The New England Journal of Medicine
`
`Review Article
`
`Drug Therapy
`
`A
`
`L A S T A I R
`
` J.J. W
`
`O O D
`
`, M.D.,
`
` Editor
`
`D
`
`RUGS
`
`
`
`IN
`
` P
`
`REGNANCY
`
`G
`
`IDEON
`
` P
`, M.D., A
` K
`ASTUSZAK
`NNE
`OREN
` S
` I
`, M.D.
`AND
`HINYA
`TO
`
`, M.S
`
`C
`
`.,
`
`B
`
`EFORE marketing a new drug, the manufac-
`turer almost never tests the product in preg-
`nant women to determine its effects on the fe-
`tus. Consequently, most drugs are not labeled for
`use during pregnancy. Typically, descriptions of drugs
`
`Physicians’ Desk Reference and
`that appear in the
`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 out-
`weigh the risks to the fetus. These typical disclaim-
`ers, 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,
` and every year, hundreds of
`1
`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
` Another reason is that with
`wanted pregnancies.
`2,3
`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
`
`From the Motherisk Program, Division of Clinical Pharmacology and
`Toxicology (G.K., A.P., S.I.), the Departments of Pediatrics and Population
`Health Sciences (G.K., S.I.), and the Research Institute (G.K., S.I.), Hos-
`pital for Sick Children; and the Departments of Pediatrics (G.K., S.I.),
`Pharmacology (G.K., S.I.), and Medicine (G.K.), University of Toronto —
`both in Toronto. Address reprint requests to Dr. Koren at the Division of
`Clinical Pharmacology, Hospital for Sick Children, 555 University Ave.,
`Toronto, ON M5G 1X8, Canada.
`©1998, Massachusetts Medical Society.
`
`1128
`
`ⴢ
`
`April 16, 1998
`
`before pregnancy. Furthermore, for pregnant wom-
`en with certain conditions once believed to be in-
`compatible 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 wom-
`en and their families.
`
`HUMAN TERATOGENESIS
`Teratogenesis is defined as the dysgenesis of fetal
`organs as evidenced either structurally or function-
`ally (e.g., brain functions).
` The typical manifesta-
`5
`tions of teratogenesis are restricted growth or death
`of the fetus, carcinogenesis, and malformations,
` de-
`6
`fined 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 demonstrat-
`ing that fetal exposure to the drug during critical pe-
`riods of development resulted in severe limb defects
`and other organ dysgenesis (e.g., kidney and heart
` Despite the high rates of malformations
`defects).
`6,7
`(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 pathogno-
`monic, 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
`
`Downloaded from www.nejm.org on March 4, 2007 . Copyright © 1998 Massachusetts Medical Society. All rights reserved.
`
`Page 1 of 10
`
`CELGENE EXHIBIT 2013
`Coalition for Affordable Drugs VI LLC (Petitioner) v. Celgene Corporation (Patent Owner)
`Case IPR2015-01103
`
`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`DRUG THERAPY
`
`in clinically effective doses, and even fewer are cur-
`rently in clinical use (Table 1). Many other com-
`monly used drugs, including salicylates, glucocorti-
`coids, and spermicides, were once thought to be
`teratogenic but have been shown to be safe in sub-
`sequent studies that were larger and better con-
`trolled 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 antihista-
`mine (doxylamine) and pyridoxine, was the most
`widely used medication in the United States for nau-
`sea 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 with-
`drawn from the market by its manufacturer in 1982,
`which left millions of pregnant women without a
`drug approved by the Food and Drug Administra-
`
`tion (FDA) for the treatment of nausea and vomit-
`ing. The rate of hospitalization for severe nausea and
`vomiting during pregnancy increased by a factor of
`2 in both the United States and Canada after Ben-
`dectin was withdrawn from the market (Fig. 1).
`The drug was withdrawn despite a substantial
`body of evidence that the rate of major malforma-
`tions among the children of women who had re-
`ceived Bendectin during pregnancy did not differ
`from the rate in the general population.
` With-
`24,25
`drawal of the drug from the American market did
`not decrease the rate of any specific category of mal-
`formation, as would be expected for a truly terato-
`genic 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 un-
`der the trade name Diclectin. A review committee has
`advised the Canadian Minister of Health that the
` A recent study revealed that severe nau-
`drug is safe.
`27
`sea and vomiting of pregnancy often lead women to
`terminate or consider the termination of otherwise
`wanted pregnancies.
` Other formulations of dox-
`28
`
`T
` 1.
