TUCSON, ARIZ. — The use of carbamazepine, phenytoin, or phenobarbital as monotherapy by pregnant women was not associated with significant deficits in full-scale IQ in their offspring, a small study showed.
However, these data counter previous reports of lowered intellectual ability in children born to mothers who took anticonvulsant medications during pregnancy, Dr. Kenneth Lyons Jones said at the annual meeting of the Teratology Society.
“The results, then, should be considered in conjunction with previous studies when evaluating the risks associated with use of these medications during pregnancy,” cautioned Dr. Jones, chief of the division of dysmorphology and teratology in the department of pediatrics at the University of California, San Diego.
Despite the documented increased risk for certain major malformations following prenatal exposure to anticonvulsants, he said that “substantial controversy” exists regarding their impact on neurobehavioral outcomes.
“Compounding these concerns is the fact that in the United States, between 7.6 million and 12.7 million women have epilepsy, and 95% of them are being treated with anticonvulsant drugs,” Dr. Jones said. “In addition, 1 in every 200 pregnant women in the general population requires antiepileptic medication during her pregnancy.”
He and his associates studied 82 children aged 4–14 years whose mothers were on anticonvulsant monotherapy and contacted the California Teratogen Information Service (CTIS) during their pregnancy. Of the 82 children, 30 were exposed to carbamazepine, 23 were exposed to phenytoin, and 29 were exposed to phenobarbital.
The researchers randomly selected 50 matched controls born to women who contacted CTIS due to an exposure not deemed to be teratogenic.
All children underwent neuropsychiatric testing that included a measure of full-scale IQ (FSIQ). Testing was conducted at the Center for Behavioral Teratology at San Diego State University or in a location near the child's home.
Dr. Jones reported that 26% of children exposed to phenytoin were rated as affected with features of the anticonvulsant facies based on ratings of minor malformations, compared with 24% of those exposed to phenobarbital and only 3% of those exposed to carbamazepine. No one in the control group was considered affected.
In terms of the FSIQ, there were no statistically significant differences in scores between exposed children and controls, nor did any of the exposed children differ from each other, even after controlling for age and socioeconomic status.
Dr. Jones acknowledged certain limitations of the study, including the fact that the population studied was homogeneous, the sample size was small, and there was potential for selection bias.