Abstract
Teratogenic exposures are those that can cause an embryo or fetus to develop abnormally. Several factors determine whether an agent is teratogenic, including the gestational timing of the exposure, as well as the dose, route and nature of the agent itself. We review the general concepts of teratogenesis, as well as known genetic susceptibilities to teratogenic effects, with a special focus on antiepileptic drugs. We discuss general principles of risk counselling and risk reduction, and we describe several long-known teratogens, as well as several exposures recognized only recently to have teratogenic potential.
Case
A 27-year-old woman with epilepsy and her husband are planning to have a child but are concerned that their baby may be harmed by the mother's epilepsy or its treatment during pregnancy. The woman is otherwise healthy, and her seizures are well controlled with oral divalproex, 900 mg per day. What information should the couple be given? What course of action should be recommended?
Teratology is the study of abnormal prenatal development. A teratogenic exposure is one that can cause an embryo or fetus to develop abnormally. Teratogenic exposures can affect prenatal development by altering gene expression, programmed cell death (apoptosis), cell migration or proliferation, histogenesis, synthesis or function of proteins or nucleic acids, or the supply of energy sources. Some teratogenic exposures act directly on the embryo, whereas others act through intermediates produced by maternal metabolism.
The nature of an agent is an important factor in determining whether an exposure will damage a developing embryo, and certain chemicals such as thalidomide and isotretinoin do have a greater teratogenic potential than others. Nevertheless, it is inappropriate to classify some agents as teratogenic and others as not teratogenic, because teratogenicity depends on the gestational timing and on the dose and route of exposure, as well as on the nature of the agent itself. The presence of other concurrent exposures and the biological susceptibility of the mother and fetus are also important factors that determine whether or not an exposure will be teratogenic. For example, the occasional use of one or 2 ASA tablets during early pregnancy is unlikely to be harmful to the embryo or fetus; however, taking ASA during this period of time may not be safe if the dose is so great that it causes coma in the mother.
One of the most important factors affecting safety is the stage of pregnancy during which an exposure occurs. During the first 2 weeks after conception, toxic exposures are unlikely to cause malformations, because the cells of the conceptus are pluripotent at this stage. Cells that are killed by an exposure can be replaced by other cells. However, if too many cells are damaged or die, the embryo will not survive. The first 2 weeks after conception are sometimes referred to as the “all-or-none” period, because toxic exposures during this time usually kill the embryo or produce no permanent effect if the embryo survives.1 However, some mutagenic treatments during the preimplantation period produce malformations in rodent experiments, so the “all-or-none” rule does not always hold.2
Organogenesis (18–60 days post conception in humans) is the time during which the embryo is most sensitive to many teratogenic exposures and when most structural anomalies are produced. The fetal period is marked by rapid growth and maturation as well as active cell growth, proliferation and migration, particularly in the central nervous system. Teratogenic exposures during this period may cause fetal growth retardation, death or central nervous system dysfunction that may not be apparent until later childhood. Some treatments with long-recognized teratogenic potential and the stage of pregnancy during which they exert their effects are summarized in Table 1. Several treatments that have only been recognized as having teratogenic potential in the last few years are listed in Table 2.3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21 Many of these teratogenic effects have been recognized in humans not because of their frequency of occurrence but because of the characteristic pattern of anomalies that is produced in exposed children. Fig. 1 shows a child who has minor facial anomalies consistent with methimazole embryopathy (see Table 2 for details of this syndrome).
Dose is a critical feature of any teratogenic exposure. Exposures are teratogenic only when they exceed a threshold dose.22 This means that for every exposure there is a level below which damage to the embryo does not occur. As the dose increases above the threshold, the frequency and severity of teratogenic effects also increase. Almost any agent can have adverse effects on the developing embryo if given in a dose that is high enough to produce maternal toxicity. Chronic exposures usually have a greater potential to cause teratogenic effects than acute exposures at similar doses. For example, typical fetal alcohol syndrome will not occur in the child of a woman who only got drunk once during pregnancy, even if it was in the first trimester, but the risk increases substantially if the mother consumes large amounts of alcohol throughout gestation. Route of exposure also affects the dose received by the embryo. Dermal exposures that do not produce substantial systemic absorption are unlikely to cause adverse effects in the embryo.