`ABLE
`
` D
`
`
`RUGS
`
`WITH
`
` P
` T
`ROVEN
`
` E
`
`ERATOGENIC
`FFECTS
`
` H
`.*
`IN
`UMANS
`
`D
`RUG
`
`T
`ERATOGENIC
`
` E
`FFECT
`
`Aminopterin†, methotrexate
`Angiotensin-converting–enzyme inhibitors
`
`Anticholinergic drugs
`Antithyroid drugs (propylthiouracil and meth-
`imazole)
`Carbamazepine
`Cyclophosphamide
`Danazol and other androgenic drugs
`Diethylstilbestrol†
`
`CNS and limb malformations
`Prolonged renal failure in neonates, decreased skull
`ossification, renal tubular dysgenesis
`Neonatal meconium ileus
`Fetal and neonatal goiter and hypothyroidism, apla-
`sia 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‡, necrotizing
`enterocolitis
`Facial and CNS defects
`Paramethadione†
`Growth retardation, CNS deficits
`Phenytoin
`Neonatal withdrawal syndrome when drug is taken
`Psychoactive drugs (e.g., barbiturates, opioids,
`in late pregnancy
`and benzodiazepines)
`Systemic retinoids (isotretinoin and etretinate) CNS, craniofacial, cardiovascular, and other defects
`Tetracycline
`Anomalies of teeth and bone
`Thalidomide
`Limb-shortening defects, internal-organ defects
`Trimethadione†
`Facial and CNS defects
`Valproic acid
`Neural-tube defects
`Warfarin
`Skeletal and CNS defects, Dandy–Walker syndrome
`
`Hypoglycemic drugs
`Lithium
`Misoprostol
`Nonsteroidal antiinflammatory drugs
`
`*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.
` CNS denotes central nervous system.
`8
`†The drug is not currently in clinical use.
`‡Sulindac probably does not have this effect.
`
`Downloaded from www.nejm.org on March 4, 2007 . Copyright © 1998 Massachusetts Medical Society. All rights reserved.
`
`Volume 338 Number 16
`
`ⴢ
`
`1129
`
`Page 2 of 10
`
`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The New England Journal of Medicine
`
` 2.
`ABLE
`
` C
`
` D
` I
` T
`
`OMMON
`RUGS
`NITIALLY
`HOUGHT
`
`TO
`
` B
` T
`
`E
`ERATOGENIC
`
` S
` P
` S
`.
`BUT
`UBSEQUENTLY
`ROVED
`AFE
`
`T
`
`D
`RUG
`
`I
`NITIAL
`
` E
`VIDENCE
`
`
`
`OF
`
` R
`ISK
`
`S
`UBSEQUENT
`
` E
`VIDENCE
`
`
`
`OF
`
` S
`AFETY
`
`Diazepam*
`
`Oral clefts
`9
`
`Oral contraceptives
`
`Spermicides
`
`Salicylates
`
`Birth defects involving the vertebrae, anus,
`heart, trachea, esophagus, kidney, and
`limbs
`; masculinizing effects on female fe-
`13
`tuses resulting in pseudohermaphroditism
`14
`Limb defects, tumors, Down’s syndrome,
`and hypospadias
`
`17
`Cleft palate
` and congenital heart disease
`19
`
`Bendectin (doxylamine
`plus pyridoxine)
`
`Cardiac and limb defects
`22,23
`
`
`
`No increase in risk in large cohort and case–
`control studies
`
`10-12
`No association between first-trimester expo-
`sure to oral contraceptives and malforma-
`tions in general or external genital mal-
`
`formations in two meta-analyses
`15,16
`No increase in risk in a meta-analysis
`18
`
`No increase in risk in large cohort
`studies
`20,21
`No increase in risk in two meta-analyses
`
`24,25
`
`*Diazepam taken near term may cause the neonatal withdrawal syndrome or cardiorespiratory instability.
`
`this belief became evident after isotretinoin was in-
`troduced in North America in the early 1980s for
`the treatment of acne. For years before its clinical in-
`troduction, this drug had been known to cause mal-
`formations in animals.
` Despite explicit warning la-
`29
`bels, scores of children with retinoid embryopathy
`were born in the years after the drug was intro-
`duced.
` Such warnings are not sufficient, because
`30
`women taking isotretinoin may not plan their preg-
`nancies, or their birth-control methods may fail. In
`addition, some women and men are functionally il-
`literate, 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 Pre-
`vention Program includes explicit and detailed print-
`ed warnings as well as a line drawing of a malformed
`child,
` and as part of the program, women are asked
`32
`to sign a consent form indicating that they agree to
`use two effective methods of contraception before
`therapy is started. Since the program was implement-
`ed 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 cog-
`nizant of the high fetal risk.
