and no cocaine) Pregnant women exposed to cocaine plus other drugs versus other drug addictions (and no cocaine) To avoid duplication of results, after data extraction, we examined data results of included studies with a cross analysis of author groups. The outcomes considered were related to maternal and neonatal health, namely: major malformations, low birth weight, prematurity, premature rupture of membranes (PROM), abruptio placentae, meconium, gestational age, head circumference, birth weight, and birth length. Studies examining only certain subtypes of malformations or studies with unspecified addictions were excluded from the analysis.

Data Analysis All controlled studies were pooled together to calculate the relative risk (RR) and 95% confidence interval (CI95%) using a random-effect model (6). For outcomes measured in a continuous scale (i.e., head circumference, birth weight, length, gestational age), weighted mean differences were calculated, also using, in this case, the random-effect model. Chi square tests for heterogeneity were also performed for all outcomes of interest in this analysis. Sensitivity analyses were performed on the study methodology (by excluding retrospective analysis) and according to the method used to identify cocaine users (by excluding studies that use chart review or maternal interview methodologies). This was done with the aim of assessing the impact of identification bias in the study population.

388

Addis et al.

RESULTS Over 600 scientific references were considered because they deal with the human effects of cocaine used during pregnancy, with an exponential growth in number of papers published in recent years. Of these, 517 studies were retrieved and, based on a review of their methods sections, only 36 studies met the inclusion criteria for this analysis. Three studies originally included (7–9) were excluded after a more detailed analysis. This was because they included groups with multiple exposure in more than 10% of cases and it was not possible to separate the different outcomes according to single or multiple exposures. Appendix A lists all the rejected studies from this analysis and the reasons for their rejection. The following were excluded: 80 case reports, 22 editorials, 43 letters, 87 reviews or commentaries, 127 studies without fetal or pregnancy outcomes, 40 in vitro studies or placental perfusions, 54 studies with outcomes not within the scope of this meta-analysis, 19 studies without a control group, and 12 studies where cocaine users had not been separated from users of other drugs. Table 2 highlights the characteristics of the 33 included studies. Exposure was mainly ascertained through urine or meconium analysis (73%) or by chart review and/or maternal interview (27% of the studies), while outcome was mainly confirmed using physician examination or record examination. The studies reported pregnancy and neonatal outcomes of 4184 women exposed to cocaine during pregnancy. Nineteen percent of these were also exposed to others drugs (i.e., heroin, methadone, marijuana). The outcomes were compared with 31,544 drug-free pregnancy controls and with 963 women exposed to other addictions but not cocaine. Major Malformations The teratogenic potential of cocaine was investigated in 16 (47%) of included studies (Fig. 1). Major malformations were those described by Heinonen et al. (10). Comparing women where cocaine was the only drug of abuse in pregnancy with drug-free controls, the pooled relative risk resulted significant (RR: 1.70, 95% CI 1.12–2.60). Six studies reported the number of malformed children after exposure during pregnancy to cocaine plus other addictions (heroin, marijuana, methadone, etc). The pooled relative risk was also significant in this case (RR: 2.10; 95% CI 1.42–3.09). The significance disappeared when the risk for major malformations was analyzed in women exposed to cocaine alone compared to women exposed to cocaine plus other addictions. When we used women exposed to polydrugs without cocaine as control, we were not able to identify any significant risk for major malformations in the women exposed to cocaine alone or to polydrugs with cocaine. A test for heterogeneity showed that all studies detected an effect size of similar magnitude and direction with the exception of the comparison between cocaine alone versus polydrugs with cocaine exposures. In this case, a single study was responsible for the heterogeneity (11). When that study was removed, the heterogeneity test was no longer significant. Low Birth Weight Of the included studies, 18 (54%) reported data on the number newborn with low birth weight after cocaine exposure (Fig. 2). The rate of low birth weight was defined as the number of live births weighing less than 2500 g. As in the previous analysis, mothers exposed to cocaine alone or to polydrugs with cocaine show a significant pooled relative risk of low birth weight when compared to a drug-free control (RR: 2.85, 95% CI 2.28–

Fetal Effects of Cocaine

389

3.56 and 3.54, 95% CI 1.95–6.43, respectively). When we used as controls the mothers exposed to polydrugs with or without cocaine the significance disappeared. Results were homogeneous in three comparisons (cocaine alone versus polydrugs with and without cocaine and polydrugs with cocaine versus polydrugs without cocaine). The test for heterogeneity was positive in the first two comparisons with drug-free controls (cocaine alone and polydrugs with cocaine). However, in the comparison of cocaine plus polydrugs versus drug-free controls the heterogeneity was due to a single study (12). In the first comparison of cocaine alone versus drug free, the homogeneity resulted by using only prospective studies in which urine analysis was used to identify cocaine users.

Birth Weight Similar trends are shown with the data on the rate of low birth weight (Fig. 3). Use of cocaine during pregnancy (alone or with polydrugs) was associated with the risk of a weighted average birth weight reduction from 486 to 512 g. When pregnant women exposed to polydrugs with or without cocaine were used as controls, the significant birthweight reduction disappeared. Regarding the combinability of the included studies reporting birth weight after cocaine exposure, the comparisons of cocaine alone or with polydrugs versus drug-free control was homogeneous after the exclusion of outliers— Bingol (13) and Cohen (14) respectively. All other comparisons were homogeneous.

Birth Length Consistent with the data on birth weight, use of cocaine (alone or in combinations with polydrugs) during pregnancy is associated with the risk of a weighted average birth-length reduction from 2.20 to 2.65 cm (Fig. 4). Using as controls mothers exposed to polydrugs with or without cocaine, there was no difference in birth length of newborns exposed to cocaine (with or without polydrugs) during pregnancy. All comparisons were homogeneous with the exception of cocaine alone versus drug-free controls where one outlier (15) was excluded for homogeneity.

Prematurity Prematurity was defined as a live birth before 37 weeks of gestation (Fig. 5). Pregnant women exposed to cocaine alone or to cocaine plus polydrugs have a higher risk of preterm delivery (RR: 2.63 95% CI 2.13–3.26 and 3.19 95% CI 2.30–4.41, respectively) than do drug-free controls. A comparison of mothers exposed to cocaine alone versus polydrugs with cocaine or mothers with cocaine and polydrug exposure versus polydrugs without cocaine did not result in a significant risk for prematurity. When mothers exposed to cocaine alone were compared with the group of mothers exposed to polydrugs without cocaine, we noticed again a significant risk of prematurity (RR: 1.72 95% CI 1.19–2.52). All comparisons were homogeneous with the exception of the first comparison (cocaine alone versus drug-free control). This comparison became homogeneous when only prospective studies with identification of cocaine users by urine or meconium analysis were considered.

390

Table 2

Characteristics of Studies Included in the Meta-analysis

Reference 1. Bateman, 1993 (16)

2. Bingol, 1987 (13)

3. Chasnoff, 1985 (17)

4. Chasnoff, 1987 (18)

Outcomes

Study type

Data collection

Cocaine alone (361) Drug free (387)

Cohort prospective

Urine analysis Chart review

Cocaine alone (50) Drug free (340) Polydrugs with Cocaine (110) Cocaine Alone (12) Drug free (15) Polydrugs with cocaine (12) Polydrugs no cocaine (15)

Cohort retrospective

Interview with the mother Chart review

Cohort prospective

Urine analysis

Polydrugs with cocaine (52) Polydrugs no cocaine (73)

Cohort prospective

Urine analysis Chart review

Addis et al.

Birth defects Low birth weight Prematurity Abruptio placentae Gestational age Head circumference Weight Length Birth defects Head circumference Weight Length Birth defects Abruptio placentae Head circumference Weight Length Birth defects Prematurity Abruptio placentae Meconium Weight Head circumference Length

Study groups (sample size)

6. Chavez, 1989 (20) 7. Chazotte, 1991 (21)

8. Cherukuri, 1988 (22)

9. Chouteau, 1988 (23) 10. Cohen, 1991 (14)

11. Eyler, 1994 (24)

Birth defects Low birth weight Prematurity Abruptio placentae Weight Head circumference Birth defects

Cohort prospective

Urine analysis

Cocaine alone (1063) Drug free (8326) Cocaine alone (42) Drug free (42)

Case control retrospective Cohort prospective

Chart review Interview with the mother Urine analysis

Cocaine alone (55) Drug free (55)

Cohort retrospective

Interview with the mother Chart review

Cocaine alone (124) Drug free (218) Cocaine alone (56) Drug free (166) Polydrugs with cocaine (27)

Cohort retrospective Cohort retrospective

Urine analysis

Cocaine alone (168) Drug free (168)

Cohort retrospective

Chart review

Urine analysis

391

Low birth weight Gestational age Weight Birth defects Low birth weight Prematurity PROM Abruptio placentae Meconium Gestational age Head circumference Weight Length Low birth weight Prematurity Low birth weight Prematurity Gestational age Weight Birth defects Low birth weight Prematurity Abruptio placentae Gestational age Weight

Cocaine alone (53) Drug free (40)

Fetal Effects of Cocaine

5. Chasnoff, 1988 (19)

392

Table 2

Continued

Reference

Outcomes Prematurity Meconium

13. Galanter, 1992 (26)

PROMa Abruptio placentae Meconium Birth defects Low birth weight Prematurity PROM Abruptio placentae Head circumference Weight Length Birth defects Low birth weight Prematurity Abruptio placentae Meconium Head circumference Weight Length Weight

14. Gillogley, 1990 (27)

15. Hadeed, 1989 (28)

16. Isenberg, 1987 (29)

Cocaine alone (35) Drug free (1021) Polydrugs no cocaine (35) Polydrugs with cocaine (51) Polydrugs no cocaine (350)

Study type

Data collection

Cohort prospective

Urine analysis Chart review

Cohort prospective

Urine analysis

Cocaine alone (139) Drug free (293) Polydrugs with cocaine (35) Polydrugs no cocaine (125)

Cohort retrospective

Urine analysis Chart review

Cocaine alone (56) Drug free (56)

Cohort prospective

Urine analysis

Cocaine alone (13) Drug free (36)

Cohort prospective

Urine analysis

Addis et al.

