Anticonvulsant
PREGNANCY RECOMMENDATION: Compatible—Maternal Benefit >> Embryo–Fetal Risk
BREASTFEEDING RECOMMENDATION: Compatible
PREGNANCY SUMMARY
The use of phenytoin during pregnancy involves significant risk to the fetus in terms of major and minor congenital abnormalities and hemorrhage at birth. Adverse effects on neurodevelopment have also been reported. The risk to the mother, however, is also great if the drug is not used to control her seizures. The risk:benefit ratio, in this case, favors continued use of the drug during pregnancy. Frequent determinations of phenytoin levels are recommended to maintain the lowest level required to prevent seizures and possibly to lessen the likelihood of fetal anomalies. Based on recent research, consideration should also be given to monitoring folic acid levels simultaneously with phenytoin determinations and administering folic acid very early in pregnancy or before conception to those women shown to have low folate concentrations. Alternatively, adding folic acid, such as 4–5 mg/day (the dose recommended for women with a history of a fetus or infant with a neural tube defect [NTD]), should be considered.
FETAL RISK SUMMARY
Phenytoin is a hydantoin anticonvulsant introduced in 1938. The teratogenic effects of phenytoin were recognized in 1964 (1). Since this report there have been numerous reviews and studies on the teratogenic effects of phenytoin and other anticonvulsants. Based on this literature, the epileptic pregnant woman taking phenytoin, either alone or in combination with other anticonvulsants, has a 2–3 times greater risk for delivering a child with congenital defects over the general population (2–9).
It was not always known whether this increased risk was caused by antiepileptic drugs, the disease itself, genetic factors, or a combination of these, although past evidence indicated that drugs were the causative factor. Fifteen epidemiologic studies cited by reviewers in 1982 found an incidence of defects in treated epileptics varying from 2.2% to 26.1% (9). In each case, the rate for treated patients was higher than for untreated epileptics or normal controls. Animal studies have also implicated drugs and have suggested that a dose-related response may occur (9). Two studies cited below provide further evidence that the congenital defects observed in the offspring of epileptic mothers treated with anticonvulsants are caused by drugs.
A prospective study published in 1999 described the outcomes of 517 pregnancies of epileptic mothers identified at one Italian center from 1977 (10). Excluding genetic and chromosomal defects, malformations were classified as severe structural defects, mild structural defects, and deformations. Minor anomalies were not considered. Spontaneous (N = 38) and early (N = 20) voluntary abortions were excluded from the analysis, as were seven pregnancies that delivered at other hospitals. Of the remaining 452 outcomes, 427 were exposed to anticonvulsants of which 313 involved monotherapy: phenytoin (N = 31), carbamazepine (N = 113), phenobarbital (N = 83), valproate (N = 44), primidone (N = 35), clonazepam (N = 6), and other (N = 1). There were no defects in the 25 pregnancies not exposed to anticonvulsants. Of the 42 (9.3%) outcomes with malformations, 24 (5.3%) were severe, 10 (2.2%) were mild, and 8 (1.8%) were deformities. There were three malformations with phenytoin monotherapy: none were severe, one (3.2%) was mild (umbilical hernia), and two (6.4%) deformations (clubfoot and hip dislocation). The investigators concluded that the anticonvulsants were the primary risk factor for an increased incidence of congenital malformations (see also Carbamazepine, Clonazepam, Phenobarbital, Primidone, and Valproic Acid) (10).
A prospective cohort study, conducted from 1986 to 1993 at five maternity hospitals, was designed to determine if anticonvulsant agents or other factors (e.g., genetic) were responsible for the constellation of abnormalities seen in infants of mothers treated with anticonvulsants during pregnancy (11). A total of 128,049 pregnant women were screened at delivery for exposure to anticonvulsant drugs. Three groups of singleton infants were identified: (a) exposed to anticonvulsant drugs, (b) not exposed to anticonvulsant drugs but with a maternal history of seizures, and (c) not exposed to anticonvulsant drugs and with no maternal history of seizures (control group). After applying exclusion criteria, including exposure to other teratogens, 316, 98, and 508 infants, respectively, were analyzed. Anticonvulsant monotherapy occurred in 223 women: phenytoin (N = 87), phenobarbital (N = 64), carbamazepine (N = 58), and too few cases for analysis with valproic acid, clonazepam, diazepam, and lorazepam. Ninety-three infants were exposed to two or more anticonvulsant drugs. All infants were examined systematically (blinded as to group in 93% of the cases) for embryopathy associated with anticonvulsant exposure (major malformations, hypoplasia of the midface and fingers, microcephaly, and intrauterine growth restriction). Compared with controls, significant associations between anticonvulsants and anticonvulsant embryopathy were as follows: phenytoin monotherapy 20.7% (18/87), phenobarbital monotherapy 26.6% (17/64), any monotherapy 20.6% (46/223), exposed to ≥2 anticonvulsants 28.0% (26/93), and all infants exposed to anticonvulsants (monotherapy and polytherapy) 22.8% (72/316). Nonsignificant associations were found for carbamazepine monotherapy 13.8% (8/58), nonexposed infants with a maternal history of seizures 6.1% (6/98), and controls 8.5% (43/508). The investigators concluded that the distinctive pattern of physical abnormalities observed in infants exposed to anticonvulsants during gestation was due to the drugs, rather than to epilepsy itself (11).
A study published in 1990 provided evidence that, at least in some cases, the teratogenic effects of phenytoin are secondary to elevated levels of oxidative metabolites (epoxides) (12). Epoxides are normally eliminated by the enzyme epoxide hydrolase, but in some individuals, low activity of this enzyme is present. By measuring the enzyme’s activity in a number of subjects, the investigators proposed a trimodal distribution that is regulated by a single gene with two allelic forms. The three phenotypes proposed were: low activity (homozygous for the recessive allele), intermediate activity (heterozygous), and high activity (homozygous for the dominant allele). In the prospective portion of the study, 19 pregnant women with epilepsy, who were being treated with phenytoin monotherapy, had an amniocentesis performed and the microsomal epoxide hydrolase activity in amniocytes was determined. Four of the 19 had low activity (<30% of standard), whereas 15 had normal activity (>30% of standard). As predicted, only the four fetuses with low activity had clinical evidence of the fetal hydantoin syndrome (FHS) (12).
