DRUGS AND PREGNANCY.

Bertis B Little PhD.

Preface.

The purpose of this volume was originally to condense and update Drugs and Pregnancy Drugs and Pregnancy second edition. However, the book has evolved into a larger project since its inception in 2000. The total length of the original typescript was approximately three times the number of pages for which the publisher contracted. First strategy suggested was to eliminate the large number of references, but this was not acceptable. The final compromise developed was to post the full bibliography and supporting materials on a website second edition. However, the book has evolved into a larger project since its inception in 2000. The total length of the original typescript was approximately three times the number of pages for which the publisher contracted. First strategy suggested was to eliminate the large number of references, but this was not acceptable. The final compromise developed was to post the full bibliography and supporting materials on a website for the book http://www.drugsandpregnancy.com.

It is intended for the content of this book to be updated and corrected as necessary approximately four times per year. These additions to the book will be posted to the website above, which will be maintained by the publisher. In addition, a searchable index of proprietary and generic names is provided on the website.

In addition, a link to TERIS is provided on the website, and the reader is strongly encouraged to use TERIS in counseling patients who have been exposed to a medication during pregnancy. The information in TERIS has been vetted by leading authorities in the field, and is tantamount to high quality peer reviewed literature. It was through working on the development of TERIS (writing agent summaries, knowledgebase design) from 1985 to 1989 that I became deeply interested in human teratology.

Links to other sites that may be of use to readers of this book will also be included on the website. A feature under development for http://www.drugsandpregnancy.com is a current literature alert window. We have developed an automated search agent that will identify new publications (journal articles, books, etc.) relevant to human teratology, and will post those to the website under CURRENT LITERATURE ALERTS. This will a.s.sist the reader in maintaining access to up-to-date information. The main advantage of having a website accompany the published book is currency of information. Books usually take a year or longer to reach the reader after the author has completed the typescript, and may already be out-of-date by the time of release.

I thank the publisher for providing this option to (1) maintain scholarly references for the book's content, (2) provide readers the most recent information available, and (3) refer readers to other authoritative sources such as TERIS.

BBL.

July 2006 This page intentionally left blank

Acknowledgements

My friend and colleague, Rick Weideman, PharmD of the VA Medical Center, Dallas even gave vacation time to help with the completion of this volume. Beverly A DelHomme, JD, my wife, has read and edited the typescript of this book more than a couple of times over the past several years. Marie Kelly, MD, friend and colleague, made significant contributions to the completion of this volume through editing, updating sources, and information during her personal holiday time. My Staff a.s.sistant, Nona Williamson, did computer MEDLINE searches, typed chapters, checked bibliography, and sought out difficult-to-find literature. Eva Malina, PhD also a.s.sisted in production of this volume during her summer vacation by reading and marking page proofs. Donna Savage, University Librarian, and her staff have patiently worked with Nona and me to acquire numerous uncommon reference sources.

Finally, I wish to thank the Illinois Poison Control Board for allowing reproduction of their comprehensive list of antidotes, and to Saunders/Elsevier Publishing for allowing the adaptation of Figure 14.1.

This page intentionally left blank

1.

Introduction to drugs in pregnancy Magnitude of the problem 2.Prenatal diagnosis 14.Clinical evaluation 3.Counseling and evaluation of the Human teratology principles 4.drug-exposed pregnant patient 14.Animal studies in clinical evaluation 6.Food and Drug Administration Human studies 6.cla.s.sification of drugs and Known human teratogens 8.informed consent 18.Critical time periods 8.Informed consent and Potential adverse effects 9.post-exposure counseling 19.Maternal physiology during Summary 20.pregnancy 12.Key references 21.Pharmac.o.kinetics in pregnancy 13.Birth defects occur among 3.55 percent of infants examined at birth or neonatally (Polifka and Friedman, 2002) but prevalence of birth defects may be as high as 8 percent, according to a universal disease registry from British Columbia (Baird et al et al., 1989).

Using the estimated teratogenic causes of birth defects in Fig. 1.1, it may be extrapolated that as many as 1 percent of congenital anomalies are caused by drugs, chemicals and other exogenous agents (i.e., approximately one in 400 infants has a birth defect with a teratogenic etiology). These estimates have not changed over the past decade and a half, perhaps because genomic research eclipses research in clinical teratology, as suggested by a recent review (Polifka and Friedman, 2002). Nonetheless, much research remains to be done because the magnitude of the problem of medication use during pregnancy may be somewhat underestimated because 6570 percent of birth defects have an unknown etiology. This may include unreported medically prescribed medication with teratogenic potential, use of alcohol and/or drugs of abuse, and other preventable causes of birth defects (i.e., congenital anomalies and other pregnancy complications due to drug and chemical exposure are unique because they are potentially preventable).