12. Fulroth, 1989 (25)

Study groups (sample size)

18. Kelley, 1991 (15)

19. Kistin, 1996 (31)

20. Little, 1989 (32) 21. McCalla, 1991 (33)

22. MacGregor, 1987 (34)

23. Mastrogiannis, 1990 (35)

Low birth weight Prematurity PROM Abruptio placentae Meconium Gestational age Weight Low birth weight Prematurity Abruptio placentae Gestational age Head circumference Weight Length Birth defects Low birth weight Abruptio placenta Prematurity Birth defects Meconium PROM Abruptio placentae Low birth weight Prematurity Abruptio placenta Gestational age Weight Abruptio placentae PROM Meconium

Cocaine alone (63) Drug free (123) Polydrugs with cocaine (137) Polydrugs no cocaine (27)

Cohort retrospective

Chart review

Cocaine alone (30) Drug free (30)

Cohort retrospective

Chart review

Cocaine alone (64) Drug free (13,043)

Cohort retrospective

Urine analysis Chart review

Polydrugs with cocaine (53) Drug free (100) Cocaine alone (128) Drug free (983)

Cohort prospective Crosssectional study Cohort prospective

Interview with the mother Chart review Urine analysis

Cohort retrospective

Urine analysis

Cocaine alone (24) Drug free (70) Polydrugs with cocaine (46)

Cocaine alone (40) Drug free (46)

Fetal Effects of Cocaine

17. Keith, 1989 (30)

Chart review

393

394

Table 2

Continued

Reference 24. Matera, 1990 (36) 25. Nair, 1994 (37)

26. Neerhof, 1989 (11)

PROM Abruptio placentae Low birth weight Meconium 14 Weight Length Birth defects Prematurity Abruptio placentae Gestational age Head circumference Weight Low birth weight Low birth weight Prematurity Abruptio placentae Head circumference Weight Length

Study groups (sample size)

Study type

Data collection

Cocaine alone (51) Drug free (350) Polydrugs with cocaine (55) Polydrugs no cocaine (86)

Cohort prospective Cohort prospective

Urine analysis

Cocaine alone (113) Drug free (88) Polydrugs with cocaine (24)

Cohort prospective

Urine analysis

Polydrugs with cocaine (34) Drug free (590) Cocaine alone (21) Drug free (600)

Cohort prospective Cohort prospective

Interview with the mother

Meconium analysis

Meconium analysis

Addis et al.

27. Richardson, 1991 (12) 28. Rosengren, 1993 (38)

Outcomes

30. Singer, 1994 (40)

31. Sprauve, 1997 (41)

32. Stafford, 1994 (42) 33. Zuckerman, 1989 (43)

a

Prematurity Gestational age Head circumference Weight Length Low birth weight Prematurity Gestational age Weight Length Birth defect Low birth weight Prematurity Abruptio placentae Birth defects Birth defects Gestational age Head circumference Weight Length

Drug free (50) Polydrugs with cocaine (50) Polydrugs no cocaine (50)

Cohort prospective

Urine analysis Chart review

Cocaine alone (100) Drug free (100)

Cohort retrospective

Urine analysis Interview with the mother

Cocaine alone (483) Drug free (3158)

Cohort retrospective

Urine analysis

Cocaine alone (40) Drug free (40) Drug free (1010) Polydrugs with cocaine (114) Polydrugs no cocaine (202)

Cohort prospective Cohort prospective

Urine analysis

Fetal Effects of Cocaine

29. Ryan, 1987 (39)

Urine analysis Chart review

PROM ⫽ premature rupture of membranes.

395

396

Figure 1 Major malformations.

Addis et al.

Fetal Effects of Cocaine

Figure 2 Low birth weight.

397

Figure 3 Birth weight.

Fetal Effects of Cocaine

399

Figure 4 Birth length.

Gestational Age The analysis of a continuous variable like gestational age confirmed the findings on prematurity after cocaine exposure during pregnancy (Fig. 6). There was a significant reduction of the gestational age when we compare the mothers exposed to cocaine (alone or with other drugs) versus drug free control groups. Mothers using cocaine alone did not have a significant reduction of gestational age when compared with women exposed to polydrugs with cocaine. Similar results were found for polydrugs with cocaine exposure versus

Figure 5 Prematurity.

Fetal Effects of Cocaine

Figure 6 Gestational age.

401

402

Addis et al.

polydrugs without cocaine exposure. Of the included studies, only one reported the data on gestational age comparing women exposed to cocaine alone versus polydrugs without cocaine. After the exclusion of one outlier in the comparison of cocaine with polydrugs versus drug-free control, all comparisons were homogeneous.

Abruptio placentae Of the included studies, 19 reported data on abruptio placentae after cocaine exposure (Fig. 7). The risk for this pregnancy complication was significant when we considered the studies that compared mothers exposed to cocaine alone or in combination with other addictions versus drug-free controls or versus exposure to polydrugs but not cocaine. No risk of abruptio placentae was noted when we considered mothers exposed to cocaine alone versus women exposed to cocaine plus other drugs. The pooled risk for abruptio placentae was higher in studies comparing cocaine alone or with polydrugs versus polydrugs without cocaine. All comparisons were homogeneous.

Premature Rupture of Membranes (PROM) This analysis included 7 studies (Fig. 8). The pooled relative risk for PROM was significantly higher for pregnant women exposed to cocaine alone or with polydrugs compared to drug-free controls or to polydrugs but not cocaine. The only comparison not found to be significant was ‘‘cocaine alone versus polydrugs with cocaine.’’ In all comparisons, the test for heterogeneity showed that all studies detected an effect size of similar magnitude and direction.

Head Circumference Thirteen studies reported measurements of head circumference in children exposed to cocaine during pregnancy (Fig. 9). Cocaine use (alone or with polydrugs) during pregnancy was associated with a significant risk of reduction of head circumference in the newborns. Cocaine alone or with other addictions did not lead to a significant reduction of head circumference when compared to polydrug (but not cocaine) exposure. All these comparisons were found to be homogeneous. Only two studies had data eligible for the comparison between exposure to cocaine alone versus cocaine with polydrugs, which showed no significant reduction of head circumference. Finally, only one study reported data comparing head circumference after cocaine alone versus polydrugs with no cocaine. No significant risk was detected in this last comparison.

Sensitivity Analysis Both retrospective analyses and studies identifying exposure by maternal interview or chart extraction may result in bias as to the correct exposure type. For this reason the analysis was repeated, excluding, where possible, all retrospective studies and analyses

Figure 7 Abruptio placentae.

404

Addis et al.

Figure 8 Premature rupture of membranes.

where the identification of cocaine exposure was done by chart extraction or maternal interview. After the exclusions of those studies, the risk of major malformation was still significantly higher in mothers exposed to cocaine than in drug-free controls (RR: 1.09, 95% CI 1.09–4.93) but nonsignificant in the comparison of pregnancy with cocaine alone versus polydrugs without cocaine exposures. Using the same exclusion criteria, we recalculated the relative risk or weighted average of low birth weight (RR: 2.82, 95% CI 1.72–

Fetal Effects of Cocaine

405

Figure 9 Head circumference.

4.64), birth weight (weighted average: ⫺485g 95% CI ⫺550.8 to ⫺379.9 g), birth length (weighted average: ⫺2.18 cm 95% CI ⫺2.84 to ⫺1.52 cm), prematurity (RR: 2.44 95% CI 1.90–3.12), gestational age (weighted average: ⫺1.24 weeks, 95% CI ⫺1.61 to ⫺0.8 weeks), abruptio placentae (RR: 5.89 95% CI 1.90–3.12), PROM (RR: 1.58 95% CI 1– 2.51), and head circumference (weighted average ⫺1.7 cm 95% CI ⫺2.0 to ⫺1.35 cm) in women exposed to cocaine alone versus drug-free control. In none of these comparisons with cocaine versus drug-free controls did the significance disappear.

406

Addis et al.

DISCUSSION Because very large numbers of women in the reproductive age group consume cocaine, there are serious concerns regarding the long-term health of millions of children exposed in utero to this recreational drug (2). An issue widely recognized by researchers and clinicians is that, in addition to cocaine, many other risk factors occur and may adversely affect infants exposed in utero to cocaine. These include other drugs of abuse and cigarettes, alcohol, poor prenatal care, single motherhood, and poverty, to mention a few (2,4). Any attempt to separate the effects of cocaine from those of other drugs of abuse must therefore address the role of the various confounders. In our original meta-analysis, performed a decade ago (5), there was often insufficient statistical power to address differences between cocaine-exposed fetuses and controls. Because of the tremendous volume of research published during the last 10 years, we had the opportunity to address several important adverse effects that are widely attributed to cocaine. By separating studies into those where women were exposed mainly to cocaine, to cocaine plus other drugs, mostly to other drugs, or no drugs (who were drug-free), we have created a population-based control for many of the confounding factors mentioned above that are known to affect infants adversely. The rates of major malformations among children of cocaine users are significantly higher than among drug-free women. However, when these babies are compared to those exposed to other drugs of abuse, the effect size is nullified, strongly suggesting that the various confounders, also occurring in polydrug/no cocaine users, are responsible for this effect. An identical trend has emerged for other endpoints, including gestational age, birth weight, birth length, and head circumference as well as the rates of prematurity and low birth weight. The two adverse effects measured by this systematic review that are significantly associated with cocaine itself are the rates of abruptio placentae and premature rupture of membranes. In both cases, cocaine users had higher risks for these adverse events than women consuming other drugs of abuse. These results are different from those in our original analysis (5), where there was insufficient power to prove placental abruption as directly associated with cocaine. It is important to recognize that this analysis did not include reports of neurobehavioral endpoints, which are currently the focus of much research and clinical attention. A separate systematic review will be needed to address long-term brain development in these children.