Genetic susceptibility to teratogenic effects
Both maternal and fetal genotypes can affect placental transport, absorption, metabolism, distribution and receptor binding of an agent, influencing its teratogenicity. Such gene–teratogen interactions may explain why the range of effects seen after identical exposures can be so broad. The extent to which teratogenic effects depend on the fetal or maternal genotype is much more easily determined in experimental animals than in humans; however, there is evidence of differences in genetic susceptibility within human populations that result in greater damage from a teratogenic exposure in one individual than in another.23,24 Some examples of the interplay between genetic variants in humans and the development of congenital anomalies follow.
Cigarette smoking
The association of cigarette smoking by pregnant women with low birth weight and preterm delivery of their infants is well recognized.25,26,27,28,29 Although maternal smoking during pregnancy is associated with low birth weight regardless of maternal genotype, greater reductions in birth weight have been found among infants whose mothers smoked and had certain variants of the CYP1A1 gene.29 These variants influence the activity of enzymes involved in the metabolism of some of the chemicals found in cigarette smoke.29,30
Children of women who smoke during pregnancy have also been found to have orofacial clefts, limb reduction defects, congenital foot deformities, urinary tract abnormalities, craniosynostosis and other congenital anomalies 1.5–2 times more frequently than expected in some studies.31,32,33,34,35,36,37 Other studies suggest that maternal smoking poses a higher risk of producing oral clefts in infants of women who carry genetic variants involved in the metabolism of cigarette smoke.38,39,40 Further research is needed to define the interaction of genetic predisposition, maternal smoking and birth defects.
Alcohol
Excessive maternal drinking during pregnancy has been associated with a wide spectrum of birth defects in children. The most severe manifestation is fetal alcohol syndrome, which typically occurs among children whose mothers abuse alcohol in large amounts daily. Features of fetal alcohol syndrome include growth retardation, microcephaly, mental retardation, neurobehavioural deficits and facial dysmorphism.41 Lower levels of maternal alcohol consumption during pregnancy are associated with less severe neurodevelopmental anomalies in children.42 The finding that concordance for alcohol-related birth defects is higher among monozygotic twins than among dizygotic twins born to women who drink during pregnancy supports the importance of genetic factors in determining susceptibility to fetal alcohol syndrome.43
Ethanol is metabolized by alcohol dehydrogenase (ADH) and cytochrome P4502E1 (CYP2E1) to acetaldehyde, which is then oxidized to acetate by aldehyde dehydrogenase (ALDH). Genetic variants of each of these enzymes occur.44 A protective role for one particular variant at the ADH2 locus has been demonstrated for both their offspring's neurobehavioural development and intrauterine growth in a prospective study of African American women who drank during pregnancy.45
Antiepileptic drugs
Maternal treatment with most antiepileptic drugs (AEDs) has been associated with an increased risk of congenital anomalies in the offspring. These anomalies include central nervous system, cardiovascular, genitourinary and gastrointestinal defects (see Table 3).46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62 An anticonvulsant embryopathy that includes growth retardation, developmental delay, midface hypoplasia and distal digital hypoplasia has been observed among the children of women treated with several different AEDs.63,64,65 The features of this syndrome may vary somewhat with the different AEDs, but it does not occur among the children of women with epilepsy who are not treated with an AED during pregnancy.56,65 In addition, neural tube defects occur in 1%–2% of infants whose mothers were treated with valproic acid and 0.5%–1.7% of infants whose mothers were treated with carbamazepine during pregnancy.21,51,66,67,68
The overall risk of congenital anomalies among the infants of epileptic mothers treated with AEDs during pregnancy is 2–3 times higher than the “baseline” risk that attends every pregnancy, which has been estimated to be between 3% and 5%.1,63,64,65 The extent to which this increased risk of congenital anomalies can be attributed to genetic factors, severity of maternal epilepsy, AEDs or a combination of these factors has been difficult to ascertain. The risk of congenital anomalies is even greater in those infants whose mothers must take more than one AED concomitantly during pregnancy.51,69,70 Although the real magnitude of this increased risk is unknown, investigators in one study found that women with epilepsy who took only one AED during pregnancy had a 3% risk of giving birth to an infant with congenital anomalies compared with a 2% risk in untreated women with epilepsy.69 This risk increased to 5% in women who took 2 AEDs during pregnancy, 10% when 3 AEDs were taken and 20% when as many as 4 AEDs were taken. The combination of valproic acid, carbamazepine and phenobarbital, in particular, may be more teratogenic than other combinations of AEDs.51
The teratogenic risk associated with the use of AEDs during pregnancy for indications other than epilepsy may be different. In addition, the information available about the new AEDs, such as felbamate, lamotrigine, gabapentin, oxcarbazepine, tiagabine, topiramate or vigabatrin, is too limited at the present time to determine whether or not the therapeutic use of these medications during pregnancy poses a greater or smaller teratogenic risk.