` Many of these women
`32
`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 contracep-
`tive methods.
`33
`
`CURRENT TRENDS IN PREVENTING FETAL
`EXPOSURE TO TERATOGENS
`The advent of effective injectable hormonal con-
`traceptives has made it possible to minimize the risk
`of an unplanned pregnancy during therapy with a
`known teratogen. This approach was first implement-
`ed in South America, where sexually active women
`with cutaneous leprosy were injected with medroxy-
`
`ylamine in combination with pyridoxine are available
`in other countries (e.g., South Africa, Spain, and
`Thailand).
`
`Isotretinoin
`The experience with thalidomide led drug regula-
`tors, drug manufacturers, and the medical commu-
`nity to believe that appropriate labeling of terato-
`genic drugs, with warnings not to take them around
`the time of conception, would be effective in pre-
`venting fetal exposure to the drugs. The naiveté of
`
`Prescriptions (⫻10⫺3)
`
`140
`
`120
`
`100
`
`80
`
`60
`
`40
`
`02
`
`0
`
`Hospitalization
`
`Bendectin
`
`Diclectin
`
`20
`
`15
`
`10
`
`5
`
`Hospitalizations (per 1000 live births)
`
`0
`1979
`
`1980
`
`1981
`
`1982
`
`1983
`
`1986
`
`1987
`
`1988
`
`1989
`
`1984
`1985
`Year
`
`Figure 1.
` Rates of Hospitalization among Pregnant Women
`with Severe Nausea and Vomiting and Numbers of Prescrip-
`tions for Bendectin and Diclectin in North America, 1979
`through 1989.
`Bendectin was withdrawn from the U.S. market in 1982, where-
`as Diclectin, the same drug, remained on the market in Cana-
`da. Adapted from Neutel and Johansen with the permission of
`the publisher.
`
`26
`
`1130
`
`ⴢ
`
`April 16, 1998
`
`Downloaded from www.nejm.org on March 4, 2007 . Copyright © 1998 Massachusetts Medical Society. All rights reserved.
`
`Page 3 of 10
`
`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`DRUG THERAPY
`
`progesterone before receiving a prescription for tha-
`lidomide.
` Yet numerous new cases of thalidomide-
`34
`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 pre-
`vent 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 contra-
`ception are likely to fail.
` Implants should be con-
`35
`sidered 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., alco-
`hol or cocaine). Furthermore, women taking terato-
`genic drugs who are not sexually active should be in-
`formed 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 an-
`imals 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 oc-
`currence of thalidomide-associated embryopathy led
`to the erroneous belief that human teratogenicity
`could not be predicted on the basis of studies in an-
`imals. However, every drug that has since been
`found to be teratogenic in humans has caused simi-
`lar 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 an-
`imals probably prevented a disaster similar to that of
`thalidomide.
`29
`However, there are drugs that have teratogenic ef-
`fects in animals when administered in high doses
`that are not teratogenic in humans given clinically
`relevant doses. For example, high doses of gluco-
` or benzodiazepines
` can cause
`corticoids
`19,70-75
`76,77
`oral clefts in animals, but clinically relevant doses in
`humans have no such effects.
` Similarly, sa-
`10-12,75
`
`licylates
` cause cardiac malformations in animals
`78-80
`but not in humans.
` Such discrepancies have led
`20,21
`to unwarranted anxiety on the part of women, their
`families, and physicians and may have contributed to
`unnecessary terminations of pregnancies.
` Although
`3
`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 terato-
`genicity and to assess the relation between drug ex-
`posure and fetal outcome. The first accounts of ad-
`verse fetal outcomes after exposure to a marketed
`drug are usually published in the form of case re-
`ports. These reports can be either very useful or use-
`less 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
`) or causes a rare malfor-
`30
`mation (e.g., ear agenesis
`), then a small number of
`81
`cases can establish a strong association. Warfarin,
`9
`diethylstilbestrol,
` and isotretinoin
` were originally
`82
`83
`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 tha-
`lidomide). To 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 terato-
`genicity.
`Epidemiologic studies are typically designed to de-
`termine whether mothers who took a specific drug
`during pregnancy have a larger number of mal-
`formed children than mothers who did not (cohort
`studies) or whether mothers of children with a spe-
`cific malformation took the drug more often than
`mothers of children without the malformation (case–
`control studies).
`With the international development of teratology-
` a new source of data for pro-
`information services,
`84
`spective observational research has emerged. Preg-
`nant women taking prescription or over-the-counter
`drugs voluntarily call these centers for risk-assess-
`ment 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.