Appendix A List of Rejected Studies and Reason for Rejection Case Reports/Case Series (80) Abramowicz, (1991)(3); Apple, (1990)(27); Bakht, (1990)(32); Beltran, (1995)(41); Brown, Jr., (1993)(50); Burkett, (1990)(57); Carlan, (1991)(63); Cohen, (1991)(91); Collins, (1989)(95); Deoliveira, (1991)(110); Dollberg, (1989)(121); Dominguez, (1991)(122); Evelyn, (1992)(138); Geggel, (1989)(160); George, (1995)(161); GieronKorthals, (1994)(163); Gomez-Anson, (1994)(171); Gonsoulin, (1990)(173); Good, (1992)(175); Greenfield, (1991)(183); Greenland, (1989)(184); Hall, (1992)(188); Hannig,

Fetal Effects of Cocaine

407

(1991)(192); Harris, (1993)(196); Hepper, (1995)(205); Ho, (1994)(208); Hoeger, (1996)(209); Hoyme, (1990)(216); Hsu, (1992)(217); Iriye, (1994)(230); Iriye, (1995)(229); Jack, (1990)(232); Jasnosz, (1994)(238); Jawahar, (1997)(239); Johnson, (1995)(245); Kankirawatana, (1993)(254); Kapur, (1991)(255); Lampley (1996)(282); Lessick, (1991)(289); Lezcano, (1994)(293); Liu, (1992)(302); Martinez, (1994)(315); Meeker, (1990)(324); Mendelson, (1992)(325); Mercado, (1989)(326); Miller, (1990)(329); Mishra, (1995)(338); Mittleman, (1989b)(340); Moen, (1993)(341); Moriya, (1995)(348); Murphy, (1993)(352); Nolte, (1991)(367); Nyirjesy, (1993)(371); Okoruwa, (1995)(373); Ostrea, Jr., (1994)(375); Perlow, (1990)(380); Peters, (1992)(381); Porat, (1991)(387); Potter, (1994)(388); Rais-Bahrami, (1990)(394); Reznik, (1989)(400); Robin, (1994)(410); Sarpong, (1992)(427); Seballos, (1994)(432); Sheinbaum, (1992)(435); Sims, (1989)(442); Singer, (1995)(444); Skopp, (1997)(448); Spinazzola, (1992)(458); Spires, (1989)(459); Spital, (1991)(460); Streissguth, (1991)(468); Sturner, (1991)(469); Sumner, (1993)(471); Sztulman, (1990)(476); Thatcher, (1989)(478); Towers, (1993)(480); van den Anker, (1993)(489); Viscarello, (1992)(491); Wilson, (1993)(504). Editorials (22) Alexander, (1991)(13); Amoury, (1993)(15); Anonymous, (1990a)(20); Anonymous, (1991a)(23); Baxley, (1990)(36); Brouhard, (1994)(49); Chasnoff, (1989a)(71); Fantel, (1990)(141); Heagarty, (1990)(200); Howard, (1995)(214); Jentzen, (1993)(240); Jos, (1995)(247); Kearney, (1995)(259); Konkol, (1994a)(273); Lewis, (1989)(292); Lone, (1991)(303); Neuspiel, (1993b)(360); Nora, (1990)(368); Plessinger, (1993)(384); Salamy, (1990)(424); Tudehope, (1989)(482); Zuckerman, (1994)(515). Letters (43) Abdeljaber, (1990)(1); Ahmed, (1989)(7); Anonymous, (1990)(17); Bays, (1991)(37); Carraccio, (1994)(64); Chavkin, (1990)(80); Cottler, (1992)(102); DiGregorio, (1993)(116); Downing, (1991)(125); Fogarty, (1991)(146); Friedman, (1995)(153); Goldin, (1989)(169); Good, (1994)(174); Graham, (1989a)(178); Gratacos, (1993)(182); Hansen, (1990)(193); Herschman, (1991)(207); Hunt, (1990)(220); Jerome, (1995)(241); Kain, (1992)(249); Levy, (1993)(290); Maynard, (1990)(319); Miele, (1992)(328); Mittleman, (1989a)(339); Moore, (1993)(346); Morfesis, (1994)(347); Neuspiel, (1992a)(358); Neuspiel, (1992b)(362), Neuspiel, (1995)(361); Nucci, (1994)(369); Page, (1992)(377); Petrulis, (1992)(382); Racine, (1994)(393); Sackoff, (1992)(423); Shaw, (1991)(434); Sher, (1989)(437); Stephens, (1989)(464); Sugarman, (1995)(470);Tabor, (1990)(477); Thorp, Jr. (1991)(479); van den Anker, (1991)(487); van den Anker, (1992)(488);Yang, (1995)(508). Reviews and Commentaries (87) Abel, (1991)(2); Adams, (1990)(5); Anonymous, (1989a)(18); Anonymous, (1993)(25); Aronson, (1990)(29); Berger, (1990)(44); Brent, (1994)(48); Buehler, (1995)(54); Buehler, (1996)(55); Byrne, (1992)(60); Chan, (1997)(68); Chao, (1996)(69); Chasnoff, (1989b)(72); Chasnoff, (1993)(73); Chavkin, (1993)(81); Church, (1993)(88); Coles, (1992)(92); Coles, (1993)(93); Cornish, (1996)(101); Day, (1993a)(106); Day, (1993b)(107); Dixon, (1989)(118); Dow Edwards, (1991)(124); Dow-Edwards, (1993)(123); Dumas, (1992)(127); Dungy-Poythress, (1995)(128); Elhassani, (1990)(132);

408

Addis et al.

Ellen, (1993)(135); Fantel, (1993)(140); Ferner, (1993)(143); Finnegan, (1994)(145); Frank, (1993)(152); Garcia, (1990)(158); Gaskill, (1992)(159); Giacoia, (1990)(162); Gintautiene, (1990)(168); Goldstein, (1990)(170); Goodwin, (1990)(176); Grossman, (1993)(185); Horger, (1990)(211); Janke, (1990)(237); Johnson, (1992)(242); Kaltenbach, (1994)(251); Kandall, (1991)(252); Karmel, (1991)(257); Keller, (1993)(261); Kenner, (1997)(263); Khoury, (1991)(264); Klitsch, (1994)(270); Knisely, (1991)(272); Koren, (1989)(278); Koren, (1993)(275); Landry, (1996)(284); Lauder, (1991)(286); Lewis, (1991)(291); Litt, (1997)(299); Lunsford, (1995)(306); Lutiger, (1991)(307); Martin, (1992a)(312); Martin, (1992b)(313); Meyer, (1996)(327); Miller, (1994)(333); Miller, (1997)(330); Murphey, (1993)(351); Neuspiel, (1993a)(359); Ostrea, (1992)(374); Pokorni, (1996)(385); Richardson, (1993)(405); Richardson, (1996)(406); Ripple, (1992)(408); Rizk, (1996)(409); Sexson, (1993)(433); Shepard, (1993)(436); Slade, (1993)(449); Spear, (1993)(456); Sprauve, (1996)(461); Strauss, (1997)(467); Sun. (1997)(472); Swadi, (1993)(474); Szeto, (1991)(475); Warner, (1997)(494); Webster, (1993)(497); Westdorp, (1995)(502); Wightman, (1991)(503); Wootton, (1994)(507); Zuckerman, (1992)(514); Zuckerman, (1991)(512). Studies not reporting fetal or pregnancy outcomes (127) Abusada, (1993)(4); Alemi, (1996)(10); Alemi, (1996)(11); Alemi, (1996)(12); Amaro, (1990)(14); Anonymous, (1990c)(22); Anonymous, (1996)(26); Archie, (1997)(28); Ball, (1997)(33); Bauchner, (1990)(35); Behnke, (1997)(39); Bendersky, (1996)(42); Berenson, (1991)(43); Billman, (1996)(46); Browne, (1992)(52); Browne, (1994)(51); Brunader, (1991)(53); Burke, (1993)(56); Callahan, (1992)(62); Cartwright, (1991)(65); Casanova, (1994)(66); Chasnoff, (1990)(78); Charpentier, (1996)(70); Condie, (1989)(96); Cornelius, (1993)(99);Cornelius, (1994)(100); Czeizel, (1990)(105); de Feo, (1995)(108); DelaneyBlack, (1996)(109); DePetrillo, (1995)(111); DeVane, (1991)(113); DiGregorio, (1994)(117); Dudish, (1996)(126); Egelko, (1996)(131); Elk, (1994)(133); Elk, (1995)(134); Ellerbrock, (1995)(136); Emery, (1995)(137); Fetters, (1996)(144); Forman, (1992)(149); Forman, (1994)(147); Frank, (1992)(151); Galanter, (1992)(157); Gillmore, (1992)(165); Graham, (1989b)(180); Graham, (1991)(179); Grant, (1994)(181); Gomez, (1996)(172); Hall, (1996)(187); Hawley, (1995)(198); Heller, (1996)(202); Henderson, (1989)(203); Henderson, (1997)(204); Hurd, (1991)(222); Hurt, (1996)(223); Hurt, (1997a)(225); Hurt, (1997b)(226); Hutchins, (1997)(227); Ingersoll, (1996)(228); Jacobson, (1991)(234); Jacobson, (1996)(235); Jain, (1993)(236); Johnson, (1997)(244); Joyce, (1995)(248); Kain, (1995)(250); Karmel, (1996)(256); Killeen, (1995)(265); Klein, (1992)(268); Klein, (1994)(267); Kline, (1997)(269); Konkol, (1994b)(274); Koren, (1992a)(276); Koren, (1992b)(277); Koren, (1992c)(279); Land, (1990)(283); Lindsay, (1990)(297); Lanehart, (1994)(285); Lauria, (1997)(287); Lindsay, (1991)(295); Lindsay, (1992)(296); Lindsay, (1997)(294); Little, (1990)(300); Marques, (1993)(310); Martin, (1996)(311); Martinez Crespo, (1994)(314); Mayes, (1996)(318); McCalla, (1992)(322); McFarlin, (1996)(323); Miller, (1993)(334); Minkoff, (1990)(335); Moore, (1996)(345); Moriya, (1994)(349); Moser, (1993)(350); Napiorkowski, (1996)(355); Neuspiel, (1993c)(365); O’Connor, (1997)(372); Peeke, (1994)(378); Pelham, (1992)(379); Phibbs, (1991)(383); Polzin, (1991)(386); Quinn, (1992)(391); Reddin, (1991)(398); Regalado, (1996)(399); Richardson, (1996a)(406); Richardson, (1996b)(402); Rico, (1990)(407); Rodriguez, (1996)(414); Rogers, (1991)(416); Rosenkranz, (1990)(418); Ruiz, (1994)(420); Ryan, (1994)(422); Scafidi, (1996)(428); Schutzman, (1991)(431); Singer, (1997)(443); Smith, (1989)(453); Strano-Rossi, (1996)(466); Svikis, (1997)(473); Tron-