In contrast to the above study, a 1999 study with mice involving fluconazole and phenytoin was unable to provide support for the theory that toxic intermediates, such as epoxides, were the cause of phenytoin-induced congenital defects (13). Because fluconazole inhibits the cytochrome P450 pathway responsible for phenytoin metabolism, the authors of this study reasoned that the drug combination could provide a test of the hypothesis. Pretreatment of mice with a nonembryotoxic fluconazole dose, however, significantly doubled (from 6.2% to 13.3%) the incidence of phenytoin-induced cleft palate. Administering both drugs closely together significantly increased the incidence of resorptions but not malformations. This lack of effect on malformations may have been related to the increased embryolethality of the combination. The mechanism for the teratologic interaction between the drugs was unknown (13).
In a surveillance study of Michigan Medicaid recipients involving 229,101 completed pregnancies conducted between 1985 and 1992, 332 newborns had been exposed to phenytoin during the 1st trimester (F. Rosa, personal communication, FDA, 1993). A total of 15 (4.5%) major birth defects were observed (13 expected), including (observed/expected) 5/3 cardiovascular defects, 1/0 spina bifida, and 1/1 hypospadias. No anomalies were observed in three other defect categories (oral clefts, polydactyly, and limb reduction defects) for which specific data were available.
A recognizable pattern of malformations, now known as the FHS, was partially described in 1968 when Meadow (14) observed distinct facial abnormalities in infants exposed to phenytoin and other anticonvulsants. In 1973, two groups of investigators, in independent reports, described unusual anomalies of the fingers and toes in exposed infants (15,16). The basic syndrome consists of variable degrees of hypoplasia and ossification of the distal phalanges and craniofacial abnormalities. Clinical features of the FHS, not all of which are apparent in every infant, are as follows (14–16):
Craniofacial
Broad nasal bridge
Wide fontanelle
Low-set hairline
Broad alveolar ridge
Metopic ridging
Short neck
Ocular hypertelorism
Microcephaly
Cleft lip and/or palate
Abnormal or low-set ears
Epicanthal folds
Ptosis of eyelids
Coloboma
Coarse scalp hair
Limbs
Small or absent nails
Hypoplasia of distal phalanges
Altered palmar crease
Digital thumb
Dislocated hip
Impaired growth (physical and mental) and congenital heart defects are often observed in conjunction with FHS.
Numerous other defects have been reported to occur after phenytoin exposure in pregnancy. Janz, in a 1982 review (17), stated that nearly all possible types of malformations may be observed in the offspring of epileptic mothers. This statement is supported by the large volume of literature describing various anomalies that have been attributed to phenytoin with or without other anticonvulsants (1–9,14–58).
In one of the above cases the mother took phenytoin (300 mg/day) during the first 7 months of gestation and phenobarbital (90 mg/day) during the last 2 months (54). The full-term female infant had features of FHS (hypoplastic nails and flat nasal bridge), hydrocephalus, a left porencephalic cyst, and an encephalocele. She died at 2.5 months of age of bronchopneumonia. An autopsy of the brain revealed both gross and microscopic abnormalities. The defects were consistent with changes produced early in fetal development and with tissue destruction resulting in the second half of gestation (54).
A possible association between phenytoin and the rare defect, holoprosencephaly, was reported in 1993 (56). Because of psychomotor and petit mal seizures, the mother was treated with phenytoin (350 mg/day) and primidone (500 mg/day) during gestation. Except for microcephaly, fetal sonography detected no pathology. The female infant was born after 36 weeks of gestation with both weight and length above the 50th percentile. The Apgar scores were 3, 9, and 10 at 1, 5, and 10 minutes, respectively. The occipitofrontal head circumference (31 cm) was below the 25th percentile. Malformations evident on examination were microcephaly, narrow forehead, hypertelorism, hypoplastic midface with anteverted nostrils, smooth philtrum, and thin vermillion border lip. Distal phalanges and nails on all fingers and toes were noted to be hypoplastic. A sacral dimple was also noted. Ultrasound examination revealed a right-sided renal duplication, hepatomegaly, a mild ventricular septal defect, and bilateral hip dysplasia. An electroencephalogram revealed significantly delayed visual-evoked potentials. A partial lobar holoprosencephaly with ventral fusion of the cerebral hemispheres was noted on magnetic resonance imaging of the brain. Coarse gyri and horizontal cleavage, but no sagittal cleavage, was noted in the frontal lobe. The corpus callosum was absent. Although both parents suffered from epilepsy, there was no consanguinity and the infant’s karyotype was normal (46,XX), thus a genetic cause of the holoprosencephaly was thought to be unlikely (56).
Authors of correspondence relating to the above report noted that they had also identified a case of holoprosencephaly in a stillborn infant exposed in utero to anticonvulsants (59). Unfortunately, the clinical examination did not record the presence or absence of minor physical features characteristic of the FHS. Based on their experience, however, establishing an association between phenytoin and the defect would be very difficult because of the infrequent in utero exposure to anticonvulsants (1:250 births), the even lower frequency of exposure to phenytoin monotherapy (1:844 births), and the rarity of holoprosencephaly (1:10,000) (59).
Thanatophoric dwarfism was found in a stillborn infant exposed throughout gestation to phenytoin (200 mg/day), phenobarbital (300 mg/day), and amitriptyline (>150 mg/day) (60). The cause of the malformation could not be determined, but drug and genetic causes were considered.
A 2000 study, using data from the MADRE (an acronym for malformation and drug exposure) surveillance project, assessed the human teratogenicity of anticonvulsants (61). Among 8005 malformed infants, cases were defined as infants with a specific malformation, whereas controls were infants with other anomalies. Of the total group, 299 were exposed in the 1st trimester to anticonvulsants. Among these, exposure to monotherapy occurred in the following: phenytoin (N = 24), phenobarbital (N = 65), mephobarbital (N = 10), carbamazepine (N = 46), valproic acid (N = 80), and other agents (N = 16). No statistically significant associations were found with phenytoin monotherapy or polytherapy. Although the study confirmed some previously known associations, several new associations with anticonvulsants were discovered that require independent confirmation (see also Carbamazepine, Mephobarbital, Phenobarbital, and Valproic Acid) (61).