`
`Volume 338 Number 16
`
`ⴢ
`
`1131
`
`Downloaded from www.nejm.org on March 4, 2007 . Copyright © 1998 Massachusetts Medical Society. All rights reserved.
`
`Page 4 of 10
`
`

`
`The New England Journal of Medicine
`
`T
`
` 3.
`ABLE
`
` T
` E
`ERATOGENIC
`FFECTS
`
`
`
` D
`
`OF
`RUGS
`
`IN
`
` ANIMALS AND HUMANS.*
`
`DRUG
`
`EFFECTS IN ANIMALS
`
`EFFECTS IN HUMANS
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Stillbirths and increased fetal loss in
`sheep and rabbits37
`
`Prolonged renal failure and hypotension in the newborn, decreased skull ossification, hypo-
`calvaria, and renal tubular dysgenesis38
`
`Neural-tube defects 40
`
`Growth retardation involving weight, length, and head circumference43; placental
`abruption44,45 and uterine rupture
`
`Fetal alcohol syndrome: prenatal and postnatal growth deficiency, CNS anomalies (micro-
`cephaly, behavioral abnormalities, and mental retardation), characteristic pattern of facial
`features (short palpebral fissures, hypoplastic philtrum, and flattened maxilla), and major
`organ-system malformations49; with age, facial features may become less distinctive, but
`short stature, microcephaly, and behavioral abnormalities persist50
`Retinoid embryopathy resulting in some or all of the following abnormalities30: CNS defects
`(hydrocephalus, optic-nerve blindness, retinal defects, microphthalmia, posterior fossa de-
`fects, and cortical and cerebellar defects); craniofacial defects (microtia or anotia, low-set
`ears, hypertelorism, depressed nasal bridge, microcephaly, micrognathia, and agenesis or
`stenosis of external ear canals); cardiovascular defects (transposition of great vessels, te-
`tralogy of Fallot, and ventricular or atrial septal defects); thymic defects (ectopia and hy-
`poplasia or aplasia); and miscellaneous defects (limb reduction, decreased muscle tone,
`spontaneous abortion, and behavioral abnormalities)
`Ebstein’s anomaly and other heart defects52,53
`Fetal Minamata disease: diffuse neuronal disintegration with gliosis, cerebral palsy, micro-
`cephaly, strabismus, blindness, speech disorders, motor impairment, abnormal reflexes,
`and mental retardation56
`
`Fetal hydantoin syndrome60: prenatal and postnatal growth deficiency, motor or mental de-
`ficiency, short nose with broad nasal bridge, microcephaly, hypertelorism, strabismus,
`epicanthus, wide fontanelles, low-set 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 intelligence61
`Limb-shortening defects,63 loss of hearing, abducens paralysis, facial paralysis, anotia, micro-
`tia, renal malformations, congenital heart disease
`Neural-tube defects 66,67
`Fetal warfarin syndrome: skeletal defects (nasal hypoplasia and stippled epiphyses), limb hy-
`poplasia (particularly in distal digits), low birth weight (⬍10th percentile), hearing loss,
`and ophthalmic anomalies 69; CNS defects with exposure after first trimester; dorsal mid-
`line dysplasia (agenesis of corpus callosum and Dandy–Walker malformations) or ventral
`midline dysplasia (optic atrophy) 6
`
`Angiotensin-con-
`verting–enzyme
`inhibitors
`Carbamazepine
`
`Cocaine
`
`Ethanol
`
`Cleft palate, dilated cerebral ventricles,
`and growth retardation in mice 39
`Dose-dependent decrease in uterine
`blood flow, fetal hypoxemia, hyper-
`tension, and tachycardia in sheep41;
`reduced fetal weight, fetal edema, and
`increased resorption in rats and mice42
`Microcephaly, growth deficiency, and
`limb anomalies in dogs, chickens,
`and mice46-48
`
`Isotretinoin
`
`CNS, head, limb, and cardiovascular
`defects in rats and rabbits29
`
`Lithium
`Methyl mercury
`
`Phenytoin
`
`Heart defects in rats51
`CNS abnormalities in rats 54; growth re-
`tardation, motor disturbances, mi-
`croencephaly, and brain lesions in
`rhesus monkeys55
`Cleft palate, micromelia, renal defects,
`and hydrocephalus in rabbits, mice,
`and rats57-59
`
`Thalidomide†
`
`Valproic acid
`Warfarin†
`
`Limb-shortening defects in rabbits
`(most sensitive species)62
`Exencephaly in hamsters and mice 64,65
`Maxillonasal hypoplasia and skeletal
`anomalies in rats68
`
`*CNS denotes central nervous system.