Fetal Effects of Cocaine

409

ick, (1996)(481); Ursitti, (1997)(483); Vega, (1997)(490); Wasserman, (1993)(495); Weeman, (1995)(498); Weese-Mayer, (1991)(499); Welch, (1993)(501); Wingert, (1994)(506); Zimmerman, (1994)(511); Zuckerman, (1989a)(513). In vitro studies/placenta perfusion (40) Ahmed, (1990)(9); Ahmed, (1991)(8); Bailey, (1994)(30); Bailey, (1996)(31); Cejtin, (1990)(67); Chiarotti, (1996)(85); Cook, (1996)(97); Dicke, (1993)(114); Dicke, (1994)(115); Gilbert, (1990)(164); Harker, (1994)(194); Hurd, (1993)(221); Johnson, (1996)(246); Krishna, (1993)(281); Malek, (1995)(309); Miller, (1996)(332); Mirochnick, (1997)(337); Monga, (1993)(343); Monga, (1994)(342); Moore, (1992)(344); Oyler, (1996)(376); Prasad, (1994)(389); Ramamoorthy, (1993a)(396); Ramamoorthy, (1993b)(397); Ramamoorthy, (1995)(395); Richards, (1990)(401); Roby, (1996)(411); Roe, (1990)(415); Saraf, (1995)(426); Schenker, (1993)(429); Simone, (1994b)(441); Simone, (1996)(438); Simone, (1997)(439); Smith, (1992)(452); Smith, (1995)(454); Sosnoff, (1996)(455); Sternfeld, (1997)(465); Wang, (1996)(493); Winecker, (1997)(505); Yelian, (1994)(510). Outcomes beyond the scope of this meta-analysis (54) Ahluwalia, (1992)(6); Anonymous, (1989b)(16); Anonymous, (1990b)(21); Anonymous, (1990d)(19); Anonymous, (1991b)(24); Bhushan, (1994)(45); Burkhead, (1995)(59); Calhoun, (1991)(61); Chen, (1991)(83); Chiu, (1990)(86); Cohen, (1991)(90); Cohen, (1994)(89); Coles, (1992)(94); Cordero, (1990)(98); Cotton, (1994)(103); Curry, (1995)(104); Dixon, (1989)(119); Durand, (1990)(129); Dusick, (1993)(130); Johnson, (1992)(243); Frank, (1990)(150); Fritz, (1993)(155); Gingras, (1994)(167); GottbrathFlaherty, (1995)(177); Handler, (1991)(189); Handler, (1994)(190); Hanlon-Lundberg, (1996)(191); Harris, (1992)(195); Harsham, (1994)(197); Hoskins, (1991)(213); Hume, Jr., (1994)(218); Hurt, (1995);(224); Jacobson, (1994)(233); Kandall, (1993)(253); Legido, (1992)(288); Lopez, (1995)(304); Lounsbury, (1989)(305); Miller, Jr., (1995)(331); Mirochnick, (1995)(336); Needlman, (1993)(356); Neuspiel, (1991)(363); Neuspiel, (1994)(364); Nulman, (1994)(370); Prichep, (1995)(390); Richardson, (1991)(403); Richardson, (1994)(404); Rodning, (1989)(412); Rodriguez, (1993)(413); Schneider, (1992)(430); Skolnick, (1990)(447); Singer, (1994)(446); van de Bor, (1990)(486); Wehbeh, (1995)(500); Yawn, (1994)(509). Studies without control group (19) Beeram, (1993)(38);Beltran, (1994)(40); Burkett, (1994)(58); Chasnoff, Fries, (1993)(154); Hofkosh, (1995)(210); Horn, (1992)(212); Howard, Hume, Jr., (1989)(219); Knight, (1994)(271); Kramer, (1990)(280); Link, McCalla, (1995)(320); Ney, (1990)(366); Racine, (1993)(392); Rosenstein, Slutsker, (1993)(450); Smit, (1994)(451); Weathers, (1993)(496).

(1989c)(77); (1995)(215); (1991)(298); (1990)(419);

Cocaine users not separated from other drugs (12) Doberczak, (1991)(120); Feldman, (1992)(142); Forman, (1993)(148); Handler, (1991)(189); Hawthorne, (1993)(199); Hernandez, (1992)(206); Kaye, (1989)(258); Nanda, (1990)(354); Samuels, (1993)(425); Spence, (1991)(457); van Baar, (1994)(485); Van Baar, (1990)(484).

410

Addis et al.

REFERENCES 1. Simone C, Derewlany LO, Koren G. Cocaine transfer across a cocaine exposed placental cotyledon perfused in vitro. Life Sci 1995; 57:PL137–PL140. 2. Volpe JJ. Effect of cocaine use on the fetus. N Engl J Med 1992; 327:399–407. 3. Koren G. Cocaine and the human fetus: the concept of teratophilia (comment). Neurotoxicol Teratol 1993; 15:301–304; discussion 311–312. 4. Graham K, Koren G. Characteristics of pregnant women exposed to cocaine in Toronto between 1985 and 1990. Can Med Assoc J 1991; 144:563–568. 5. Lutiger B, Graham K, Einarson TR, Koren G. Relationship between gestational cocaine use and pregnancy outcome: a meta-analysis. Teratology 1991; 44:405–414. 6. DerSimonion R, Laird N. Meta-analysis in clinical trials. Contr Clin Trials 1986; 7:177– 188. 7. Forman R, Klein J, Meta D, et al. Maternal and neonatal characteristics following exposure to cocaine in Toronto. Reprod Toxicol 1993; 7:619–622. 8. Handler A, Kistin N, Davis F, Ferre C. Cocaine use during pregnancy: perinatal outcomes. Am J Epidemiol 1991; 133:818–825. 9. Spence MR, Williams R, DiGregorio GJ, et al. The relationship between recent cocaine use and pregnancy outcome. Obstet Gynecol 1991; 78:326–329. 10. Heinonen OP, Slone D, Shapiro S. Birth Defects and Drugs in Pregnancy. Littleton, MA: PSG Publishing, 1977. 11. Neerhof MG, MacGregor SN, Retzky SS, Sullivan TP. Cocaine abuse during pregnancy: peripartum prevalence and perinatal outcome. Am J Obstet Gynecol 1989; 161:633–638. 12. Richardson GA, Day NL. Maternal and neonatal effects of moderate cocaine use during pregnancy. Neurotoxicol Teratol 1991; 13:455–460. 13. Bingol N, Fuchs M, Diaz V, et al. Teratogenicity of cocaine in humans [published erratum appears in J Pediatr 1987; 110(3):350]. J Pediatr 1987; 110:93–96. 14. Cohen HR, Green JR, Crombleholme WR. Peripartum cocaine use: estimating risk of adverse pregnancy outcome. Int J Gynaecol Obstet 1991; 35:51–54. 15. Kelley SJ, Walsh JH, Thompson K. Birth outcomes, health problems, and neglect with prenatal exposure to cocaine. Pediatric Nurs 1991; 17:130–136. 16. Bateman DA, Ng SK, Hansen CA, Heagarty MC. The effects of intrauterine cocaine exposure in newborns. Am J Public Health 1993; 83:190–193. 17. Chasnoff IJ, Burns W, Schnoll SH, Burns KA. Cocaine use in pregnancy. N Engl J Med 1985; 313:666–669. 18. Chasnoff IJ, Burns KA, Burns W. Cocaine use in pregnancy; perinatal morbidity and mortality. Neurotoxicol Teratol 1987; 161:291–293. 19. Chasnoff IJ, Chisum GM, Kaplan WE. Maternal cocaine use and genitourinary tract malformation. Teratology 1988; 37:201–204. 20. Chavez GF, Mulinare J, Cordero JF. Maternal cocaine use during early pregnancy as a risk factor for congenital urogenital anomalies. JAMA 1989; 262:795–798. 21. Chazotte C, Forman L, Gandhi J. Heart rate patterns in fetuses exposed to cocaine. Obstet Gynecol 1991; 78:323–325. 22. Cherukuri R, Minkoff H, Feldman J, et al. A cohort study of alkaloidal cocaine (‘‘crack’’) in pregnancy. Obstet Gynecol 1988; 72:147–151. 23. Chouteau M, Namerow PB, Leppert P. The effects of cocaine abuse on birth weight and gestational age. Obstet Gynecol 1988; 72:351–354. 24. Eyler FD, Behnke M, Conlon M, et al. Prenatal cocaine use: a comparison of neonates matched on maternal risk factors. Neurotoxicol Teratol 1994; 16:81–87. 25. Fulroth R, Phillips B, Durand DJ. Perinatal outcome of infants exposed to cocaine and/or heroin in utero. Am J Dis Child 1989; 143:905–910. 26. Galanter M, Egelko S, De Leon G, et al. Crack/cocaine abusers in the general hospital: assessment and initiation of care. Am J Psychiatry 1992; 149:810–815.

Fetal Effects of Cocaine

411

27. Gillogley KM, Evans AT, Hansen RL, et al. The perinatal impact of cocaine, amphetamine, and opiate use detected by universal intrapartum screening. Am J Obstet Gynecol 1990; 163: 1535–1542. 28. Hadeed AJ, Siegel SR. Maternal cocaine use during pregnancy: effect on the newborn infant (see comments). Pediatrics 1989; 84:205–210. 29. Isenberg S, Spierer A, Inkelis S. Ocular signs of cocaine intoxication in neonates. Am J Ophthalmol 1987; 103:211–214. 30. Keith LG, MacGregor S, Friedell S, et al. Substance abuse in pregnant women: recent experience at the Perinatal Center for Chemical Dependence of Northwestern Memorial Hospital. Obstet Gynecol 1999; 73:715–720. 31. Kistin M, Handler A, Davis F, Ferre C. Cocaine and cigarettes: a comparison of risks. Paediatr Perinat Epidemiol 1996; 10:269–278. 32. Little BB, Snell LM, Klein VR, Gilstrap LCI. Cocaine abuse during pregnancy: maternal and fetal implications. Obstet Gynecol 1989; 73:157–160. 33. McCalla S, Minkoff HL, Feldman J, et al. The biologic and social consequences of perinatal cocaine use in an inner-city population: results of an anonymous cross-sectional study. Am J Obstet Gynecol 1991; 164:625–630. 34. MacGregor S, Keith LG, Chasnoff IJ, et al. Cocaine use during pregnancy: adverse perinatal outcome. Am J Obstet Gynecol 1987; 157:686–690. 35. Mastrogiannis DS, Decavalas GO, Verma U, Tejani N. Perinatal outcome after recent cocaine usage. Obstet Gynecol 1990; 76:8–11. 36. Matera C, Warren WB, Moomjy M, et al. Prevalence of use of cocaine and other substances in an obstetric population. Am J Obstet Gynecol 1990; 163:797–801. 37. Nair P, Rothblum S, Hebel R. Neonatal outcome in infants with evidence of fetal exposure to opiates, cocaine, and cannabinoids. Clin Pediatr 1994; 33:280–285. 38. Rosengren SS, Longobucco DB, Bernstein BA, et al. Meconium testing for cocaine metabolite: prevalence, perceptions, and pitfalls. Am J Obstet Gynecol 1993: 168:1449–1456. 39. Ryan L, Ehrlich S, Finnegan L. Cocaine abuse in pregnancy. Neurotoxicol Teratol 1987; 9: 295–299. 40. Singer L, Arendt R, Song LY, et al. Direct and indirect interactions of cocaine with childbirth outcomes. Arch Pediatr Adolesc Med 1994; 148:959–964. 41. Sprauve ME, Lindsay MK, Herbert S, Graves W. Adverse perinatal outcome in parturients who use crack cocaine. Obstet Gynecol 1997; 89:674–678. 42. Stafford JR Jr, Rosen TS, Zaider M, Merriam JC. Prenatal cocaine exposure and the development of the human eye (see comments). Ophthalmology 1994; 101:301–308. 43. Zuckerman B, Frank DA, Hingson R, et al. Effects of maternal marijuana and cocaine use on fetal growth (see comments). N Engl J Med 1989; 320:762–768. 44. Abramowicz JS, Sherer DM, Woods JR Jr. Acute transient thrombocytopenia associated with cocaine abuse in pregnancy. Obstet Gynecol 1991; 78:499–501. 45. Apple FS, Roe SJ. Cocaine-associated fetal death in utero. J Anal Toxicol 1990; 14;259– 260. 46. Bakht FR, Kirshon B, Baker T, Cotton DB. Postpartum cardiovascular complications after bromocriptine and cocaine use. Am J Obstet Gynecol 1990; 162:1065–1066. 47. Beltran RS, Coker SB. Transient dystonia of infancy, a result of intrauterine cocaine exposure? Pediatr Neurol 1995; 12:354–356. 48. Brown RE Jr, Galford R. Postoperative ischemia after cocaine ingestion. J Arkansas Med Soc 1993; 89:589–594. 49. Burkett G, Bandstra ES, Cohen J, et al. Cocaine-related maternal death. Am J Obstet Gynecol 1990; 163:40–41. 50. Carlan SJ, Stromquist C, Angel JL, et al. Cocaine and indomethacin: fetal anuria, neonatal edema, and gastrointestinal bleeding. Obstet Gynecol 1991; 78:501–503. 51. Cohen LS, Sabbagha RE, Keith LG, Chasnoff IJ. Doppler umbilical velocimetry in women with polydrug abuse including cocaine. Int J Gynaecol Obstet 1991; 36:287–290.