The Lamotrigine Pregnancy Registry, an ongoing project conducted by the manufacturer, was first published in January 1997. The final report was published in July 2010 (62). The Registry is now closed. Among 68 prospectively enrolled pregnancies exposed to phenytoin and lamotrigine, with or without other anticonvulsants, 62 were exposed in the 1st trimester resulting in 56 live births, 1 birth defect, 1 spontaneous abortion, and 4 elective abortions. There were six exposures in the 2nd/3rd trimesters resulting in six live births without birth defects (62).
Twelve case reports have been located that, taken in sum, suggest phenytoin is a human transplacental carcinogen (18–28,63). Tumors reported to occur in infants after in utero exposure to phenytoin include the following:
Neuroblastoma (6 cases) (18–22,63)
Ganglioneuroblastoma (1 case) (23)
Melanotic neuroectodermal tumor (1 case) (24)
Extrarenal Wilms’ tumor (1 case) (25)
Mesenchymoma (1 case) (26)
Lymphangioma (1 case) (27)
Ependymoblastoma (1 case) (28)
Children exposed in utero to phenytoin should be closely observed for several years because tumor development may take that long to express itself. A 1989 study, however, found no in utero exposures to phenytoin among 188 cases of childhood neuroblastoma diagnosed between 1969 and 1988 at their center (64). But the investigators did conclude that an increased risk of the neoplasm in children with FHS was still a possibility.
Phenytoin and other anticonvulsants (e.g., phenobarbital) may cause early hemorrhagic disease of the newborn (18,65–79). Hemorrhage occurs during the first 24 hours after birth and may be severe or even fatal. The exact mechanism of the defect is unknown but may involve phenytoin induction of fetal liver microsomal enzymes that deplete the already low reserves of fetal vitamin K (79). This results in suppression of the vitamin K-dependent coagulation factors II, VII, IX, and X. Phenytoin-induced thrombocytopenia has also been reported as a mechanism for hemorrhage in the newborn (76). A 1985 review summarized the various prophylactic treatment regimens that have been proposed (79):
Administering 10 mg of oral vitamin K daily during the last 2 months of pregnancy
Administering 20 mg of oral vitamin K daily during the last 2 weeks of pregnancy
Avoiding salicylates and administering vitamin K during labor
Cesarean section if a difficult or traumatic delivery is anticipated
Administering IV vitamin K to the newborn in the delivery room plus cord blood clotting studies
Although all of the above suggestions are logical, none has been tested in controlled trials. The reviewers recommended immediate IM vitamin K and close observation of the infant (see also Phytonadione) (79).
Of interest, a 1995 study suggested that anticonvulsant-induced vitamin K deficiency may be the mechanism that causes the maxillonasal hypoplasia seen in the FHS (80). They proposed early vitamin K supplementation of at-risk pregnancies to prevent this disfiguring malformation.
Liver damage was observed in an infant exposed during gestation to phenytoin and valproic acid (81). Although they were unable to demonstrate which anticonvulsant caused the injury, the authors concluded that valproic acid was the more likely offending agent.
Phenytoin may induce folic acid deficiency in the epileptic patient by impairing gastrointestinal absorption or by increasing hepatic metabolism of the vitamin (82–84). Whether phenytoin also induces folic acid deficiency in the fetus is less certain because the fetus seems to be efficient in drawing on available maternal stores of folic acid (see Folic Acid). Low maternal folate levels, however, have been proposed as one possible mechanism for the increased incidence of defects observed in infants exposed in utero to phenytoin. In a 1984 report, two investigators studied the relationship between folic acid, anticonvulsants, and fetal defects (82). In the retrospective part of this study, a group of 24 women treated with phenytoin and other anticonvulsants produced 66 infants, 10 (15%) with major anomalies. Two of the mothers with affected infants had markedly low red blood cell folate concentrations. A second group of 22 epileptic women was then supplemented with daily folic acid, 2.5–5.0 mg, starting before conception in 26 pregnancies and within the first 40 days in 6. This group produced 33 newborns (32 pregnancies—1 set of twins) with no defects, a significant difference from the nonsupplemented group. Loss of seizure control caused by folic acid lowering of phenytoin serum levels, which is known to occur, was not a problem in this small series (82).
Negative associations between phenytoin and folate deficiency have been reported (83,84). In one study, mothers were given supplements with an average folic acid dose of 0.5 mg/day from the 6th to 16th week of gestation until delivery (84). Defects were observed in 20 infants (15%) from the 133 women taking anticonvulsants, which is similar to the reported frequency in pregnant patients not given supplements. Folate levels were usually within the normal range for pregnancy.
The effects of exposure (at any time during the 2nd or 3rd month after the last menstrual period) to folic acid antagonists on embryo–fetal development were evaluated in a large, multicenter, case–control surveillance study published in 2000 (85). The report was based on data collected between 1976 and 1998 from 80 maternity or tertiary care hospitals. Mothers were interviewed within 6 months of delivery about their use of drugs during pregnancy. Folic acid antagonists were categorized into two groups: group I—dihydrofolate reductase inhibitors (aminopterin, methotrexate, sulfasalazine, pyrimethamine, triamterene, and trimethoprim); group II—agents that affect other enzymes in folate metabolism, impair the absorption of folate, or increase the metabolic breakdown of folate (carbamazepine, phenytoin, primidone, and phenobarbital). The case subjects were 3870 infants with cardiovascular defects, 1962 with oral clefts, and 1100 with urinary tract malformations. Infants with defects associated with a syndrome were excluded as were infants with coexisting NTDs (known to be reduced by maternal folic acid supplementation). Too few infants with limb reduction defects were identified to be analyzed. Controls (N = 8387) were infants with malformations other than oral clefts and cardiovascular, urinary tract, and limb reduction defects and NTDs, but included infants with chromosomal and genetic defects. The risk of malformations in control infants would not have been reduced by vitamin supplementation, and none of the controls used folic acid antagonists. For group I cases, the relative risks (RRs) of cardiovascular defects and oral clefts were 3.4 (95% confidence interval [CI] 1.8–6.4) and 2.6 (95% CI 1.1–6.1), respectively. For group II cases, the RRs of cardiovascular and urinary tract defects, and oral clefts were 2.2 (95% CI 1.4–3.5), 2.5 (95% CI 1.2–5.0), and 2.5 (95% CI 1.5–4.2), respectively. Maternal use of multivitamin supplements with folic acid (typically 0.4 mg) reduced the risks in group I cases, but not in group II cases (85).
