Newer-Generation Antiepileptic Drugs and the Risk of Major Birth Defects
Mølgaard-Nielsen D, Hviid A.
JAMA. 2011;305:1996-2002
Study Summary
Less than 1% of pregnant women are on antiepileptic drugs (AEDs). These drugs are used not only to treat epilepsy, but also to manage migraine headaches, pain syndromes, and even mood disorders. Previous research demonstrated that older AEDs are associated with an increased risk for birth defects if women took them during the first trimester of pregnancy. However, newer AED preparations have been available since the 1990s, and it is not clear whether they have the same teratogenic potential.
This study used publicly maintained birth and health records in Denmark to evaluate live births from 1996 through 2008. The investigators identified women who were pregnant and taking one of the newer-generation antiepileptic drugs during the first trimester. The drugs of interest were lamotrigine, oxcarbazepine, topiramate, gabapentin, and leviteracitam. Mølgaard-Nielsen and Hviid evaluated the association between exposure to any one of these newer AEDs during the first trimester and subsequent risk for major birth defects among their offspring. The investigators were able to control for many potential confounders, including year of birth, maternal parity, maternal age, location, maternal education, socioeconomic status, smoking exposure, and similar demographic factors. Other potential confounders that were included in the analysis were maternal use of other medications before the women or healthcare providers realized that the women were pregnant (including exposure to older AEDs in the first trimester). The potential dose-dependent teratogenic effects of lamotrigine were evaluated by dividing the women into those who received less than 250 mg/day vs those who received more than 250 mg/day.
A total of 1532 infants were born to women who took a newer-generation AED during the first trimester. The comparison group included more than 800,000 infants born during the study period to women who were not exposed to AEDs. In the mothers exposed to new AEDs, 49 infants (3.2%) were born with major birth defects. Of the nonexposed infants, 2.4% had major birth defects. This bivariate (unadjusted) difference was statistically significant. However, the 95% confidence interval for the prevalence odds ratio was 1.02 to 1.80. After adjusting the analysis for potential confounders, no significant difference in the risk for birth defects between infants who were exposed and not exposed to newer-generation AEDs was found (prevalence odds ratio, 0.99; 95% confidence interval, 0.72-1.36). Even when the analyses were stratified by 3 individual drugs (lamotrigine, oxcarbazepine, and topiramate), no single drug was associated with risk for major birth defect. In addition, the investigators conducted extensive sensitivity analyses and found no association between exposure to AEDs and major birth defects, regardless of analytic approach. They concluded that there is no association between the use of newer-generation AEDs during the first trimester of pregnancy and the risk for a major birth defect.
Viewpoint
These data fall into the "file away for future use" category. Obviously, most pediatricians will not be prescribing AEDs to women, but adolescents do become pregnant. Therefore, these data may be used prospectively by neurologists during discussions with adolescents and their parents about choosing AEDs. Primary care providers can use these data to counsel patients when concerns arise about medication use and plans for pregnancy. The Scandinavian epidemiologic databases are comprehensive, allowing for powerful analyses such as this one. Although this study design is not considered as strong as a randomized controlled trial, this is one area in which ethics would not allow a randomized controlled trial, so epidemiologic studies will likely remain the best data available to answer this question.
Abstract