Content area
Abstract
[...]patients may also have higher and more consistent blood levels of the drug than patients receiving other antiepileptic agents, both because the severity of their epilepsy improves their compliance and because of the need for more frequent hepatic and hematologic monitoring with carbamazepine and valproic acid. Fetal exposure to antiepileptic drugs during the first trimester of pregnancy increases the risk of major malformations two to three times above the base-line risk of 2 to 3 percent in unexposed populations.1 However, the risk of major malformations among infants exposed to antiepileptic drugs during gestation is lower when the mother has received monotherapy (2.3 to 6.7 percent) than when she has received poly therapy (4.9 to 15.6 percent).2 3 4 5 On the basis of an analysis of prescription data, Rosa has suggested a specific association between carbamazepine and spina bifida among women enrolled in Michigan's Medicaid program at the time of delivery.6 Although his article addresses an important topic, we feel that it is analytically flawed and makes use of an inappropriate data set. Unfortunately, all carbamazepine-exposed infants with spina bifida were also exposed to other antiepileptic drugs. Because phenytoin, phenobarbital, and primidone were not associated with spina bifida in this study when analyzed separately, the author has assumed that the effects of carbamazepine alone could be evaluated independently, regardless of exposure to other antiepileptic drugs. The important topic of the teratogenic effects of antiepileptic drugs should be evaluated in prospective studies in which drug levels are monitored, particularly during the first trimester of pregnancy; only one pregnancy per woman should be evaluated; and pregnancy outcome should be assessed by investigators blinded to maternal exposure.