412

Addis et al.

52. Collins E, Hardwick RJ, Jeffery H. Perinatal cocaine intoxication. Med J Aust 1989; 150: 331–332, 334. 53. Deoliveira IJ, Cratty BJ. Survey of ten infants exposed prenatally to maternal cocaine use. Int J Rehab Res 1991; 14:265–266. 54. Dollberg S, Armon Y, Gur I, et al. Hyponatremia in a neonate of a cocaine abusing mother. J Toxicol Clin Toxicol 1989; 27:287–292. 55. Dominguez R, Aguirre Vila-Coro A, et al. Brain and ocular abnormalities in infants with in utero exposure to cocaine and other street drugs (see comments). Am J Dis Child 1991; 145: 688–695. 56. Evelyn AC, Fine PM. Effects of AIDS, cocaine, and family violence on children in out of home care. Nebraska Med J 1992; 77:245–252. 57. Geggel RL, McInerny J, Estes NA. Transient neonatal ventricular tachycardia associated with maternal cocaine use. Am J Cardiol 1989; 63:383–384. 58. George K, Smith JF, Curet LB. Doppler velocimetry and fetal heart rate pattern observations in acute cocaine intoxication: a case report. J Reprod Med 1995; 40:65–67. 59. Gieron-Korthals MA, Helal A, Martinez CR. Expanding spectrum of cocaine induced central nervous system malformations. Brain Dev 1994; 16:253–256. 60. Gomez-Anson B, Ramsey RG. Pachygyria in a neonate with prenatal cocaine exposure: MR features. J Comput Assist Tomogr 1994; 18:637–639. 61. Gonsoulin W, Borge D, Moise KJ Jr. Rupture of unscarred uterus in primigravid woman in association with cocaine abuse. Am J Obstet Gynecol 1990; 163:526–527. 62. Good WV, Ferriero DM, Golabi M, Kobori JA. Abnormalities of the visual system in infants exposed to cocaine. Ophthalmology 1992; 99:341–346. 63. Greenfield SP, Rutigliano E, Steinhardt G, Elder JS. Genitourinary tract malformations and maternal cocaine abuse. Urology 1991; 37:455–459. 64. Greenland VC, Delke I, Minkoff HL. Vaginally administered cocaine overdose in a pregnant woman. Obstet Gynecol 1989; 74:476–477. 65. Hall TR, Zaninovic A, Lewin D, et al. Neonatal intestinal ischemnia with bowel perforation: an in utero complication of maternal cocaine abuse. AJR 1992; 158:1303–1304. 66. Hannig VL, Phillips JA. Maternal cocaine abuse and fetal anomalies: evidence for teratogenic effects of cocaine. South Med J 1991; 84:498–499. 67. Harris SR, Osborn JA, Weinberg J, et al. Effects of prenatal alcohol exposure on neuromotor and cognitive development during early childhood: a series of case reports. Phys Ther 1993; 73:608–617. 68. Hepper PG. Human fetal behaviour and maternal cocaine use: a longitudinal study. Neurotoxicology 1995; 16:139–143. 69. Ho J, Afshani E, Stapleton FB. Renal vascular abnormalities associated with prenatal cocaine exposure. Clin Pediatr 1994; 33:155–156. 70. Hoeger PH, Haupt G, Hoelzle E. Acute multifocal skin necrosis: synergism between invasive streptococcal infection and cocaine-induced tissue ischaemia? Acta Derm Venereol 1996; 76:239–241. 71. Hoyme HE, Jones KL, Dixon SD, et al. Prenatal cocaine exposure and fetal vascular disruption. Pediatrics 1990; 85:743–747. 72. Hsu CD, Chen S, Feng TI, Johnson TR. Rupture of uterine scar with extensive maternal bladder laceration after cocaine abuse. Am J Obstet Gynecol 1992; 167:129–130. 73. Iriye BK, Bristow RE, Hsu CD, et al. Uterine rupture associated with recent antepartum cocaine abuse. Obstet Gynecol 1994; 83:840–841. 74. Iriye BK, Asrat T, Adashek JA, Carr MH. Intraventricular haemorrhage and maternal brain death associated with antepartum cocaine abuse. Br J Obstet Gynaecol 1995; 102:68–69. 75. Jack BW, Davis S, Culpepper L, Hunt VR. Cocaine abuse in maternal-child health care (clinical conference). J Fam Pract 1990; 31:477–478, 481–482, 485–488. 76. Jasnosz KM, Hermansen MC, Snider C, Sang K. Congenital complete absence (bilateral

Fetal Effects of Cocaine

77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91.

92. 93.

94. 95. 96.

97. 98.

99.

413

agenesis) of the diaphragm: a rare variant of congenital diaphragmatic hernia. Am J Perinatol 1994; 11:340–343. Jawahar D, Leo PJ, Anandarao M, Pachter BR. Cocaine-associated intestinal gangrene in a pregnant woman. Am J Emerg Med 1997; 15:510–512. Johnson MO, Lobo ML. Case study of the home health management of a child with congenital anomalies associated with prenatal cocaine abuse. J Pediatr Nurs 1995; 10:375–382. Kankirawatana P, Tennison MB, D’Cruz O, Greenwood RS. Mobius syndrome in infant exposed to cocaine in utero. Pediatr Neurol 1993; 9:71–72. Kapur RP, Shaw CM, Shepard TH. Brain hemorrhages in cocaine-exposed human fetuses. Teratology 1991; 44:11–18. Lampley EC, Williams S, Myers SA. Cocaine-associated rhabdomyolysis causing renal failure in pregnancy. Obstet Gynecol 1996; 87:804–806. Lessick M, Vasa R, Israel J. Severe manifestations of oculoauriculovertebral spectrum in a cocaine exposed infant. J Med Genet 1991; 28:803–804. Lezcano L, Antia DE, Sahdev S, Jhaveri M. Crossed renal ectopia associated with maternal alkaloid cocaine abuse: a case report. J Perinatol 1994; 14:230–233. Liu SS, Forrester RM, Murphy GS, et al. Anaesthetic management of a parturient with myocardial infarction related to cocaine use. Can J Anaesth 1992; 39:858–861. Martinez JM, Fortuny A, Comas C, et al. Body stalk anomaly associated with maternal cocaine abuse. Prenatal Diagn 1994; 14:669–672. Meeker JE, Reynolds PC. Fetal and newborn death associated with maternal cocaine use. J Anal Toxicol 1990; 14:379–382. Mendelson MA, Chandler J. Postpartum cardiomyopathy associated with maternal cocaine abuse. Am J Cardiol 1992; 70:1092–1094. Mercado A, Johnson G Jr, Calver D, Sokol RJ. Cocaine, pregnancy, and postpartum intracerebral hemorrhage. Obstet Gynecol 1989; 73:467–468. Miller BM, Rosario PG, Prakash K, et al. Neonatal intestinal perforation: the ‘‘crack’’ connection. Am J Gastroenterol 1990; 85:767–769. Mishra A, Landzberg BR, Parente JT. Uterine rupture in association with alkaloidal (‘‘crack’’) cocaine abuse. Am J Obstet Gynecol 1995; 173:243–244. Mittleman RE, Cofino JC, Hearn WL. Tissue distribution of cocaine in a pregnant woman [published erratum appears in J Forens Sci 1989; 34(4):807]. J Forens Sci 1989; 34:481– 486. Moen MD, Caliendo MJ, Marshall W, Uhler ML. Hepatic rupture in pregnancy associated with cocaine use. Obstet Gynecol 1993; 82:687–689. Moriya F, Chan KM, Noguchi TT, Parnassus WN. Detection of drugs-of-abuse in meconium of a stillborn baby and in stool of a decreased 41-day-old infant. J Forens Sci 1995; 40:505– 508. Murphy JLJ. Hypertension and pulmonary oedema associated with ketamine administration in a patient with a history of substance abuse. Can J Anaesth 1993; 40:160–164. Nolte KB. Cocaine, fetal loss, and the role of the forensic pathologist. J Forens Sci 1991; 36:926–929. Nyirjesy P, Kamnani AJ, Scharf ML, Suh B. Fulminant tuberculosis complicating pregnancy in a patient infected with the human immunodeficiency virus. J Maternal Fetal Med 1993; 2:75–78. Okoruwa E, Shah R, Gerdes K. Apnea and vomiting in an infant due to cocaine exposure. Iowa Med 1995; 85:449–450. Ostrea EM Jr, Romero A, Knapp DK, et al. Postmortem drug analysis of meconium in earlygestation human fetuses exposed to cocaine: clinical implications. J Pediatr 1994; 124:477– 479. Perlow JH, Schlossberg DL, Strassner HT. Intrapartum cocaine use: a case report. J Reprod Med 1990; 35:978–980.

414

Addis et al.