Knowledge of the effects of prenatal exposure and the window of opportunity for intervention are the key factors in evaluation and prevention of morbidity and mortality due to drug and chemical exposure during pregnancy. Chapters 215 summarize information currently available regarding drug exposure during pregnancy, with detailed 2 2 Introduction to drugs in pregnancy 3%.

4%.

Cytogenetic 2%.

1%.

Mendelian and 5%.

mutation Unknown (polygenic, etc.) Maternal infections 15%.

Maternal metabolic disease 70%.

Problems of constraint Medications, chemicals, radiation Figure 1.1 Causes of birth defects Causes of birth defects drug-specific information obtained from the current medical literature, clinical experience, and science.

Clinicians find it difficult to use the narrow window of opportunity to intervene in medication use during pregnancy because pregnant women do not present for prenatal care until embryogenesis is complete (i.e., after 58 days postconception). Intervention is further complicated because many women are not aware of the potential adverse affects of drugs and chemicals on pregnancy. For example, more than 60 percent of gravidas had never heard of fetal alcohol syndrome and were not aware of the adverse effects of alcohol on pregnancy in several surveys. Patient education prior to conception is obviously the best intervention, but very little funding is available for this. In addition, social and cultural barriers must also be overcome for the patient education process to be successful.

Even the most well-educated obstetrical patients have culturally based 'folk etiologies'

that they believe explain the occurrence of birth defects and other adverse pregnancy outcomes that are usually not correlated with medically founded causes. The author has counseled gravid physicians who were not entirely correct in their understanding of prenatal development and how the environment can be a disruptive influence. Folk or culture-specific explanations and educational background must therefore be considered when counseling the obstetrical patient of specific risks to pregnancy, including exposures to medications, drugs, and chemicals.

MAGNITUDE OF THE PROBLEM.

Women ingest a variety of medications or drugs during pregnancy, usually related to a medical condition that was being treated before the pregnancy was recognized.

Prevalence of medication use varied from less than 10 percent of pregnant women to Clinical evaluation Clinical evaluation 3.more than 95 percent. Frequently, more than one medication will be used. For example, in one comprehensive study in the United States of tens of thousands of patients, women received an average of 3.1 prescriptions for medications other than vitamins or iron during their pregnancies. Similar prevalences were observed in Brazil, Australia, New Zealand, and Egypt. The high end estimates are probably closer to the prevalence in 2004, and this is an international pattern and problem. Medication use during pregnancy is clearly a frequent event. However, safety may be questionable or simply unknown in many instances (Polifka and Friedman, 2002), primarily because of the paucity of clinical teratology research conducted over the last two decades (Lo and Friedman, 2002).

Three scenarios describe inadvertent drug exposure during pregnancy: (1) some medications are taken before the pregnancy is recognized; (2) some medications are taken without the physician's advice once the pregnancy is recognized; and (3) some are taken with physician's advice. In practice, the predominant case is for physicians to be faced with determining whether or not a medication or drug may be harmful to a pregnant woman or her unborn child after the exposure has occurred.

Also of concern are nonmedical exposures to drugs. Nonmedical exposures to drugs during pregnancy occur in suicide gestures (technically a subcategory of substance abuse) and substance abuse (i.e., recreational use). Suicide gestures occur among approximately 1 percent of pregnant women. Substance abuse during pregnancy is much more prevalent than suicide gestures, and is discussed in Chapter 15. Briefly, an estimated 1020 percent of pregnant women use an illicit substance and/or alcohol during their pregnancies. Cocaine seems to be the most frequently used substance in 2004.

CLINICAL EVALUATION.