`†Initial studies in animals failed to show teratogenicity; hence, documentation in humans preceded that in animals.
`
`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 limit-
`ed in size, meta-analyses of studies of similar design
`are becoming more frequent. A detailed, stepwise
`methodologic approach to meta-analysis of terato-
`logic studies has been described.
` The appropriate
`24
`use of this approach depends to a large extent on es-
`tablishing 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, be-
`cause it is becoming clear that the long-term effects
`of teratogenic drugs on neurobehavioral develop-
`
`ment can have a more devastating effect on children
`and their families than structural anomalies. To date,
`several drugs have been shown to affect brain devel-
`opment, including carbamazepine, isotretinoin, phen-
`ytoin, valproic acid, and warfarin (Table 1). Carba-
`mazepine 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 de-
`velopment, but recent studies have suggested that
`even phenotypically normal children may have ab-
`normal neurodevelopment.
` Warfarin was initially
`87
`associated with chondrodysplasia punctata and men-
`tal retardation and has subsequently been found to
`cause the Dandy–Walker brain malformation in an
`estimated 1 to 2 percent of exposed fetuses.
`6,69
`
`1132
`
`ⴢ
`
`April 16, 1998
`
`Downloaded from www.nejm.org on March 4, 2007 . Copyright © 1998 Massachusetts Medical Society. All rights reserved.
`
`Page 5 of 10
`
`

`
`DRUG THERAPY
`
`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 popu-
`lation is 3 percent, then at least 220 pregnancies
`with the specific exposure and a similar number of
`control pregnancies 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 condi-
`tions 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 under-
`lying 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-the-counter drug.89 Women who have given
`birth to malformed children may be more likely to
`remember the course of their pregnancies, in the ef-
`fort to understand what went wrong, than women
`who have given birth to healthy children, thus giv-
`ing rise to false positive associations. The initial sug-
`gestions that benzodiazepines, spermicides, and Ben-
`dectin, for example, were teratogenic were based on
`retrospective case–control studies subsequently re-
`futed by other, larger studies (Table 2).
`With improved epidemiologic methods, the reli-
`ability of the case–control design has improved. For
`example, recruiting mothers of infants with a differ-
`ent 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 caus-
`es spina bifida) on the information they reported.90
`
`Nonrandomized Observational Studies
`With prospective observational studies, the treat-
`ment decisions have not been made by the investi-
`gators collecting the data. As a result, the indica-
`tions 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 con-
`founding factors are not randomly distributed be-
`tween 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 develop-
`ment require longer follow-up than observational
`studies of other abnormalities, and interpretation of
`the results is often complicated by numerous con-
`founding factors. Maternal and paternal IQ, socio-
`economic 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 re-
`ports. 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 manu-
`facturers are likely to report adverse fetal outcomes,
`not uneventful ones. For example, the pivotal study
`that described retinoid embryopathy contained two
`parts: prospectively collected data from a study co-
`hort, with a malformation rate of 36 percent, and data
`from voluntary retrospective reporting to the manu-
`facturer, with a malformation rate of 80 percent.30
`
`Meta-Analyses
`A common concern regarding the use of meta-
`analysis 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 neg-
`ative studies (i.e., those that do not reject the null
`hypothesis) are less likely to be published than posi-
`tive 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 malforma-
`tions, which is in the range of the teratogenic risk as-
`sociated with thalidomide.2 After counseling, this es-
`
`Volume 338 Number 16
`
`ⴢ 1133
`
`Downloaded from www.nejm.org on March 4, 2007 . Copyright © 1998 Massachusetts Medical Society. All rights reserved.
`
`Page 6 of 10
`
`

`
`The New England Journal of Medicine
`
`timate was substantially reduced, thereby preventing
`numerous terminations of otherwise wanted preg-
`nancies.2,3 The same women correctly estimated the
`risk of major malformations in the general popula-
`tion (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? Nu-
`merous lay publications misinform women by as-
`signing risks to drugs not known to be teratogenic
`in humans.2,3 Women often report that their physi-
`cians have encouraged them to terminate otherwise
`wanted pregnancies just to be on the safe side, sug-
`gesting 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 situa-
`tion and specific circumstances. For example, wom-
`en with epilepsy that has been treated effectively
`with phenytoin since their childhood may be glad to
`hear that although the drug is teratogenic, the over-
`all 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 chil-
`dren 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