100. Peters AJ, Abrams RM, Burchfield DJ, Gilmore RL. Seizures in a fetal lamb after cocaine exposure: a case report. Epilepsia 1992; 33:1001–1004. 101. Porat R, Brodsky N. Cocaine: a risk factor for necrotizing enterocolitis. J Perinatol 1991; 11:30–32. 102. Potter S, Klein J, Valiante G, et al. Maternal cocaine use without evidence of fetal exposure. J Pediatr 1994; 125:652–654. 103. Rais-Bahrami K, Naqvi M. Hydranencephaly and maternal cocaine use: a case report. Clin Pediatr 1990; 29:729–730. 104. Reznik VM, Anderson J, Griswold WR, et al. Successful fibrinolytic treatment of arterial thrombosis and hypertension in a cocaine-exposed neonate. Pediatrics 1989; 84:735–738. 105. Robin NH, Zackai EH. Unusual craniofacial dysmorphia due to prenatal alcohol and cocaine exposure. Teratology 1994; 50:160–164. 106. Sarpong S, Headings V. Sirenomelia accompanying exposure of the embryo to cocaine. South Med J 1992; 85:545–547. 107. Seballos RJ, Mendel SG, Mirmiran-Yazdy A, et al. Sarcoid cardiomyopathy precipitated by pregnancy with cocaine complications. Chest 1994; 105:303–305. 108. Sheinbaum KA, Badell A. Physiatric management of two neonates with limb deficiencies and prenatal cocaine exposure. Arch Phys Med Rehabil 1992; 73:385–388. 109. Sims ME, Walther FJ. Neonatal ultrasound casebook: antenatal brain injury and maternal cocaine use. J Perinatol 1989; 9:349–350. 110. Singer L, Arendt R, Minnes S, et al. Increased psychological distress in post-partum, cocaineusing mothers. J Substance Abuse 1995; 7:165–174. 111. Skopp G, Potsch L. A case report on drug screening of nail clippings to detect prenatal drug exposure. Ther Drug Monit 1997; 19:386–389. 112. Spinazzola R, Kenigsberg K, Usmani SS, Harper RG. Neonatal gastrointestinal complications of maternal cocaine abuse. NY State J M Med 1992; 92:22–23. 113. Spires MC, Gordon EF, Choudhuri M, et al. Intracranial hemorrhage in a neonate following prenatal cocaine exposure. Pediatr Neurol 1989; 5:324–326. 114. Spital A, Greenwell R. Severe hyperkalemia during magnesium sulfate therapy in two pregnant drug abusers. South Med J 1991; 84:919–921. 115. Streissguth AP, Grant TM, Barr HM, et al. Cocaine and the use of alcohol and other drugs during pregnancy. Am J Obstet Gynecol 1991; 164:1239–1243. 116. Sturner WQ, Sweeney KG, Callery RT, Haley NR. Cocaine babies: the scourge of the ’90s. J Forens Sci 1991; 36:34–39. 117. Sumner GS, Mandoki MW, Matthews-Ferrari K. A psychiatric population of prenatally cocaine-exposed children. J Am Acad Child Adolesc Psychiatry 1993; 32:1003–1006. 118. Sztulman L, Ducey JJ, Tancer ML. Intrapartum, intranasal cocaine and use and acute fetal distress: a case report. J Reprod Med 1990; 35:917–918. 119. Thatcher SS, Corfman R, Grosso J, et al. Cocaine use and acute rupture of ectopic pregnancies. Obstet Gynecol 1989; 74:478–479. 120. Towers CV, Pircon RA. Nageotte MP, et al. Cocaine intoxication presenting as preeclampsia and eclampsia. Obstet Gynecol 1993; 81:545–547. 121. van den Anker JN, Van Vught EE, Zandwijken GRJ, et al. Severe limb abnormalities: analysis of a cluster of five cases born during a period of 45 days. Am J Med Genet 1993; 45: 659–667. 122. Viscarello RR, Ferguson DD, Nores J, Hobbins JC. Limb-body wall complex associated with cocaine abuse: further evidence of cocaine’s teratogenicity. Obstet Gynecol 1992; 80:523– 526. 123. Wilson BE, Hobbs WN. Cocaine toxicity in glycogen storage disease (letter). West J Med 1993; 159:508–509. 124. Alexander LL. Fetal addiction: health problem or criminal offense? (editorial). J Natl Med Assoc 1991; 83:663.

Fetal Effects of Cocaine

415

125. Amoury RA. Necrotizing enterocolitis: a continuing problem in the neonate. World J Surg 1993; 17:363–373. 126. Anonymous. American Academy of Pediatrics Committee on Substance Abuse: Drugexposed infants. Pediatrics 1990; 86:639–642. 127. Anonymous. Alternative case-finding methods in a crack-related syphilis epidemic—Philadelphia. MMWR 1991; 40:77–80. 128. Baxley EG. Cocaine abuse in pregnancy—a myriad of unanswered questions (editorial). J S C Med Assoc 1990; 86:555–556. 129. Brouhard BH. Cocaine ingestion and abnormalities of the urinary tract (editorial; comment). Clin Pediatr 1994; 33:157–158. 130. Chasnoff IJ. Cocaine and pregnancy—implications for the child [editorial]. West J Med 1989; 150:456–458. 131. Fantel AG, Shepard TH. Prenatal cocaine exposure (editorial). Reprod Toxicol 1990; 4:83. 132. Heagarty MC. Crack cocaine. A new danger for children (editorial). Am J Dis Child 1990; 144:756–757. 133. Howard BJ, O’Donnell KJ. What is important about a study of within-group differences of ‘‘cocaine babies’’? (editorial). Arch Pediatr Adolesc Med 1995; 149:663–664. 134. Jentzen J. Medical complications of cocaine abuse [editorial]. Am J Clin Pathol 1993; 100: 475–476. 135. Jos PH, Marshall MF, Perlmutter M. The Charleston policy on cocaine use during pregnancy: a cautionary tale. J Law Med Ethics 1995; 23:120–128. 136. Kearney MH, Murphy S, Irwin K, Rosenbaum M. Salvaging self: a grounded theory of pregnancy on crack cocaine. Nurs Res 1995; 44:208–213. 137. Konkol RJ. Is there a cocaine baby syndrome? (editorial). J Child Neurol 1994; 9:225– 226. 138. Lewis KD, Bennett B, Schmeder NH. The care of infants menaced by cocaine abuse. Am J Maternal Child Nurs 1989; 14:324–329. 139. Lone P. Silencing crack addiction. Am J Maternal Child Nurs 1991; 16:202–205. 140. Neuspiel DR. On pejorative labeling of cocaine exposed children (editorial). J Subst Abuse Treat 1993; 10:407. 141. Nora JG. Perinatal cocaine use: maternal, fetal, and neonatal effects. Neonatal Netw 1990; 9:45–52. 142. Plessinger MA, Woods JRJ. Maternal, placental, and fetal pathophysiology of cocaine exposure during pregnancy. Clin Obstet Gynecol 1993; 36:267–278. 143. Salamy A, Eldredge L, Anderson J, Bull D. Brain-stem transmission time in infants exposed to cocaine in utero. J Pediatr 1990; 117:627–629. 144. Tudehope DI. Perinatal cocaine intoxication: a precaution (editorial). Med J Aust 1989; 150: 290–291. 145. Zuckerman B, Frank DA. Prenatal cocaine exposure: nine years later (editorial; comment). J Pediatr 1994; 124:731–733. 146. Abdeljaber M, Nolan BM, Schork MA. Maternal cocaine use during pregnancy: effect on the newborn infant (letter; comment). Pediatrics 1990; 85:630. 147. Ahmed MS, Spong CY, Geringer JL, et al. Prospective study on cocaine use prior to delivery. JAMA 1989; 262:1880. 148. Anonymous. Effects of maternal marijuana and cocaine use on fetal growth (letter; comment). N Engl J Med 1989; 321:979. 149. Bays J. Fetal vascular disruption with prenatal exposure to cocaine or methamphetamine (letter; comment). Pediatrics 1991; 87:416–418. 150. Carraccio C, Papadimitriou J, Feinberg P. Subcutaneous fat necrosis of the newborn: link to maternal use of cocaine during pregnancy (letter). Clin Pediatr 1994; 33:317–318. 151. Chavkin W, Kandall SR. Between a ‘‘rock’’ and a hard place: perinatal drug abuse. Pediatrics 1990; 85:223–225.

416

Addis et al.

152. Cottler LB, Compton WM. Re: ‘‘Cocaine use during pregnancy: perinatal outcomes’’ (letter; comment). Am J Epidemiol 1992; 135:1425–1427. 153. DiGregorio GJ, Barbieri EJ, Ferko AP, Ruch EK. Prevalence of cocaethylene in the hair of pregnant women (letter). J Anal Toxicol 1993; 17:445–446. 154. Downing GJ, Horner SR, Kilbride HW. Characteristics of perinatal cocaine-exposed infants with necrotizing enterocolitis (letter). Am J Dis Child 1991; 145:26–27. 155. Fogarty ME. Re: ‘‘A review of the literature on cocaine abuse in pregnancy’’ (letter; comment). Nurs Res 1991; 40:235. 156. Friedman EH. Neurobiology of cocaine binging in pregnancy (letter; comment). Am J Obstet Gynecol 1995; 172:1322–1323. 157. Goldin K. Cocaine abuse in pregnancy (letter; comment). JAMA 1989; 262:771–771. 158. Good WV, Ferriero D. Ocular effects of prenatal cocaine exposure (letter; comment). Ophthalmology 1994; 101:1321; discussion 132–1321. 159. Graham K, Koren G. Maternal cocaine use and risk of sudden infant death (letter; comment). J Pediatr 1989; 115:333–333. 160. Gratacos E, Torres PJ, Antolin E. Use of cocaine during pregnancy (letter). N Engl J Med 1993; 329:667. 161. Hansen RL. Diagnosis, treatment, and follow-up of newborns addicted to cocaine. West J Med 1990; 153:646. 162. Herschman Z, Aaron C. Prolongation of cocaine effect (letter; comment). Anesthesiology 1991; 74:631–632. 163. Hunt CE. Respiratory pattern abnormalities and prenatal cocaine exposure (letter). Am J Dis Child 1990; 144:138–139. 164. Jerome L. Adopted children exposed to cocaine in utero: confounding factors (letter; comment). Can Med Assoc J 1995; 152:1187–1188. 165. Kain ZN, Kain TS, Scarpelli EM. Effect of cocaine use on the fetus (letter; comment). N Engl J Med 1992; 327:1393–1394. 166. Levy M. Is cocaine a risk factor to necrotizing enterocolitis? (letter). Clin Pediatr 1993; 32: 700–701. 167. Maynard EC. Maternal abuse of cocaine and heroin (letter; comment). Am J Dis Child 1990; 144:520–521. 168. Miele NF. Controversial costs of cocaine: (letter; comment). JAMA 1992; 267:507; discussion 507–508. 169. Mittleman RE, Cofino J, Hearn WL. Addendum on ‘‘Cocaine in a pregnant woman’’ (letter). J Forens Sci 1989; 34:807. 170. Moore CM, Brown S, Negrusz A, et al. Determination of cocaine and its major metabolite, benzoylecgonine, in amniotic fluid, umbilical cord blood, umbilical cord tissue, and neonatal urine: a case study (letter). J Anal Toxicol 1993; 17:62–62. 171. Morfesis FA. Association between newborn birth weight and prenatal care for cocaine users (letter; comment). JAMA 1994; 271:1161. 172. Neuspiel DR. Cocaine-associated abnormalities may not be causally related (letter; comment). Am J Dis Child 1992; 146:278–279. 173. Neuspiel DR, Hamel SC. Neurobehavioral sequelae of fetal cocaine exposure (letter; comment). J Pediatr 1992; 120:661–662. 174. Neuspiel DR. Screening neonates for intrauterine cocaine exposure (letter; comment). J Pediatr 1995; 126:323–324. 175. Nucci P, Brancato R. Ocular effects of prenatal cocaine exposure (letter; comment). Ophthalmology 1994; 101:1321; discussion 132–1321; discussion 132 176. Page D. Controversial costs of cocaine (letter; comment). JAMA 1992; 267:507; discussion 507–507; discussion 508. 177. Petrulis AS. Hypertension in pregnancy (letter; comment). N Engl J Med 1992; 327:733– 733.