The pharmacokinetics and placental transport of phenytoin have been extensively studied and reviewed (86–88). Plasma concentrations of phenytoin may fall during pregnancy. Animal studies and recent human reports suggest a dose-related teratogenic effect of phenytoin (89,90). Although these results are based on a small series of patients, it is reasonable to avoid excessively high plasma concentrations of phenytoin. Close monitoring of plasma phenytoin concentrations is recommended to maintain adequate seizure control and prevent potential fetal hypoxia.
Placental function in women taking phenytoin has been evaluated (91). No effect was detected from phenytoin as measured by serum human placental lactogen, 24-hour urinary total estriol excretion, placental weight, and birth weight.
In a study evaluating thyroid function, no differences were found between treated epileptic pregnant women and normal pregnant controls (92). Thyroxine levels in the cord blood of anticonvulsant-exposed infants were significantly lower than in controls, but this was shown to be caused by altered protein binding and not altered thyroid function. Other parameters studied—thyrotropin, free thyroxine, and triiodothyronine—were similar in both groups.
The effect of phenytoin on maternal and fetal vitamin D metabolism was examined in a 1984 study (93). In comparison to normal controls, several significant differences were found in the level of various vitamin D compounds and in serum calcium, but the values were still within normal limits. No alterations were found in alkaline phosphatase and phosphate concentrations. The authors doubted whether the observed differences were of major clinical significance.
Phenytoin may be used for the management of digitalis-induced arrhythmias that are unresponsive to other agents and for refractory ventricular tachyarrhythmias (94–96). This short-term use has not been reported to cause problems in the exposed fetuses. The drug has also been used for anticonvulsant prophylaxis in severe preeclampsia (97).
A study published in 2009 examined the effect of antiepileptic drugs (AEDs) on the head circumference in newborns (98). Significant reductions in mean birth-weight-adjusted mean head circumference (bw-adj-HC) was noted for monotherapy with carbamazepine and valproic acid. No effect on bw-adj-HC was observed with gabapentin, phenytoin, clonazepam, and lamotrigine. A significant increase in the occurrence of microcephaly (bw-adj-HC smaller than 2 standard deviations below the mean) was noted after any AED polytherapy but not after any monotherapy, including carbamazepine and valproic acid. The potential effects of these findings on child development warrant study (98).
In a 1988 study designed to evaluate the effect of in utero exposure to anticonvulsants on intelligence, 148 Finnish children of epileptic mothers were compared with 105 controls (99). Previous studies had either shown intellectual impairment from this exposure or no effect. Of the 148 children of epileptic mothers, 129 were exposed to anticonvulsant therapy during the first 20 weeks of pregnancy, 2 were only exposed after 20 weeks, and 17 were not exposed. In those mothers treated during pregnancy, 103 received phenytoin (monotherapy in 54 cases), all during the first 20 weeks. The children were evaluated at 5.5 years of age for both verbal and nonverbal measures of intelligence. A child was considered mentally deficient if the results of both tests were <71. Two of the 148 children of epileptic mothers were diagnosed as mentally deficient and 2 others had borderline intelligence (the mother of 1 of these latter children had not been treated with anticonvulsant medication). None of the controls was considered mentally deficient. Both verbal (110.2 vs. 114.5) and nonverbal (108.7 vs. 113.2) intelligence scores were significantly lower in the study group children than in controls. In both groups, intelligence scores were significantly lower when seven or more minor anomalies were present. However, the presence of hypertelorism and digital hypoplasia, two minor anomalies considered typical of exposure to phenytoin, was not predictive of low intelligence (99).
A prospective, controlled, blinded observational 1994 study compared the global IQ and language development of children exposed in utero to either phenytoin (N = 36) or carbamazepine (N = 34) monotherapy to their respective matched controls (100). The cognitive tests were administered to the children between the ages of 18 and 36 months. The maternal IQ scores and socioeconomic status in the phenytoin subjects and their controls were similar (90 vs. 93.9 and 40.8 vs. 40.9, respectively) as they were in the carbamazepine subjects and controls (96.5 vs. 96.0 and 44.7 vs. 46.1, respectively). Compared with controls, phenytoin-exposed children had a significantly lower mean global IQ than their matched controls (113.4 vs. 103.1). The verbal comprehension and expressive language scores were also significantly lower (0.2 vs. 1.1 and –0.47 vs. 0.2, respectively). In contrast, no significant differences were measured either in IQ or language development scores between carbamazepine-exposed children and their matched controls. No correlation between the daily dose (mg/kg) of either anticonvulsant and global IQ was found. Major malformations were observed in two phenytoin-exposed children (cleft palate and hypospadias; meningomyelocele and hydrocephalus), none of the phenytoin controls, two carbamazepine-exposed children (missing last joint of right index finger and nail hypoplasia; hypospadias), and one carbamazepine control (pulmonary atresia). The study results suggested that phenytoin had a clinically important negative effect on neurobehavioral development that was independent of maternal or environmental factors (100). In subsequent correspondence relating to the above study (101,102), various perceived problems were cited and were addressed in a reply (103).
A 2013 study reported dose-dependent associations with reduced cognitive abilities across a range of domains at 6 years of age after fetal valproate exposure (104). Analysis showed that IQ was significantly lower for valproate (mean 97, 95% CI 94–101) than for carbamazepine (mean 105, 95% CI 102–108; p = 0.0015), lamotrigine (mean 108, 95% CI 105–110; p = 0.0003), or phenytoin (mean 108, 95% CI 104–112; p = 0.0006). Mean IQs were higher in children whose mothers had taken folic acid (mean 108, 95% CI 106–111) compared with children whose mothers had not taken the vitamin (mean 101, 95% CI 98–104; p = 0.0009) (104).