Clinical evaluation of potentially teratogenic and/or toxic exposures during pregnancy must consider three separate components of normal pregnancy: maternal, embryonic, and fetal. Marked differences in the physiology of these components exist because of differences in the purposes of the cells, or the end points of cell division (replacement versus morphogenesis versus hyperplastic growth) and the metabolic capabilities of the mother and the developing conceptus. In the embryo, organs are being formed, and drugs cannot be metabolized at adult or fetal rates, if at all. The embryo is not a little fetus. The fetus is not a little adult. Most of the fetal period is occupied with growth in size of organs, not usually their formation, and these are growing very rapidly. Exceptions exist (e.g., thyroid, s.e.xual organs, brain cell 'arrangement'), but this is generally true for the fetus. Fetal enzyme systems involved in drug metabolism are only beginning to function, and some will not be active until after the neonatal period (e.g., cholinesterase). Pregnant women have the full enzyme complement for metabolizing drugs, but most such systems have lower activity during pregnancy, as does cholinesterase (Pritchard, 1955), which metabolizes cocaine. In addition, gender differences in the nonpregnant state also exist [e.g., alcohol dehydrogenase (ADH) among adult females is only 55 percent of adult males'

activity]. Therefore, the responses of adults, fetuses, embryos, and pregnant women to drugs (pharmacodynamics, pharmac.o.kinetics) differ markedly (Little, 1999). Therefore, it is important to differentiate the effects of drugs and chemicals upon these distinctly different components of pregnancy. We shall repeatedly observe that many drugs and chemicals have different effects on these three components of pregnancy.

4.Introduction to drugs in pregnancy HUMAN TERATOLOGY PRINCIPLES A teratogen is usually defined as any agent, physical force, or other factor (e.g., maternal disease) that can induce a congenital anomaly through alteration of normal development during any stage of embryogenesis (Polifka and Friedman, 2002). Agents include drugs and other chemicals. Physical forces include ionizing radiation and physical restraint (e.g., amniotic banding). Teratogenic maternal diseases include disorders such as diabetes mellitus and phenylketonuria. Agents that cause defects during the postem-bryonic (fetal) period are termed to have the potential for producing adverse 'fetal effects.' However, not all agents or factors that are teratogens have adverse fetal effects, and vice versa.

A simplified overview of the differences between the embryonic and fetal periods should be presented during consultation to clarify status for the patient. The period of the embryo should be described as the growth of cells that all look alike (i.e., are undifferentiated) into specialized cells that are arranged in special ways (i.e., organs, specialized tissues). These specialized cell lines or lineages grow in number and change in structure and arrangement, giving rise to organs and tissues. Some organs and tissues are formed earlier than are others. For example, the brain and spine form earlier than do the face and endocrine system. After embryogenesis (5860 days postconception) is completed, the conceptus is a fetus (Fig. 1.2). With few exceptions, the morphological architecture for a normal (or abnormal) human is laid down during the embryonic period, and these structures simply grow in size and develop normal physiologic function during the fetal period.

Congenital anomalies can be induced during the fetal period through a fetal effect, although they are usually induced during the critical embryonic period. For example, a structure that was formed normally during embryogenesis can be damaged during the fetal period, and the resulting malformation may appear to have arisen during morphogenesis. A cla.s.sic example of a fetal effect is hemorrhaging due to Coumadin exposure, which may induce brain or eye defects despite the fact that these structures were formed normally during the embryonic period.

Of all the human teratogens, thalidomide is the most notorious and heuristic example of how such agents might not be identified. In the case of thalidomide, the animal models normally used in drug screening failed to identify this drug as a dangerous substance for use during pregnancy before it was released to the market. In the human experience, it was one of the most potent teratogens ever discovered. Although laboratory studies cannot replace large, well-controlled, human epidemiologic studies, they do play an important role in screening drugs and chemicals for their potential to cause human birth defects during pregnancy. Isotretinoin (Accutane) is the only human teratogen ever discovered through laboratory research. It was known before isotretinoin was ever released on the market that this drug had a high potential for inducing congenital anomalies and pregnancy loss, and this fact was clearly displayed on the manufacturer's package insert. Unfortunately, inadvertent exposures to isotretinoin during early human pregnancy have confirmed laboratory findings. More than 100 pregnancies have been exposed to date, and a pattern of anomalies known as isotretinoin embryopathy has been observed in more than 40 percent of the offspring. Other human teratogens were discovered by astute clinicians who recognized patterns or constellations of anomalies Embryonic period (in weeks) Fetal period (in weeks) Full term 1.2.3.4.5.6.7.8.12.16.2036 38.Period of dividing zygote, Ear Ear Brain Heart CNS.