Fetal Effects of Cocaine

417

178. Racine AD, Joyce TJ, Anderson R. The association between prenatal care and birth weight among women exposed to cocaine in New York City: a correction. JAMA 1994; 271:1161– 1162. 179. Sackoff J, Kline J, Kinney A, Grunebaum A. Cocaine use in obstetric patients underreported (letter). Am J Public Health 1992; 82:1043. 180. Shaw GM, Malcoe LH, Lammer EJ, Swan SH. Maternal use of cocaine during pregnancy and congenital cardiac anomalies (letter). J Pediatr 1991; 118:167–168. 181. Sher J. Women and crack addition (letter). J Am Med Womens Assoc 1989; 44:166. 182. Stephens BG. Fetal development and cocaine (letter). Am J Forens Med Pathol 1989; 10: 268–269. 183. Sugarman K, Herman M. Crack cocaine and HIV in the inner city (letter; comment). N Engl J Med 1995; 332:1233; discussion 1234–1235. 184. Tabor BL, Saffici A. Cocaine effects on fetal behavioral state (letter; comment). Am J Obstet Gynecol 1990; 163:1364–1365. 185. Thorp JM Jr. Cocaine abuse is associated with abruptio placentae and decreased birth weight, but not shorter labor (letter; comment). Obstet Gynecol 1991; 77:807–808. 186. van den Anker JN, Cohen-Overbeek TE, Wladimiroff JW, Sauer PJ. Prenatal diagnosis of limb-reduction defects due to maternal cocaine use (letter). Lancet 1991; 338:1332. 187. van den Anker JN, Sauer PJ. Effect of cocaine use on the fetus (letter; comment). N Engl J Med 1992; 327:1394. 188. Yang YQ, Lee MP, Schenken RS, et al. Effects of binding on human transplacental transfer of cocaine (letter; comment). Am J Obstet Gynecol 1995; 172:720–722. 189. Abel EL. The future of cocaine babies: Primary care and early intervention. J Pediatr Health Care 1991; 5:321–323. 190. Adams C, Eyler FD, Behnke M. Nursing intervention with mothers who are substance abusers. J Perinat Neonat Nurs 1990; 3:43–52. 191. Anonymous. Urogenital anomalies in the offspring of women using cocaine during early pregnancy—Atlanta, 1968–1980. MMWR 1989; 38:536–541. 192. Anonymous. Cocaine in pregnancy. ACOG Committee Opinion: Committee on Obstetrics: Maternal and Fetal Medicine no. 114—September 1992 (replaces no. 81, March 1980). Int J Gynaecol Obstet 1993; 41:102–105. 193. Aronson RA, Hunt LH. Cocaine use during pregnancy: implications for physicians. Wis Med J 1990; 89:105–110. 194. Berger CS, Sorensen L, Gendler B, Fitzsimmons J. Cocaine and pregnancy: a challenge for health care providers. Health Social Work 1990; 15:310–316. 195. Brent RL, Beckman DA. The contribution of environmental teratogens to embryonic and fetal loss. Clin Obstet Gynecol 1994; 37:646–670. 196. Buehler BA. Cocaine: how dangerous is it during pregnancy? Nebr Med J 1995; 80:116– 117. 197. Buehler BA, Conover B, Andres RL. Teratogenic potential of cocaine (review). Semin Perinatol 1996; 20:93–98. 198. Byrne MW, Lerner HM. Communicating with addicted women in labor. Am J Maternal Child Nurs 1992; 17:22–26. 199. Chan L, Pham H, Reece EA. Pneumothorax in pregnancy associated with cocaine use (review). Am J Perinatol 1997; 14:385–388. 200. Chao CR. Cardiovascular effects of cocaine during pregnancy (review). Semin Perinatol 1996; 20:107–114. 201. Chasnoff IJ. Drug use in pregnancy. N Y State J Med 1989; 89:255. 202. Chasnoff IJ. Missing pieces of the puzzle (comment). Neurotoxicol Teratol 1993; 15:287– 288; discussion 311–312. 203. Chavkin W, Paone D, Friedmann P, Wilets I. Reframing the debate: toward effective treatment for inner city drug-abusing mothers. Bull N Y Acad Med 1993; 70:50–68.

418

Addis et al.

204. Church MW. Does cocaine cause birth defects: Neurotoxicol Teratol 1993; 15:289; discussion 311–289; discussion 312. 205. Coles CD. Discussion: measurement issues in the study of effects of substance abuse in pregnancy. NIDA Res Monogr 1992; 117:248–258. 206. Coles CD. Saying ‘‘goodbye’’ to the ‘‘crack baby’’ (comment). Neurotoxicol Teratol 1993; 15:290–292; discussion 311–312. 207. Cornish JW, O’Brien CP. Crack cocaine abuse: an epidemic with many public health consequences. [review]. Annu Rev Public Health 1996; 17:259–273. 208. Day NL, Cottreau CM, Richardson GA. The epidemiology of alcohol, marijuana, and cocaine use among women of childbearing age and pregnancy women. Clin Obstet Gynecol 1993; 36:232–245. 209. Day NL, Richardson GA. Cocaine use and crack babies: science, the media, and miscommunication (comment). Neurotoxicol Teratol 1993; 15:293–294; discussion 311–312. 210. Dixon SD. Effects of transplacental exposure to cocaine and methamphetamine on the neonate. West J Med 1989; 150:436–442. 211. Dow-Edwards DL. Cocaine effects on fetal development: a comparison of clinical and animal research findings. Neurotoxicol Teratol 1991; 13:347–352. 212. Dow-Edwards D. The puzzle of cocaine’s effects following maternal use during pregnancy: still unsolved (comment). Neurotoxicol Teratol 1993; 15:295–296; discussion 311–312. 213. Dumas L. Assessing the cocaine-addicted mother: guidelines for the initial visit. Home Healthcare Nurse 1992; 10:12–18. 214. Dungy-Poythress LJ. Cocaine effects on pregnancy and infant outcome: do we really know how bad it is? J Assoc Acad Minority Physicians 1995; 6:46–50. 215. Elhassani SB. Cocaine use and effect: a major perinatal risk factor in the nineteen nineties. J SC Med Assoc 1990; 86:532–535. 216. Ellen L, Brock MD. Perinatal cocaine abuse as a public health problem. Ceskoslovensk Pediatr 1993; 48:249–251. 217. Fantel AG. Puzzle of cocaine’s effects following maternal use during pregnancy: are there reconcilable differences? (comment). Neurotoxicology Teratol 1993; 15:297; discussion 311–297; discussion 312. 218. Ferner RE. Adverse drug reaction bulletin. Adverse Drug React Bull 1993; 161:607–610. 219. Finnegan LP. The teratogenicity of the drugs of abuse: a symposium. Drug Alcohol Depend 1994; 36:81. 220. Frank DA, Zuckerman BS. Children exposed to cocaine prenatally: pieces of the puzzle (comment). Neurotoxicol Teratol 1993; 15:298–300; discussion 311–312. 221. Garcia SA. Birth penalty: societal responses to perinatal chemical dependence. J Clin Ethics 1990; 1:135–140; discussion 140–145. 222. Gaskill SJ. The spectrum of radiologic abnormalities in the neonatal CNS (comment). AJNR 1992; 13:1272–1272. 223. Giacoia GP. Cocaine babies in Oklahoma. J Okla State Med Assoc 1990; 83:64–67. 224. Gintautiene K, Longmore W, Abadir AR, et al. Cocaine-induced deaths in pediatric population. Proc West Pharmacol Soc 1990; 33:247–248. 225. Goldstein FJ. Toxicity of cocaine. Compendium 1990; 11:710, 712, 714–716. 226. Goodwin FK. From the Alcohol, Drug Abuse, and Mental Health Administration. JAMA 1990; 263:1610. 227. Grossman J, Schottenfeld RS, Viscarello R, Pakes J. Cocaine abuse during pregnancy. NIDA Res Monogr 1993; 132:302. 228. Horger EO, Brown SB, Condon CM. Cocaine in pregnancy: confronting the problem. J SC Med Assoc 1990; 86:527–531. 229. Janke JR. Prenatal cocaine use: effects on perinatal outcome. J Nurse Midwife 1990; 35: 74–77.