Maternal seizures during the first trimester of pregnancy are also associated with an increased risk of congenital anomalies in the children. In one study, congenital anomalies were found in 12% of infants of AED-treated mothers who experienced seizures in the first trimester of pregnancy compared with only 4% of infants whose AED-treated mothers did not have seizures.71 Seizures during pregnancy have also been associated with higher fetal (30%–50%) and maternal mortality rates.72
Women who are treated with AEDs and give birth to an infant with congenital anomalies are at increased risk of having another affected child in subsequent pregnancies.73 A number of AEDs are metabolized through an epoxide hydrolase pathway,74 and significantly lower levels of epoxide hydrolase activity have been found in children with fetal hydantoin syndrome than in unaffected children born to women with epilepsy who were treated with phenytoin during pregnancy.75 It has also been suggested that genetic differences in folate metabolism may account for the increased risk of congenital anomalies, particularly neural tube defects, in the children of women with epilepsy treated with AEDs. In one study, epileptic mothers of children who were diagnosed with fetal anticonvulsant syndrome were more likely than epileptic mothers of unaffected children to be homozygous for the C677T variant of the MTHFR gene.76
Women with epilepsy who are planning a pregnancy should be advised to take 4–5 mg of folic acid per day at least 3 months before conception and throughout pregnancy.72 They should be told about their increased risk of giving birth to an infant with congenital anomalies if they continue AED therapy while they are pregnant. They should also be informed of the potential maternal and fetal risks associated with discontinuation of AEDs and the recurrence of seizures during pregnancy. If AED withdrawal is attempted, it should be completed at least 6 months before conception, because the risk of seizure recurrence is greatest during the first 6 months of AED withdrawal.72 If treatment cannot be discontinued, then the patient's physician should prescribe the AED that best controls the patient's seizures at the lowest effective dose. The use of a single agent is preferable to using multiple AEDs.
Women with epilepsy who become pregnant while being treated should be advised of their increased risk of giving birth to an infant with congenital anomalies. AED therapy should not be discontinued, because the frequency of seizures may increase. Prenatal diagnostic tests, such as serum alpha-fetoprotein screening at 15–22 weeks and ultrasonography at 18–22 weeks of pregnancy, should be offered, and amniocentesis for amniotic alpha-fetoprotein and acetylcholinesterase measurement at 16–20 weeks should be considered if appropriate.72 Serum concentrations of the AED should be monitored regularly to help avoid toxic and subtherapeutic treatment.77
Many of the commonly used AEDs can decrease fetal vitamin K–dependent coagulation factors (II, VII, IX and X) and cause early hemorrhagic disease of the newborn.72,78 It is therefore recommended that women who are treated with AEDs receive oral supplementation with vitamin K, 10 mg/d, at least one month before their delivery.72,77
Teratogen risk counselling
Counselling for possible teratogenic risks should be provided by physicians or other health care professionals who have training and experience in clinical teratology. Difficult or complex cases should be referred to the appropriate specialists (see the Additional resources box at the end of this article). When estimating a patient's teratogenic risk, the counsellor must determine whether a pregnant woman's exposure significantly increases her risk of giving birth to an infant with major malformations or mental retardation, or both, above the risk expected without such exposure. This “baseline” risk that attends every pregnancy is usually quoted as 3%–5%,1 but the baseline risk may be higher because of a woman's age, medical or family history, or other exposures.