Eye Eye Heart implantation and Palate Teeth bilaminar embryo Arm Leg External genitalia Central nervous system Heart Arms Eyes Legs Human teratology principles Teeth Palate Usually not External genitalia susceptible to teratogens Ear Prenatal death Major morphological abnormalities Physiological defects and minor morphological abnormalities 5.Figure 1.2 Critical times for the development of various organs and structures. Redrawn from Airens ES, Simonis AM. De invlsed Chemischestoffen op het angebaren kind. Natuuren Technieke 1974; Critical times for the development of various organs and structures. Redrawn from Airens ES, Simonis AM. De invlsed Chemischestoffen op het angebaren kind. Natuuren Technieke 1974; 43 43.

6.Introduction to drugs in pregnancy in small clinical series of infants whose mothers used or were exposed to certain drugs or chemicals during early pregnancy. Epidemiological studies of infants whose mothers used certain drugs or chemicals during embryogenesis, as well as research with pregnant animals, have served primarily to confirm clinical observations.

Maternal complications and fetal effects due to drug or chemical exposures are not considered under the rubric of cla.s.sical teratology, but the discovery of drugs and other agents with such potential adverse effects parallels the pattern of the discovery of human teratogens.

ANIMAL STUDIES IN CLINICAL EVALUATION.

Animal models are poor predictors of whether or not a drug or chemical is teratogenic in humans. The accuracy and precision (sensitivity and specificity) of animal models in the prediction of human teratogenicity is dependent upon how close the experimental animal species is to humans. Nonhuman primates are better predictors of human teratogenicity and fetotoxicity than are rodent models because primates are genetically more closely related to humans. Animal teratology experiments are further complicated because doses used are many times greater than those given to humans, even approaching maternally toxic doses. Extremely high doses and toxic effects on the mother confound the interpretation of fetal outcome. Metabolism and absorption of drugs and chemicals are different between species because of differences in placentation, pharmac.o.kinetics, pharmacodynamics, embryonic development timing, and innate predisposi-tion to various congenital anomalies. Sensitivity and specificity of rodent studies are less than 60 percent (Schardein, 2000). Rodent animal teratology studies are undertaken by the US Food and Drug Administration (FDA) as part of an accepted drug-approval process to evaluate the safety of medications for use during human pregnancy, despite their very poor ability to predict human teratogens. Nonhuman primate teratology studies are considerably better predictors of which medications may be harmful when given during human pregnancy, with sensitivity and specificity of 90 percent or greater.

Nonhuman primate studies are, however, orders of magnitude more expensive than rodent teratology studies, and few drugs are evaluated in primates. Unfortunately, the ultimate a.s.sessment of the safety of medication use in pregnancy must come from human studies (Schardein, 2000; Shepard, 2004). Human teratogens are discovered only after numerous children have been damaged, and an astute clinician recognizes a pattern (syndrome) of congenital anomalies, and makes the link to an exposure during pregnancy. These differences are well recognized. For example, of approximately 2000 drugs and chemicals tested in animal models, 55 percent were found to have teratogenic effects (Shepard, 2004). But the number of human teratogens is approximately 50.

HUMAN STUDIES.

Human teratogens are identified through careful interpretation of data obtained from case reports, clinical series, and epidemiologic studies. A recurrent pattern of anomalies in babies who experienced similar well-defined exposures at similar points during embryogenesis are suggestive that the agent in question may be teratogenic. Case reports are important in raising causal hypotheses; however, most hypotheses are subsequently Human studies Human studies 7.proven incorrect. For example, a high incidence of environmental exposure to spermicides by pregnant women and congenital anomalies in offspring is a coincidental occurrence, despite what the legal literature states.

An example of a teratogen that was identified through epidemiologic studies, case reports, and animal studies is carabamazepine. For several decades, carabamazepine was a.s.sumed to be safer for the treatment of epilepsy during pregnancy than phenytoin or the other hydantoins. In 1993, a case report was published that reported a suicide attempt by a nonepileptic gravida during the period of spinal closure. The result was a fetus with a very large meningomyelocele (Little et al et al., 1993). In 1989, Jones et al et al. published a casecontrol study of carabamazepine and concluded that the study drug was the cause of an increased frequency of birth defects. Other epidemiologic studies throughout the 1990s were conducted, and in 2006 the a.s.sociation of neural tubes defects with carabamazepine exposure during early pregnancy is generally accepted as causal, and the risk is quantified at about 1 percent, compared to about 0.1 percent in the general population.