The relationship between maternal anticonvulsant therapy, neonatal behavior, and neurologic function in children was reported in a 1996 study (105). Among newborns exposed to maternal monotherapy, 18 were exposed to phenobarbital (including primidone), 13 to phenytoin, and 8 to valproic acid. Compared with controls, neonates exposed to phenobarbital had significantly higher mean apathy and optimality scores. Phenytoin-exposed neonates also had a significantly higher mean apathy score. However, the neonatal optimality and apathy scores did not correlate with neurologic outcome of the children at 6 years of age. In contrast, those exposed to valproic acid had optimality and apathy scores statistically similar to controls but a significantly higher hyperexcitability score. Moreover, the hyperexcitability score correlated with later minor and major neurologic dysfunction at age 6 years (105).
A two-part 2000 study evaluated the effects of prenatal phenobarbital and phenytoin exposure on brain development and cognitive functioning in adults (106). Subjects and controls, delivered at a mean 40 weeks’ gestation, were retrospectively identified from birth records covering the years between 1957 and 1972. Maternal diseases of the subjects included epilepsy (treated with anticonvulsants) and other conditions in which anticonvulsants were used as sedatives (nausea, vomiting, or emotional problems), whereas the matched control group had no maternal pathologies. Only those exposed prenatally to phenobarbital alone or phenobarbital plus phenytoin had sufficient subjects to analyze. The mean occipitofrontal circumference for phenobarbital-exposed neonates was not different from controls (34.49 vs. 34.50 cm), but it was significantly smaller for phenobarbital plus phenytoin subjects compared with phenobarbital alone or controls (33.82 cm). In the follow-up part of the study, no differences in adult cognitive functioning (intelligence, attention, and memory) were found between the exposed and control groups. More subjects than controls, however, were mentally slow (4 vs. 2; 1 control had autism) and more had persistent learning problems (12% vs. 1%). The investigators concluded that phenobarbital plus phenytoin reduced occipitofrontal circumference but may only affect cognitive capacity in susceptible offspring (106).
BREASTFEEDING SUMMARY
Phenytoin is excreted into breast milk. Milk:plasma ratios range from 0.18 to 0.54 (86,106–110). The pharmacokinetics of phenytoin during lactation have been reviewed (86). The reviewers concluded that little risk to the nursing infant was present if maternal levels were kept in the therapeutic range. However, methemoglobinemia, drowsiness, and decreased sucking activity were reported in one infant (111). Except for this one case, no other reports of adverse effects with the use of phenytoin during lactation have been located.
In a 2010 study, 199 children, who had been breastfed while their mothers were taking a single antiepileptic drug (carbamazepine, lamotrigine, phenytoin, or valproate), were evaluated at 3 years of age cognitive outcome (112). Mean adjusted IQ scores for exposed children were 99 (95% CI 96–103), whereas the mean adjusted IQ scores of nonbreastfed infants were 98 (95% CI 95–101).
The American Academy of Pediatrics classifies phenytoin as compatible with breastfeeding (113).
References
1.Janz D, Fuchs V. Are anti-epileptic drugs harmful when given during pregnancy? German Med Monogr 1964;9:20–3.
2.Hill RB. Teratogenesis and antiepileptic drugs. N Engl J Med 1973;289:1089–90.
3.Janz D. The teratogenic risk of antiepileptic drugs. Epilepsia 1975;16:159–69.
4.Bodendorfer TW. Fetal effects of anticonvulsant drugs and seizure disorders. Drug Intell Clin Pharm 1978;12:14–21.
5.Committee on Drugs, American Academy of Pediatrics. Anticonvulsants and pregnancy. Pediatrics 1977;63:331–3.
6.Nakane Y, Okuma T, Takahashi R, Sato Y, Wada T, Sato T, Fukushima Y, Kumashiro H, Ono T, Takahashi T, Aoki Y, Kazamatsuri H, Inami M, Komai S, Seino M, Miyakoshi M, Tanimura T, Hazama H, Kawahara R, Otuski S, Hosokawa K, Inanaga K, Nakazawa Y, Yamamoto K. Multi-institutional study of the teratogenicity and fetal toxicity of antiepileptic drugs: a report of a collaborative study group in Japan. Epilepsia 1980;21:663–80.
7.Andermann E, Dansky L, Andermann F, Loughnan PM, Gibbons J. Minor congenital malformations and dermatoglyphic alterations in the offspring of epileptic women: a clinical investigation of the teratogenic effects of anticonvulsant medication. In Epilepsy, Pregnancy and the Child. Proceedings of a Workshop in Berlin, September 1980. New York, NY: Raven Press, 1981.
8.Dansky L, Andermann E, Andermann F. Major congenital malformations in the offspring of epileptic patients. In Epilepsy, Pregnancy and the Child. Proceedings of a Workshop in Berlin, September 1980. New York, NY: Raven Press, 1981.
9.Hanson JW, Buehler BA. Fetal hydantoin syndrome: current status. J Pediatr 1982;101:816–8.
10.Canger R, Battino D, Canevini MP, Fumarola C, Guidolin L, Vignoli A, Mamoli D, Palmieri C, Molteni F, Granata T, Hassibi P, Zamperini P, Pardi G, Avanzini G. Malformations in offspring of women with epilepsy: a prospective study. Epilepsia 1999;40:1231–6.
11.Holmes LB, Harvey EA, Coull BA, Huntington KB, Khoshbin S, Hayes AM, Ryan LM. The teratogenicity of anticonvulsant drugs. N Engl J Med 2001;344:1132–8.
12.Buehler BA, Delimont D, Van Waes M, Finnell RH. Prenatal prediction of risk of the fetal hydantoin syndrome. N Engl J Med 1990;322:1567–72.
13.Tiboni GM, Iammarrone E, Giampietro F, Lamonaca D, Bellati U, Di Ilio C. Teratological interaction between the bis-triazole antifungal agent fluconazole and the anticonvulsant drug phenytoin. Teratology 1999;59:81–7.
14.Meadow SR. Anticonvulsant drugs and congenital abnormalities. Lancet 1968;2:1296.
