Abstract
Administration of antiepileptic drugs to pregnant animals in lower doses than those that produce congenital malformations is associated with cognitive and behavioral deficits and other poor neurodevelopmental outcomes. However, the effect of exposure of human fetuses to antiepileptic drugs on cognitive and behavioral abnormalities is unclear, and there are little data to guide the physician in choice of antiepileptic drugs in women who are pregnant or may become pregnant. This study is a planned interim analysis of data from the Neurodevelopmental Effects of Antiepileptic Drugs (NEAD) study, a prospective observational multicenter cohort study of the neurodevelopmental outcome at age 3 years of children who had been exposed in utero to one of several antiepileptic drugs. This study enrolled pregnant women with epilepsy who had been treated with carbamazepine, lamotrigine, phenytoin, or valproate between 1999 and 2004. The primary study outcome was cognitive performance of the children at 6 years of age. Of the 309 live births, cognitive assessments were conducted in 258 children.Children who had in utero exposure to valproate had significantly lower IQ scores than children exposed to the other antiepileptic drugs. After adjustment for maternal IQ, maternal age at delivery, drug dose, gestational age at birth, and maternal use of folate before conception, the mean IQ was 92 for children exposed to valproate, compared to 98 for those exposed to carbamazepine, 101 for those exposed to lamotrigine, and 99 for those exposed to phenytoin. The effect of in utero valproate exposure on IQ was dose dependent. Maternal IQ was strongly related to child IQ for each antiepileptic drug taken during pregnancy except valproate.The investigators conclude from these findings that valproate should not be a first-line antiepileptic in women of childbearing potential not only because of its known association with neural tube defects, but also because of its negative effect on cognitive impairment.