Quant.i.tative estimates of risks for birth defects (strength and statistical significance of a.s.sociations between agent exposures in pregnant women and abnormalities in their offspring) are obtained only through epidemiological studies. Human investigations are necessary to demonstrate that an agent is teratogenic. Unfortunately, such studies are not informative until the agent has already damaged a number of children. There are two types of epidemiology studies: cohort studies and casecontrol studies. In cohort studies the frequencies of certain anomalies in the offspring of women who are exposed are compared to the frequencies in those who are unexposed to the agent in question. A higher frequency of anomalies among exposed pregnancies indicates that the drug or agent should be scrutinized as a teratogen. In casecontrol studies the frequency of prenatal exposure to the agent is compared among children with and without a specific birth defect. If malformed children were more frequently exposed to a drug or agent than unaffected controls, then the drug or agent may be a teratogen. If an agent increases the risk of anomalies in the offspring only slightly, very large studies over a protracted period may be necessary to demonstrate that the increase is causal.

Epidemiologic studies have several limitations. Spurious a.s.sociations often occur because many epidemiologists lack medical or biological training, and fail to scrutinize their 'statistical a.s.sociations' for biological plausibility. Other confounders are sample size, investigations that involve small numbers of exposed or affected subjects, or situations in which the maternal disease or situation that led to the exposure may be responsible for an observed a.s.sociation with a congenital anomaly, rather than the agent itself.

Of paramount importance is that the observed a.s.sociation makes biological sense.

Exposures that produce malformations in the embryo should do so only during organogenesis or histogenesis. Affected structures should be susceptible to the teratogenic action of an agent only at specific gestational times. Systemic absorption of the agent by the mother and its presence at susceptible sites in the embryo or placenta should be demonstrable. Exposure to a greater quant.i.ty of the agent should be a.s.sociated in a doseresponse fashion with an increased frequency of abnormalities. Finally, a causal inference is supported if a reasonable pathogenic mechanism can be established for the observed effect. For example, lower birth weight is a.s.sociated with maternal antihypertensive therapy, but maternal hypertension is itself strongly a.s.sociated with decreased 8 8 Introduction to drugs in pregnancy birth weight. Is lower birth weight a.s.sociated with the blood pressure medication, or the disease of hypertension, or some combination?

KNOWN HUMAN TERATOGENS.

The list of known human teratogens is surprisingly small (Box 1.1). The most notorious human teratogen is thalidomide. It is currently available in the USA on a limited basis for treatment of several infectious diseases such as acquired immune deficiency syndrome (AIDS), tuberculosis, and leprosy. In 1996, a new thalidomide embryopathy epidemic was reported in Brazil and other South American countries (Castilla et al et al., 1996).

The astute reader will note that some of the putative teratogens do not fit precisely the definition of teratogen (i.e., exposure is not strictly confined to the period of organogenesis).

Box 1.1 Known human teratogens ACE inhibitors Coumarin derivatives Retinoids (oral) Amiodarone Cyclophosphamide Tetracycline derivatives Aminopterin Danazol Thalidomide Antiepileptic drugs Diethylstilbestrol Fluconazole Carbamazepine Lithium Methimazole Clonazepam Methotrexate Misoprostol Primidone Methylene blue Trimethadione, Phen.o.barbital Penicillamine paramethadione Phenytoin/fosphenytoin Quinine Trimethoprim Valproic acid Radioiodine ACE, Angiotensin converting enzyme Adapted from Schardein (2000), Shepard (2004), and Polifka and Friedman (2002).

CRITICAL TIME PERIODS.

In utero development is divided into three time periods of development: (1) preimplantation; (2) period of the embryo; and (3) time of the fetus. Exposure to drugs during pregnancy must be separated into these time periods because the conceptus responds differently in each of the three stages of development. development is divided into three time periods of development: (1) preimplantation; (2) period of the embryo; and (3) time of the fetus. Exposure to drugs during pregnancy must be separated into these time periods because the conceptus responds differently in each of the three stages of development.

Preimplantation No physiologic interface between the mother and the conceptus exists at conception (ovum penetration by the spermatid to form a single diploid cell). Traditionally, the first week postconception (until the blastocyst attaches to the wall of the uterus forming chorionic villi) was considered protected from drugs or medications that may be in the maternal circulation because there is no formal biological interface between the blastocyst and the mother. However, recent evidence (e.g., mitomycin) indicates that the preimplantation embryo may not be as protected as previously thought.