In order to determine a woman's teratogenic risk, the counsellor must take a thorough medical and family history. The nature, dose, route, duration and gestational timing of the exposure of concern should be documented, as well as the reason for the exposure. It is important to include information regarding the woman's exposure to other agents, such as herbal products, tobacco, alcohol and other “recreational” drugs. Many women have more than one exposure during pregnancy.
Several attempts have been made to provide lists of agents that are “safe” or “unsafe” in pregnancy.79,80 Although it is tempting to rely on these lists, they have serious limitations and can result in the mismanagement of pregnancies.81 This is because agents cannot be classified as teratogens or nonteratogens without consideration of the dose, route, duration and gestational timing of the treatment. In addition, the likelihood that a particular treatment will cause fetal damage may also depend on concomitant exposure to other agents, the maternal illness that is being treated, and both maternal and fetal susceptibility. Moreover, the level of risk (or, alternatively, the amount of assurance of safety) that is acceptable varies from patient to patient. If a physician is contemplating initiating treatment in a pregnant woman, the benefit of the therapy to her and her fetus and the availability of safe and effective alternatives are also important issues. In addition, a list of “safe” drugs implies that the agents on the list have been tested in humans for the full range of potential developmental toxicities, including fetal death, structural malformations and functional deficits. Very few drug treatments have been evaluated to this extent.
Once the magnitude of teratogenic risk has been estimated, the counsellor must communicate this risk to the patient in a manner that enables her to make informed decisions regarding future exposures and further management of the pregnancy. Pregnant patients at high risk should be advised about prenatal diagnosis for possible teratogen- induced abnormalities, but many such anomalies are not detectable by detailed ultrasonography and few can be identified by amniocentesis.
A woman who requires treatment for a chronic disease such as epilepsy, diabetes or a psychiatric illness will often benefit from planning her pregnancies in consultation with her physician, who can discuss the risks and advantages of changing the treatment to one that is safer for the embryo and fetus. The feasibility of stopping treatment, completing certain aspects of the treatment before becoming pregnant or deferring part of the treatment until after pregnancy is completed can also be considered.
Modern genetics and clinical teratology
Genetic variants have been identified in more than 20 human drug-metabolizing enzymes that may produce significant alteration in clinical responses to drug treatment. These genetic variants often exhibit different frequencies in different populations. New genomic technology has enabled scientists to scan the entire human genome for genetic variants that may affect drug metabolism or activity, and many more are likely to be found in the next few years.82,83,84
Genetic variations that result in a deficiency of enzymes essential for the elimination of drugs or chemicals would be expected to increase the teratogenicity of these agents in at-risk individuals. On the other hand, people who lack the genetic susceptibility to the teratogenic effects of a particular treatment may be able to use that treatment safely even if it is likely to be teratogenic in most cases. Pharmacogenetics should prove useful in helping clinical teratologists understand how genetic variants influence teratogenic processes in humans, estimate teratogenic risk in individual patients and target intervention strategies to reduce the risk of birth defects.85 More research is needed to identify genetic variants that predict susceptibility to teratogenic effects, so that harmful exposures can be prevented and more effective therapy provided to pregnant women and their babies.82,83,86
The case revisited
The couple described at the beginning of this article were counselled regarding their increased risk of having an infant with a major malformation or mental retardation because of the woman's epilepsy and AED treatment. The importance of seizure control and the risks of adverse maternal and fetal effects associated with seizures during pregnancy were also discussed. After consulting with her neurologist, the patient decided that she would continue treatment with divalproex when she became pregnant. She immediately began to take 5 mg of supplemental folic acid per day. To avoid high peaks of valproic acid plasma concentrations, her divalproex was given in divided doses throughout the day instead of once a day. Three months later, the patient became pregnant. Plasma concentrations of valproic acid were monitored every month. The patient had serum alpha-fetoprotein screening at 15 weeks of pregnancy, targeted ultrasonography and fetal echocardiography at 18 weeks' gestation and amniocentesis for alpha- fetoprotein and acetylcholinesterase measurement at 20 weeks. All tests were normal. The patient began taking a 10-mg oral vitamin K supplement every day a month before her expected date of delivery. She delivered a healthy infant at term after completing an otherwise uneventful pregnancy.