15.Loughnan PM, Gold H, Vance JC. Phenytoin teratogenicity in man. Lancet 1973;1:70–2.
16.Hill RM, Horning MG, Horning EC. Antiepileptic drugs and fetal well-being. In Boreus L, ed. Fetal Pharmacology. New York, NY: Raven Press, 1973:375–9.
17.Janz D. Antiepileptic drugs and pregnancy: altered utilization patterns and teratogenesis. Epilepsia 1982;23(Suppl 1):S53–63.
18.Allen RW Jr, Ogden B, Bentley FL, Jung AL. Fetal hydantoin syndrome, neuroblastoma, and hemorrhagic disease in a neonate. JAMA 1980;244:1464–5.
19.Ramilo J, Harris VJ. Neuroblastoma in a child with the hydantoin and fetal alcohol syndrome. The radiographic features. Br J Radiol 1979;52:993–5.
20.Pendergrass TW, Hanson JW. Fetal hydantoin syndrome and neuroblastoma. Lancet 1976;2:150.
21.Sherman S, Roizen N. Fetal hydantoin syndrome and neuroblastoma. Lancet 1976;2:517.
22.Ehrenbard LT, Chaganti RSK. Cancer in the fetal hydantoin syndrome. Lancet 1981;2:97.
23.Seeler RA, Israel JN, Royal JE, Kaye CI, Rao S, Abulaban M. Ganglioneuroblastoma and fetal hydantoin-alcohol syndromes. Pediatrics 1979;63:524–7.
24.Jimenez JF, Seibert RW, Char F, Brown RE, Seibert JJ. Melanotic neuroectodermal tumor of infancy and fetal hydantoin syndrome. Am J Pediatr Hematol Oncol 1981;3:9–15.
25.Taylor WF, Myers M, Taylor WR. Extrarenal Wilms’ tumour in an infant exposed to intrauterine phenytoin. Lancet 1980;2:481–2.
26.Blattner WA, Hanson DE, Young EC, Fraumeni JF. Malignant mesenchymoma and birth defects. JAMA 1977;238:334–5.
27.Kousseff BG. Subcutaneous vascular abnormalities in fetal hydantoin syndrome. Birth Defects 1982;18:51–4.
28.Lipson A. Bale P. Ependymoblastoma associated with prenatal exposure to diphenylhydantoin and methylphenobarbitone. Cancer 1985;55:1859–62.
29.Corcoran R, Rizk MW. VACTERL congenital malformation and phenytoin therapy? Lancet 1976;2:960.
30.Pinto W Jr, Gardner LI, Rosenbaum P. Abnormal genitalia as a presenting sign in two male infants with hydantoin embryopathy syndrome. Am J Dis Child 1977;131:452–5.
31.Hoyt CS, Billson FA. Maternal anticonvulsants and optic nerve hypoplasia. Br J Ophthalmol 1978;62:3–6.
32.Wilson RS, Smead W, Char F. Diphenylhydantoin teratogenicity: ocular manifestations and related deformities. J Pediatr Ophthalmol Strabismus 1970;15:137–40.
33.Dabee V, Hart AG, Hurley RM. Teratogenic effects of diphenylhydantoin. Can Med Assoc J 1975;112:75–7.
34.Anderson RC. Cardiac defects in children of mothers receiving anticonvulsant therapy during pregnancy. J Pediatr 1976;89:318–9.
35.Hill RM, Verniaud WM, Horning MG, McCulley LB, Morgan NF. Infants exposed in utero to antiepileptic drugs. A prospective study. Am J Dis Child 1974;127:645–53.
36.Stankler L, Campbell AGM. Neonatal acne vulgaris: a possible feature of the fetal hydantoin syndrome. Br J Dermatol 1980;103:453–5.
37.Ringrose CAD. The hazard of neurotropic drugs in the fertile years. Can Med Assoc J 1972;106:1058.
38.Pettifor JM, Benson R. Congenital malformations associated with the administration of oral anticoagulants during pregnancy. J Pediatr 1975;86:459–61.
39.Biale Y, Lewenthal H, Aderet NB. Congenital malformations due to anticonvulsant drugs and congenital abnormalities. Obstet Gynecol 1975; 45:439–42.
40.Aase JM. Anticonvulsant drugs and congenital abnormalities. Am J Dis Child 1974;127:758.
41.Lewin PK. Phenytoin associated congenital defects with Y-chromosome variant. Lancet 1973;l:559.
42.Yang TS, Chi CC, Tsai CJ, Chang MJ. Diphenylhydantoin teratogenicity in man. Obstet Gynecol 1978;52:682–4.
43.Mallow DW, Herrick MK, Gathman G. Fetal exposure to anticonvulsant drugs. Arch Pathol Lab Med 1980;104:215–8.
44.Hirschberger M, Kleinberg F. Maternal phenytoin ingestion and congenital abnormalities: report of a case. Am J Dis Child 1975;129:984.
45.Hanson JW, Myrianthopoulos NC, Sedgwick Harvey MA, Smith DW. Risks to the offspring of women treated with hydantoin anticonvulsants, with emphasis on the fetal hydantoin syndrome. J Pediatr 1976;89:662–8.
46.Shakir RA, Johnson RH, Lambie DG, Melville ID, Nanda RN. Comparison of sodium valproate and phenytoin as single drug treatment in epilepsy. Epilepsia 1981;22:27–33.
47.Michalodimitrakis M, Parchas S, Coutselinis A. Fetal hydantoin syndrome: congenital malformation of the urinary tract—a case report. Clin Toxicol 1981;18:1095–7.
48.Phelan MC, Pellock JM, Nance WE. Discordant expression of fetal hydantoin syndrome in heteropaternal dizygotic twins. N Engl J Med 1982;307:99–101.
49.Kousseff BG, Root ER. Expanding phenotype of fetal hydantoin syndrome. Pediatrics 1982;70:328–9.
50.Wilker R, Nathenson G. Combined fetal alcohol and hydantoin syndromes. Clin Pediatr 1982;21:331–4.
51.Kogutt MS. Fetal hydantoin syndrome. South Med J 1984;77:657–8.