Potential adverse effects 9.Embryonic development The most critical stage of development for the induction of birth defects is the period of the embryo. The period of the embryo extends the time of implantation until 5860 days postconception. The organs and tissues of the unborn baby are being formed (i.e., organogenesis) during this period. Mistakes which occur during the period of the embryo result in malformations (congenital anomalies) and are called birth defects. Teratogens are agents that cause abnormal embryonic physical or physiological development by acting during the period of the embryo, or organogenesis (Jones, 1988). Malformations lethal to the embryo present as spontaneous abortion, sometimes before pregnancy is recognized. Similarly, some substances that are directly toxic to the embryo, e.g., methotrexate, also present as spontaneous abortions. The critical times for the development of various organs and structures of the human embryo are given in Fig. 1.2 (p. 5).

Fetal development Important changes occur during the embryonic development that can also be damaged outside the period of the embryo. Traditionally, things that happened to a fetus were not considered a teratogenic effect, but some authorities have begun lumping fetal effects into this category. Changes in cellular structures such as the brain cell arrangements during neuronal migration occur during the fetal period. However, the predominant fetal event is hyperplastic growth (increase in cell number) with organs and other tissues becoming larger through cellular proliferation, and only secondarily through hypertrophy. An important example is the thyroid, which appears early in the fetal period, as does fetal endocrine function. Most of the potential adverse effects during fetal development are maldevelop-ment due to interrupted cell migration and growth r.e.t.a.r.dation (Jones, 1988). If blood flow to an organ or structure is interrupted or obstructed, structures that were normally formed during embryogenesis may be malformed during the fetal period (e.g., vascular disruption and fetal cocaine or warfarin exposure). The structure deprived of blood flow would undergo necrosis and be resorbed. This would produce a defect that may mimic an embryonic effect. However, the true origin of the defect would be fetotoxicity.

The embryo and fetus are exposed to drugs through the placenta which can: (1) metabolize certain drugs before they reach the conceptus; (2) allow 99 percent of drugs to cross by simple diffusion; (3) not transport large molecules (i.e., larger than 1000 molecular weight), unless there is an active transport system (e.g., antibodies); (4) transport neutrally charged molecules; (5) easily transport lipid-soluble drugs; and (6) not transport charged (+ or ) molecules. Poor potential for transfer back to the maternal circulation occurs for some drugs (e.g., water-soluble drugs transfer back to the mother's circulation poorly), resulting in acc.u.mulation in the embryofetal compartment.

POTENTIAL ADVERSE EFFECTS.

Spontaneous abortion As many as 50 percent of early pregnancies (058 days) end in spontaneous abortion.

Recent findings from in vitro in vitro fertilization studies suggest that the majority of these fertilization studies suggest that the majority of these 10 10 Introduction to drugs in pregnancy spontaneous abortuses are chromosomally abnormal. The risk of spontaneous abortion is 1520 percent among fetuses surviving 59126 days of gestation. The risk of spontaneous abortion/fetal death decreases to 12 percent by 1820 weeks (127140 days). Up to 28 weeks (196 days) postconception the risk for spontaneous abortion is approximately 2 percent.

Congenital anomalies The frequency of congenital anomalies detected at birth is approximately 3.55 percent (Brent and Beckman, 1990). This figure is thought to underrepresent the true frequency of anomalies by as much as twofold because 100 percent detection of anomalies is not usually reached until about 5 years of age. The frequency of congenital anomalies is sev-eralfold higher among stillbirths and miscarriages than live births, and is especially high among early (i.e., first-trimester) miscarriages.

Fetal effects Fetal effects are of four primary types: (1) damage to structures or organs that are formed normally during embryogenesis; (2) damage to systems undergoing histogenesis during the fetal period; (3) growth r.e.t.a.r.dation; or (4) fetal death or stillbirth. Any or all of these fetal effects can occur concomitantly. Fetal effects may be caused by a teratogen, but may also be caused by agents that have no apparent potential to produce abnormal embryonic development. Organs, structures, or functions formed normally during embryogenesis can be damaged by some environmental exposures during the fetal period.

Fetal growth r.e.t.a.r.dation is the most frequently observed effect of agents given during pregnancy and outside the period of embryogenesis. Sometimes it is difficult to distinguish between the effects of the agents from those of the disease ent.i.ty being treated.