Additional resources
-
Canadian teratology information services
-
· IMAGE: Info-Medicaments en Allaitement et Grossesse
-
(514) 345-2333
-
Geographic region: province of Quebec
-
Accepts calls from: health care professionals only (members of the public should call Motherisk)
-
· Motherisk program
-
(416) 813-6780
-
Geographic region: province of Ontario
-
Accepts calls from: the public and health care professionals
-
Clinical teratology resources
-
Briggs GG, Freeman RK, Yaffe SJ. Drugs in pregnancy and lactation: a reference guide to fetal and neonatal risk. 6th ed. Baltimore: Lippincott Williams & Wilkins Publishers; 2002.
-
Drugs and pregnancy. 2nd ed. Gilstrap LC III, Little BB, editors. New York: Chapman & Hall; 1998
-
Friedman JM, Polifka JE. Teratogenic effects of drugs: a resource for clinicians. 2nd ed. Baltimore: Johns Hopkins University Press; 2000.
-
Maternal-fetal toxicology: a clinician's guide. 2nd ed. Koren G, editor. New York: Marcel Dekker, Inc.; 1994.
-
Schardein JL. Chemically induced birth defects. 3rd ed. New York: Marcel Dekker; 2000.
-
Scialli AR, Lione A, Boyle Padgett GK. Reproductive effects of chemical, physical, and biologic agents–REPROTOX. Baltimore: Johns Hopkins University Press; 1995.
-
Shepard TH. Catalog of teratogenic agents. 10th ed. Baltimore: Johns Hopkins University Press; 2001.
-
Web site resources
-
· Agency for Toxic Substances and Disease Registry (ATSDR): http://www.atsdr.cdc.gov
-
· Clinical Teratology Web site: http://depts.washington.edu/~terisweb/
-
· NIEHS Center for the Evaluation of Risks to Human Reproduction (CERHR): http://cerhr.niehs.nih.gov
-
· Organization of Teratology Information Services (OTIS): http://www.otispregnancy.org/
-
· Teratology Society: http://teratology.org/
-
· US EPA Chemicals in the Environment: OPPT Chemical Fact Sheets: http://www.epa.gov/opptintr/chemfact/
Footnotes
-
This article has been peer reviewed.
Contributors: Dr. Janine E. Polifka contributed to the substance of the manuscript, drafted the article and gave approval of the version to be published. Dr. J.M. Friedman contributed to the substance of the manuscript, revised it critically for important intellectual content and gave final approval of the version to be published.
Competing interests: None declared.
Series editor: Dr. Judith Hall
References
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
- 10.
- 11.
- 12.
- 13.
- 14.
- 15.
- 16.
- 17.
- 18.
- 19.
- 20.
- 21.
- 22.
- 23.
- 24.
- 25.
- 26.
- 27.
- 28.
- 29.
- 30.
- 31.
- 32.
- 33.
- 34.
- 35.
- 36.
- 37.
- 38.
- 39.
- 40.
- 41.
- 42.
- 43.
- 44.
- 45.
- 46.
- 47.
- 48.
- 49.
- 50.
- 51.
- 52.
- 53.
- 54.
- 55.
- 56.
- 57.
- 58.
- 59.
- 60.
- 61.
- 62.
- 63.
- 64.
- 65.
- 66.
- 67.
- 68.
- 69.
- 70.
- 71.
- 72.
- 73.
- 74.
- 75.
- 76.
- 77.
- 78.
- 79.
- 80.
- 81.
- 82.
- 83.
- 84.
- 85.
- 86.