52.Krauss CM, Holmes LB, VanLang QN, Keith DA. Four siblings with similar malformations after exposure to phenytoin and primidone. J Pediatr 1984;105:750–5.
53.Pearl KN, Dickens S, Latham P. Functional palatal incompetence in the fetal anticonvulsant syndrome. Arch Dis Child 1984;59:989–90.
54.Trice JE, Ambler M. Multiple cerebral defects in an infant exposed in utero to anticonvulsants. Arch Pathol Lab Med 1985;109:521–3.
55.D’Souza SW, Robertson IG, Donnai D, Mawer G. Fetal phenytoin exposure, hypoplastic nails, and jitteriness. Arch Dis Child 1990;65:320–4.
56.Kotzot D, Weigl J, Huk W, Rott HD. Hydantoin syndrome with holoprosencephaly: a possible rate teratogenic effect. Teratology 1993;48:15–19.
57.Shaw GM, Wasserman CR, O’Malley CD, Lammer EJ, Finnell RH. Orofacial clefts and maternal anticonvulsant use. Reprod Toxicol 1995;9:97–8.
58.Sabry MA, Farag TI. Hand anomalies in fetal-hydantoin syndrome: from nail/phalangeal hypoplasia to unilateral acheiria. Am J Med Genet 1996;62:410–2.
59.Holmes LB, Harvey EA. Holoprosencephaly and the teratogenicity of anticonvulsants. Teratology 1994;49:82.
60.Rafla NM, Meehan FP. Thanatophoric dwarfism: drugs and antenatal diagnosis: a case report. Eur J Obstet Gynecol Reprod Biol 1990;38:161–5.
61.Arpino C, Brescianini S, Robert E, Castilla EE, Cocchi G, Cornel MC, de Vigan C, Lancaster PAL, Merlob P, Sumiyoshi Y, Zampino G, Renzi C, Rosano A, Mastroiacovo P. Teratogenic effects of antiepileptic drugs: use of an international database on malformations and drug exposure (MADRE). Epilepsia 2000;41:1436–43.
62.The Lamotrigine Pregnancy Registry. Final Report. September 1, 1992 through March 31, 2010. GlaxcoSmithKline, July 2010.
63.Al-Shammri S, Guberman A, Hsu E. Neuroblastoma and fetal exposure to phenytoin in a child without dysmorphic features. Can J Neurol Sci 1992;19:243–5.
64.Koren G, Demitrakoudis D, Weksberg R, Reider M, Shear NH, Sonely M, Shandling B, Spielberg SP. Neuroblastoma after prenatal exposure to phenytoin: cause and effect? Teratology 1989;40:157–62.
65.Lawrence A. Antiepileptic drugs and the foetus. Br Med J 1963;2:1267.
66.Kohler HG. Haemorrhage in newborn of epileptic mothers. Lancet 1966;1:267.
67.Douglas H. Haemorrhage in the newborn. Lancet 1966;1:816–7.
68.Monnet P, Rosenberg D, Bovier-Lapierre M. Terapeutique anticomitale administree pendant la grosses et maladie hemorragique du nouveau-ne. As cited in Bleyer WA, Skinner AL. Fetal neonatal hemorrhage after maternal anticonvulsant therapy. JAMA 1976;235:626–7.
69.Davis PP. Coagulation defect due to anticonvulsant drug treatment in pregnancy. Lancet 1970;1:413.
70.Evans AR, Forrester RM, Discombe C. Neonatal haemorrhage following maternal anticonvulsant therapy. Lancet 1970;1:517–8.
71.Stevensom MM, Bilbert EF. Anticonvulsants and hemorrhagic diseases of the newborn infant. J Pediatr 1970;77:516.
72.Speidel BD, Meadow SR. Maternal epilepsy and abnormalities of the foetus and newborn. Lancet 1972;2:839–40.
73.Truog WE, Feusner JH, Baker DL. Association of hemorrhagic disease and the syndrome of persistent fetal circulation with the fetal hydantoin syndrome. J Pediatr 1980;96:112–4.
74.Solomon GE, Hilgartner MW, Kutt H. Coagulation defects caused by diphenylhydantoin. Neurology 1972;22:1165–71.
75.Griffiths AD. Neonatal haemorrhage associated with maternal anticonvulsant therapy. Lancet 1981;2:1296–7.
76.Page TE, Hoyme HE, Markarian M, Jones KL. Neonatal hemorrhage secondary to thrombocytopenia: an occasional effect of prenatal hydantoin exposure. Birth Defects 1982;18:47–50.
77.Srinivasan G, Seeler RA, Tiruvury A, Pildes RS. Maternal anticonvulsant therapy and hemorrhagic disease of the newborn. Obstet Gynecol 1982;59:250–2.
78.Payne NR, Hasegawa DK. Vitamin K deficiency in newborns: a case report in α-1-antitrypsin deficiency and a review of factors predisposing to hemorrhage. Pediatrics 1984;73:712–6.
79.Lane PA, Hathaway WE. Vitamin K in infancy. J Pediatr 1985;106:351–9.
80.Howe AM, Lipson AH, Sheffield LJ, Haan EA, Halliday JL, Jenson F, David DJ, Webster WS. Prenatal exposure to phenytoin, facial development, and a possible role for vitamin K. Am J Med Genet 1995;58:238–44.
81.Felding I, Rane A. Congenital liver damage after treatment of mother with valproic acid and phenytoin? Acta Paediatr Scand 1984;73:565–8.
82.Biale Y, Lewenthal H. Effect of folic acid supplementation on congenital malformations due to anticonvulsive drugs. Eur J Obstet Reprod Biol 1984;18:211–6.
83.Pritchard JA, Scott DE, Whalley PJ. Maternal folate deficiency and pregnancy wastage. IV. Effects of folic acid supplements, anticonvulsants, and oral contraceptives. Am J Obstet Gynecol 1971;109:341–6.
84.Hiilesmaa VK, Teramo K, Granstrom ML, Bardy AH. Serum folate concentrations during pregnancy in women with epilepsy: relation to antiepileptic drug concentrations, number of seizures, and fetal outcome. Br Med J 1983;287:577–9.
85.Hernandez-Diaz S, Werler MM, Walker AM, Mitchell AA. Folic acid antagonists during pregnancy and the risk of birth defects. N Engl J Med 2000;343:1608–14.