Propranolol, for example, is a.s.sociated with fetal growth r.e.t.a.r.dation, but the maternal disease for which the drug is given (hypertension) is also a.s.sociated with fetal growth r.e.t.a.r.dation in the absence of antihypertensive therapy. Some agents that are teratogenic may be a.s.sociated with fetal growth r.e.t.a.r.dation. Fetal growth r.e.t.a.r.dation may also occur without embryonic damage. Risks of fetal death, stillbirth, and other adverse effects are increased with exposure to some agents during pregnancy (Table 1.1).

Neonatal and postnatal effects Prenatal exposure to some drugs is a.s.sociated with adverse neonatal effects, such as difficulty in adaptation to life outside the womb. Drugs a.s.sociated with adverse neonatal are not usually a.s.sociated with teratogenic effects. Transient metabolic abnormalities, withdrawal, and hypoglycemia are well-doc.u.mented neonatal effects of certain medications and nonmedical drugs. Examples of other adverse neonatal effects are the floppy infant syndrome with the use of benzodiazepines near term, patent ductus arteriosus with the use of prostaglandin synthestase inhibitors (non-steroidal anti-inflammatory agents or NSAIDs) such as aspirin or indomethacin, and gray baby syndrome with high-dose chloramphenicol near the time of delivery (Table 1.1). Developmental delay is frequently a.s.sociated with the action of teratogens, but is also observed in a.s.sociation with the fetal effects of drugs that are apparently not teratogenic.

Potential adverse effects 11.Table 1.1 Adverse effects other than birth defects on the human fetus a.s.sociated with drugs Adverse effects other than birth defects on the human fetus a.s.sociated with drugs Maternal medication Fetal/neonatal effect Acetaminophen Renal failure Adrenocortical hormones Adrenocortical suppression; electrolyte imbalance Alcohol Muscular hypotonia: hypoglycemia (?); withdrawal; intrauterine growth restriction (IUGR); blood changes; affect mental ability Alphaprodine Platelet dysfunction Amitriptyline Withdrawal Ammonium chloride Acidosis Amphetamines Withdrawal Antihistamines Infertility(?) Antineoplastics Transient pancytopenia IUGR Ant.i.thyroid drugs Hypothyroidism Barbiturates/diphenylhydantoin Coagulation defects; withdrawal (barbiturates only); IUGR Chloral hydrate, excess Fetal death Chloramphenicol Death ('gray baby syndrome') Chlordiazepoxide Withdrawal(?) Chloroquine Death(?) Chlorpropamide Prolonged hypoglycemia; fetal death Cocaine Vascular disruption, withdrawal, IUGR Coumarin anticoagulants Hemorrhage, death, IUGR Diazepam Hypothermia; hypotonia; withdrawal Diphenhydramine Withdrawal Ergot Fetal death Erythromycin Liver damage(?) Gold salts Complications; kernicterus Glutethimide Withdrawal Heroin/morphine/methadone Withdrawal; neonatal death Hexamethonium bromide Neonatal ileus Hykinone Blood changes; jaundice Immunosuppressants Transient immune system depression, danger of infection Insulin (shock) Fetal loss Intravenous fluids, excess Fluid and electrolyte abnormalities Iophenoxic acid Evaluation of serum protein-bound iodine (PBI) Lithium Cyanosis, flaccidity, polyhydramnios, toxicity Magnesium sulfate Central depression and neuromuscular block Meperidine Neonatal depression Mepivacaine Fetal brachycardia and depression Meprobamate r.e.t.a.r.ded development(?) Nitrofurantoin Hemolysis Novobiocin Hyperbilirubinemia(?) Oral progestogens, androgens, Advanced bone age and estrogens Phenformin Lactic acidosis(?) Phen.o.barbital, excess Neonatal bleeding; death Phenothiazines Hyperbilirubinemia(?), depression, hypothermia(?), withdrawal Polio vaccine, live Fetal loss(?) continued 12.Introduction to drugs in pregnancy Table 1.1 Continued Continued Maternal medication Fetal/neonatal effect Prednisolone Acute fetal distress, fetal death (?) Primaquine, pentaquine Hemolysis(?) Primidone Withdrawal(?) Propoxyphene Withdrawal Quinine Thrombocytopenia Reserpine Nasal congestion, lethargy, respiratory depression, brachycardia Salicylates, excess Bleeding, fetal death Sedatives Behavioral changes Smoking Premature births, IUGR, perinatal loss(?) Sulfonamides Kernicterus(?), anemia(?) Tetracyclines Deposition in bone, inhibition of bone growth in premature infants, discoloration of teeth Thiazide diuretics Thrombocytopenia, salt and water depletion, neonatal death(?) Thioureas Blood changes, affect mental ability Tolbutamide Thrombocytopenia, fetal death Vaccinations Fetal vaccinia Verapamil Transient fetal-neonatal cardiovascular Vitamin K a.n.a.logs, excess Hyperbilirubinemia Adapted from Schardein, 2000.