Fetal Effects of Cocaine

419

230. Johnson E. From the Alcohol, Drug Abuse, and Mental Health Administration. JAMA 1992; 268:854. 231. Kaltenbach KA. Effects of in-utero opiate exposure: new paradigms for old questions. Drug Alcohol Depend 1994; 36:83–87. 232. Kandall SR, Gaines J. Maternal substance use and subsequent sudden infant death syndrome (SIDS) in offspring. Neurotoxicol Teratol 1991; 13:235–240. 233. Karmel BZ, Gardner JM, Magnano CL. Neurofunctional consequences of in utero cocaine exposure. NIDA Res Monogr 1991; 105:535–536. 234. Keller S, Niebyl J. Cocaine abuse in a high risk obstetrical population. Iowa Med 1993; 83: 153–155. 235. Kenner C, D’Apolito K. Outcomes for children exposed to drugs in utero (review). J Obstet Gynecol Neonat Nurs 1997; 26:595–603. 236. Khoury MJ, James LM, Lynberg MC. Quantitative analysis of associations between birth defects and suspected human teratogens. Am J Med Genet 1991; 40:500–505. 237. Klitsch M. Prenatal exposure to tobacco, alcohol or other drugs found for more than one in 10 California newborns. Fam Plann Perspect 1994; 26:95–96. 238. Knisely JS, Spear ER, Green DJ, et al. Substance abuse patterns in pregnant women. NIDA Res Monogr 1991; 105:280–281. 239. Koren G, Graham K, Shear H, Einarson T. Bias against the null hypothesis: the reproductive hazards of cocaine. Lancet 1989; 2:1440–1442. 240. Landry SH, Whitney JA. The impact of prenatal cocaine exposure: studies of the developing infant (review). Semin Perinatol 1996; 20:99–106. 241. Lauder JM. Discussion: Neuroteratology of cocaine—relationship to developing monoamine systems. NIDA Res Monogr 1991; 114:233–247. 242. Lewis KD. Pathophysiology of prenatal drug exposure: in utero, in the newborn, in childhood, and in agencies. J Pediatr Nurs 1991; 6:185–190. 243. Litt J, McNeil M. Biological markers and social differentiation: crack babies and the construction of the dangerous mother (review). Health Care Women Int 1997; 18:31–41. 244. Lunsford BK. Pregnancy and cocaine: a charge to nurses. Kansas Nurse 1995; 70:8–9. 245. Martin ML, Khoury MJ. Cocaine and single ventricle: a population study. Teratology 1992; 46:267–270. 246. Martin ML, Khoury MJ, Cordero JF, Waters GD. Trends in rates of multiple vascular disruption defects. Altanta, 1968–1989: is there evidence of a cocaine teratogenic epidemic? Teratology 1992; 45:647–653. 247. Meyer JS, Shearman LP, Collins LM. Monoamine transporters and the neurobehavioral teratology of cocaine (review). Pharmacol Biochem Behav 1996; 55:585–593. 248. Miller WHJ, Cox SM, Harbison V, Campbell BA. Urine drug screens for drug abuse in pregnancy: problems and pitfalls. Womens Health Issues 1994; 4:152–155. 249. Miller H. Prenatal cocaine exposure and mother-infant interaction: implications for occupational therapy intervention. (review). Am J Occup Ther 1997; 51:119–131. 250. Murphey LJ, Olsen GD, Kondol RJ. Quantitation of benzoylnorecgonine and other cocaine metabolites in meconium by high-performance liquid chromatography. J Chromatogr 1993; 613:330–335. 251. Neuspiel DR. Cocaine and the fetus: mythology of severe risk (comment). Neurotoxicol Teratol 1993; 15:305–306; discussion 311–312. 252. Ostrea EM Jr. Detection of prenatal drug exposure in the pregnant woman and her newborn infant. NIDA Res Monogr 1992; 117:61–79. 253. Pokorni JL, Stanga J. Serving infants and families affected by maternal cocaine abuse: Part 1 (review). Pediatr Nurs 1996; 22:439–442. 254. Richardson GA, Day NL, McGauhey PJ. The impact of prenatal marijuana and cocaine use on the infant and child. Clin Obstet Gynecol 1993; 36;302–318.

420

Addis et al.

255. Richardson GA, Hamel SC, Goldschmidt L, Day NL. The effects of prenatal cocaine use on neonatal neurobehavioral status. Neurotoxicol Teratol 1996; 18:519–528. 256. Ripple MG, Goldberger BA, Caplan YH, et al. Detection of cocaine and its metabolites in human amniotic fluid. J Anal Toxicol 1992; 16:328–331. 257. Rizk B, Atterbury JL, Groome LJ. Reproductive risks of cocaine (review). Hum Reprod Update 1996; 2:43–55. 258. Sexson WR. Cocaine: a neonatal perspective. Int J Addict 1993; 28:585–598. 259. Shepard TH. Nutritional aspects of embryonic CNS development: in vitro and animal studies. Introduction to part I. Ann NY Acad Sci 1993; 678:1–7. 260. Slade PH. Legislative and community response to cocaine use in pregnancy. Florida Nurse 1993; 41:8, 13. 261. Spear LP. Missing pieces of the puzzle complicate conclusions about cocaine’s neurobehavioral toxicity in clinical populations: importance of animal models (comment). Neurotoxicol Teratol 1993; 15:307–309; discussion 311–312. 262. Sprauve ME. Substance abuse and HIV pregnancy (review). Clin Obstet Gynecol 1996; 39: 316–332. 263. Strauss RS. Effects of the intrauterine environment on childhood growth (review). Br Med Bull 1997; 53:81–95. 264. Sun WY, Ghen W. The impact of maternal cocaine use on neonates in socioeconomic disadvantaged population (review). J Drug Educ 1997; 27:389–396. 265. Swadi H. Adolescent substance misuse. Curr Opin Psychiatry 1993; 6:511–515. 266. Szeto HH. Discussion: methodological issues in controlled studies on effects of prenatal drugs. NIDA Res Monogr 1991; 114:37–44. 267. Warner EA, Kosten TR, O’Connor PG. Pharmacotherapy for opioid and cocaine abuse (review). Med Clin North Am 1997; 81:909–925. 268. Webster WS, Brown Woodman PDC, Richi HE. Birth defects—causes and myths. Today Life Sci 1993; 5:10–19. 269. Westdorp EJ, Salomone JA, Roberts DK, et al. Validation of a rapid urine screening assay for cocaine use among pregnant emergency patients. Acad Emerg Med 1995; 2:795– 798. 270. Wightman MJ. Criteria for placement decisions with cocaine-exposed infants. Child Welfare 1991; 70:653–663. 271. Wootton J, Miller SI. Cocaine: a review. Pediatr Rev 1994; 15:89–92. 272. Zuckerman B, Frank DA. ‘‘Crack kids’’: not broken (comment). Pediatrics 1992; 89:337– 339. 273. Zuckerman B. Selected methodologic issues in investigations of prenatal effects of cocaine: lessons from the past (review). NIDA Res Monogr 1991; 114:45–54. 274. Abusada GM, Abukhalaf IK, Alford DD, et al. Solid-phase extraction and GC/MS quantitation of cocaine, ecgonine methyl ester, benzoylecgonine, and cocaethylene from meconium, whole blood, and plasma. J Anal Toxicol 1993; 17:353–358. 275. Alemi F, Stephens RC. Computer services for patients: description of systems and summary of findings. Med Care 1996; 34:OS1–OS9. 276. Alemi F, Stephens RC, Javalghi RG, et al. A randomized trial of a telecommunications network for pregnant women who use cocaine. Med Care 1996; 34:OS10–OS20. 277. Alemi F, Stephens RC, Muise K, et al. Educating patients at home: community health rap. Med Care 1996; 34:OS21–OS31. 278. Amaro H, Fried LE, Cabral H, Zuckerman B. Violence during pregnancy and substance use. Am J Public Health 1990; 80:575–579. 279. Anonymous. Statewide prevalence of cocaine use during the perinatal period. RI Med J 1990; 73:272. 280. Anonymous. Population-based prevalence of perinatal exposure to cocaine—Georgia, 1994. MMWR 1996; 45:887–891.

Fetal Effects of Cocaine

421

281. Archie CL, Anderson MM, Gruber EL. Positive smoking history as a preliminary screening device for substance use in pregnant adolescents. J Pediatr Adolesc Gynecol 1997; 10:13– 17. 282. Ball SA, Schottenfeld RS. A five-factor model of personality and addiction, psychaitric, and AIDS risk severity in pregnant and postpartum cocaine misusers. Subst Use Misuse 1997; 32:25–41. 283. Bauchner H, Zuckerman B. Cocaine, sudden infant death syndrome, and home monitoring (editorial). J Pediatr 1990; 117:904–906. 284. Behnke M, Eyler FD, Conlon M, et al. How fetal cocaine exposure increases neonatal hospital costs. Pediatrics 1997; 99:204–208. 285. Bendersky M, Alessandri S, Gilbert P, Lewis M. Characteristics of pregnant substance abusers in two cities in the northeast. Am J Drug Alcohol Abuse 1996; 22:349–362. 286. Berenson AB, Stiglich NJ, Wilkinson GS, Anderson GD. Drug abuse and other risk factors for physical abuse in pregnancy among white non-Hispanic, black, and Hispanic women. Am J Obstet Gynecol 1991; 164:1491–1499. 287. Billman DO, Nemeth PB, Heimler R, Sasidharan P. Prenatal cocaine/polydrug exposure: effect of race on outcome. J Perinatol 1996; 16:366–369. 288. Browne SP, Tebbett IR, Moore CM, et al. Analysis of meconium for cocaine in neonates. J Chromatogr 1992; 575:158–161. 289. Browne S, Moore C, Negrusz A, et al. Detection of cocaine, norcocaine, and cocaethylene in the meconium of premature neonates. J Forens Sci 1994; 39:1515–1519. 290. Brunader RE, Brunader JA, Kugler JP. Prevalence of cocaine and marijuana use among pregnant women in a military health care setting. J Am Board Fam Pract 1991; 4:395–398. 291. Burke MS, Roth D. Anonymous cocaine screening in a private obstetric population. Obstet Gynecol 1993; 81:354–356. 292. Callahan CM, Grant TM, Phipps P, et al. Measurement of gestational cocaine exposure: sensitivity of infants’ hair, meconium, and urine [published erratum appears in J Pediatr 1992; 121(1):156]. J Pediatr 1992; 120:763–768. 293. Cartwright PS, Schorge JO, McLaughlin FJ. Epidemiologic characteristics of drug use during pregnancy: experience in a Nashville hospital. South Med J 1991; 84:867–870. 294. Casanova OQ, Lombardero N, Behnke M, et al. Detection of cocaine exposure in the neonate: analyses of urine, meconium, and amniotic fluid from mothers and infants exposed to cocaine. Arch Pathol Lab Med 1994; 118:988–993. 295. Chasnoff IJ, Landress HJ, Barrett ME. The prevalence of illicit-drug or alcohol use during pregnancy and discrepancies in mandatory reporting in Pinellas County, Florida. N Engl J Med 1990; 322:1202–1206. 296. Charpentier PA, Schottenfeld RS. A database model for studies of cocaine-dependent pregnant women and their families. NIDA Res Monogr 1996; 166:242–2