86.Nau H, Kuhnz W, Egger HJ, Rating D, Helge H. Anticonvulsants during pregnancy and lactation: transplacental, maternal and neonatal pharmacokinetics. Clin Pharmacokinet 1982;7:508–43.
87.Chen SS, Perucca E, Lee JN, Richens A. Serum protein binding and free concentrations of phenytoin and phenobarbitone in pregnancy. Br J Clin Pharmacol 1982;13:547–52.
88.Van der Klign E, Schobben F, Bree TB. Clinical pharmacokinetics of antiepileptic drugs. Drug Intell Clin Pharm 1980;14:674–85.
89.Dansky L, Andermann E, Sherwin AL, Andermann F. Plasma levels of phenytoin during pregnancy and the puerperium. In Epilepsy, Pregnancy and the Child. Proceedings of a Workshop held in Berlin, September 1980. New York, NY: Raven Press, 1981.
90.Dansky L, Andermann E, Andermann F, Sherwin AL, Kinch RA. Maternal epilepsy and congenital malformation: correlation with maternal plasma anticonvulsant levels during pregnancy. In Epilepsy, Pregnancy and the Child. Proceedings of a Workshop held in Berlin, September 1980. New York, NY: Raven Press, 1981.
91.Hiilesmaa VK. Evaluation of placental function in women on antiepileptic drugs. J Perinat Med 1983;11:187–92.
92.Carriero R, Andermann E, Chen MF, Eeg-Oloffson O, Kinch RAH, Klein G, Pearson Murphy BE. Thyroid function in epileptic mothers and their infants at birth. Am J Obstet Gynecol 1985;151:641–4.
93.Markestad T, Ulstein M, Strandjord RE, Aksnes L, Aarskog D. Anticonvulsant drug therapy in human pregnancy: effects on serum concentrations of vitamin D metabolites in maternal and cord blood. Am J Obstet Gynecol 1984;150:254–8.
94.Tamari I, Eldar M, Rabinowitz B, Neufeld HN. Medical treatment of cardiovascular disorders during pregnancy. Am Heart J 1982;104:1357–63.
95.Rotmensch HH, Elkayam U, Frishman W. Antiarrhythmic drug therapy during pregnancy. Ann Intern Med 1983;98:487–97.
96.Rotmensch HH, Rotmensch S, Elkayam U. Management of cardiac arrhythmias during pregnancy: current concepts. Drugs 1987;33:623–33.
97.Ryan G, Lange IR, Naugler MA. Clinical experience with phenytoin prophylaxis in severe preeclampsia. Am J Obstet Gynecol 1989;161:1297–304.
98.Almgren M, Kallen B, Lavebratt C. Population-based study of antiepileptic drug exposure in utero—influence on head circumference in newborns. Seizure 2009;18:672–5.
99.Gaily E, Kantola-Sorsa E, Granstrom M-L. Intelligence of children of epileptic mothers. J Pediatr 1988;113:677–84.
100.Scolnik D, Nulman I, Rovet J, Gladstone D, Czuchta D, Gardner HA, Gladstone R, Ashby P, Weksberg R, Einarson T, Koren G. Neurodevelopment of children exposed in utero to phenytoin and carbamazepine monotherapy. JAMA 1994;271:767–70.
101.Jeret JS. Neurodevelopment after in utero exposure to phenytoin and carbamazepine. JAMA 1994;272:850.
102.Loring DW, Meador KJ, Thompson WO. Neurodevelopment of children exposed in utero to phenytoin and carbamazepine. JAMA 1994;272:850–1.
103.Koren G. Neurodevelopment of children exposed in utero to phenytoin and carbamazepine. JAMA 1994;272:851.
104.Meador KJ, Baker GA, Browning N, Cohen MJ, Bromley RL, Clayton-Smith J, Kalayjian LA, Kanner A, Liporace JD, Pennell PB, Privitera M, Loring DW, for the NEAD Study Group. Fetal antiepileptic drug exposure and cognitive outcomes at age 6 years (NEAD study): a prospective observational study. Lancet Neurol 2013;12:244–52.
105.Koch S, Jager-Roman E, Losche G, Nau H, Rating D, Helge H. Antiepileptic drug treatment in pregnancy: drug side effects in the neonate and neurological outcome. Acta Paediatr 1996;84:739–46.
106.Dessens AB, Cohen-Kettenis PT, Mellenbergh GJ, Koppe JG, van de Poll NE, Boer K. Association of prenatal phenobarbital and phenytoin exposure with small head size at birth and with learning problems. Acta Paediatr 2000;89:533–41.
107.Horning MG, Stillwell WG, N owling J, Lertratanangkoon K, Stillwell RN, Hill RM. Identification and quantification of drugs and drug metabolites in human breast milk using GC-MS-COM methods. Mod Probl Pediatr 1975;15:73–9.
108.Svensmark O, Schiller PJ. 5-5-Diphenylhydantoin (Dilantin) blood level after oral or intravenous dosage in man. Acta Pharmacol Toxicol 1960;16:331–46.
109.Kok THHG, Taitz LS, Bennett MJ, Holt DW. Drowsiness due to clemastine transmitted in breast milk. Lancet 1982;1:914–5.
110.Steen B, Rane A, Lonnerholm G, Falk O, Elwin CE, Sjoqvist F. Phenytoin excretion in human breast milk and plasma levels in nursed infants. Ther Drug Monit 1982;4:331–4.
111.Finch E, Lorber J. Methaemoglobinaemia in the newborn: probably due to phenytoin excreted in human milk. J Obstet Gynaecol Br Emp 1954;61:833.
112.Meador KJ, Baker GA, Browning N, Clayton-Smith J, Combs-Cantrell DT, Cohen M, Kalayjian LA, Kanner A, Liporace JD, Pennell PB, Privitera M, Loring DW, for the NEAD Study Group. Effects of breastfeeding in children of women taking antiepileptic drugs. Neurology 2010;75:1954–60.
113.Committee on Drugs, American Academy of Pediatrics. The transfer of drugs and other chemicals into human milk. Pediatrics 2001;108:776–89.