MATERNAL PHYSIOLOGY DURING PREGNANCY.

Profound physiological changes occur during pregnancy. Maternal enzymes, particularly cholinesterases (Pritchard, 1955), have lowered activity. Maternal blood volume increases dramatically during pregnancy, by perhaps 4050 percent, to support the requirements of the developing fetus (Cunningham et al et al., 2001). Distribution of drugs in this increased blood volume may lower serum concentrations. Absorption of drugs occurs with about the same kinetics as in the nonpregnant adult; however, renal clearance is increased and enzyme activity is downregulated. Decreased enzyme activity levels are exacerbated somewhat by the increased blood volume, decreasing the overall effective serum concentration of a given dose. In turn, increased renal output may effect an increased clearance index for most drugs. Drugs that are tightly bound to the serum proteins have little opportunity to cross the placenta or enter breast milk. Consequently, increased demands are placed on cardiovascular, hepatic, and renal systems. In addition, the gravid uterus is vulnerable to a variety of effects not present in the nonpregnant state, such as hemorrhage, rupture, or preterm contraction.

Increased demands imposed on these physiological systems by pregnancy may, under normal conditions, be dealt with in an uncomplicated manner. However, conditions of disease or other stress weaken these key systems and they may be unable to function normally. For example, cocaine abuse during pregnancy actually targets these key systems that are already stressed from the gravid state of the woman.

Hence, it would be expected that cocaine use during pregnancy would place cardiovascular, renal, and hepatic systems at greater risk than those of the nonpregnant Pharmac.o.kinetics in pregnancy Pharmac.o.kinetics in pregnancy 13.adult. Indeed, these expectations are borne out in the observations of cocaine use during pregnancy.

PHARMAc.o.kINETICS IN PREGNANCY.

The quant.i.ty of pharmac.o.kinetic data during pregnancy is extremely limited. Only two investigations examined for this review made explicit quant.i.tative recommendations for dose or schedule during pregnancy (Caritis et al et al., 1989; Wisner et al et al., 1993). Frequently, results are conflicting between studies of the same drug. Across all investigations reviewed, area under the curve was decreased in 41 percent of the studies, volume of distribution was increased in 30 percent, and peak plasma concentration was decreased in 34 percent. Steady-state plasma concentration was decreased in 44 percent of the studies, as was half-life in 41 percent. Clearance was increased in 55 percent of the studies (Table 1.2).

No general statement about pharmac.o.kinetic changes during pregnancy can be made.

The individual drug must be considered. These changes in pharmac.o.kinetics cause decreases in drug plasma concentrations. When pharmac.o.kinetic data are altered in this way, increased doses or schedules are needed to maintain effective systemic drug levels.

However, this summary information is biased by a lack of information on many therapeutic agents used during pregnancy and because some drugs are represented more than once among the investigations reviewed. Still, the physiologic changes during pregnancy and their effects on the disposition of medications given during gestation found in this review are consistent with previous surveys of the literature (Amon and Huller, 1984a, b; c.u.mmings, 1983; Kafetzis et al et al., 1983; Mattison et al et al., 1992; Philipson 1978; Reynolds 1991). Multiple confounders make it difficult to interpret available pharmac.o.kinetic data in pregnancy. Many studies have had very small sample sizes, frequently fewer than 10 pregnant women. Comparison groups have varied in composition. Studies have used nonpregnant women, adult males, the same patients 68 weeks postpartum, or published pharmac.o.kinetic data. None of the studies reviewed gave maternal weight-Table 1.2 Pharmac.o.kinetics in pregnancy Index Studies reporting pharmac.o.kinetic data changes a.s.sociated with pregnancy n Decrease No change Increase Studies not reporting pharmac.o.kinetic data (%) AUC.

17.7.5.5.44 (72.1).

V.

23.3.11.9.38 (62.3).

d C.

30.10.17.3.31 (50.8).

max C.

42.19.4.19.19 (31.1).

ss t 39.16.17.6.22 (36.1).

1/2.

t 9.3.4.2.52 (85.2).

max Cl 44.5.15.24.17 (27.9).

PPB.

7.6.1.0.

54 (88.5).