Pregnancy outcome after in utero exposure to valproate: Evidence of dose relationship in teratogenic effect

Background: Valproate is a first-line antiepileptic agent and is also used in the treatment of bipolar disorder and migraine. It is a known human teratogen. The objective of the study was to evaluate the teratogenic risk of valproate. Methods: All callers who contacted the Israeli Teratology Information Service (TIS) between 1994 and 2004 for information about gestational exposure to valproate were enrolled in the study. After the expected date of delivery, these women were followed up by telephone interview about their pregnancy outcome using a structured questionnaire. Data obtained from women who contacted the TIS about valproate exposure during pregnancy were then compared with data obtained from callers who were counselled for nonteratogenic exposures over the same timeframe. The main outcome measure was the rate of major congenital anomalies. Results: The outcomes of 154 valproate-exposed pregnancies (96.1% at least in the first trimester) were compared with those of 1315 pregnancies of women in the TIS database who were counselled for nonteratogenic exposures. The rate of major anomalies (some multiple) in the valproate group exposed in the first trimester was higher compared with controls after exclusion of genetic or cytogenetic anomalies (8 of 120 [6.7%] vs 31 of 1236 [2.5%], p = 0.018, relative risk [RR] = 2.66, 95% CI 1.25, 5.65). There were no cases of neural tube defect in the valproate-exposed group. Five of the eight major anomalies in the valproate group were cardiovascular, two of eight were mentally retarded, two of five male infants with major anomalies had hypospadias and three of eight were suspected of having fetal valproate syndrome. A daily dose ≥1000 mg was associated with the highest teratogenic risk (7 of 32 [21.9%] vs 31 of 1236 [2.5%], RR = 8.72, 95% CI 4.16, 18.30). In the subgroup exposed to polytherapy there was a 4-fold increase in the rate of major anomalies compared with controls. All major anomalies were in the group treated for epilepsy. Conclusion: When valproate treatment cannot be avoided in the first trimester of pregnancy, the lowest effective dose should be prescribed, preferably as monotherapy, to minimize its teratogenic risk.