Abstract
Commentary
Identifying an association between SGA and seizures during pregnancy is novel. An association between women with epilepsy taking antiepileptic drugs (AEDs) and a doubled risk for SGA among in utero AED-exposed offspring has been found (1). Remarkably, the usual confounder for similar trials— maternal AED use—was not present in the Chen et al. investigation, since many women in this geographic region are untreated, and women with epilepsy who were taking medications were excluded from the group used for analysis (2). The study was performed by cross-referencing two national databases in Taiwan, those of women who had single births and those of women who were diagnosed with epilepsy between 2001 and 2003. The diagnosis of epilepsy in this subset of women was supported by documentation that included at least three consensus diagnoses of epilepsy or convulsions within the 2 years prior to the index delivery.
The risk of SGA among the offspring of women with epilepsy who had seizures during pregnancy, compared to women with epilepsy who did not have seizures during pregnancy, was mildly elevated, at an odds ratio of 1.34. Even the fairly tight 95% confidence intervals (i.e., 1.01–1.84) do not exclude the possibility of a moderate risk for an increased rate of SGA, although they do exclude a doubled risk. The authors also compared the women with epilepsy and a healthy control group and found an increased risk of low birth weight and preterm delivery among the women who had seizures, even after controlling for socioeconomic factors. Again, AED use was not a potential confounder; in fact, 84% of the women with epilepsy in this study (850 out of 1,016) were not taking AEDs. This dataset is one that likely would not be available in other parts of the world where similar surveillance of pregnant women with epilepsy is ongoing (e.g., the European Pregnancy Registry [EURAP] study and the North American Pregnancy Registry), as most patients in these trials are taking AEDs.
Additional valuable data could come from this dataset, such as whether there is a risk of birth defects related to seizures and epilepsy during pregnancy—a question that has long been unsatisfactorily answered, because the data have been confounded by the risks imparted by AEDs or by the likelihood that the epilepsy in the untreated group is mild or even questionable, making the comparison groups fundamentally different at baseline. Of note, previous studies have shown a lower risk of adverse pregnancy outcomes in untreated women, but the data have been confounded by the possibility that these women had less severe epilepsy. This study adds credence to this theory, since it appears that women without seizures (i.e., with milder epilepsy) did better than those with severe (active) epilepsy. The finding indicating that epilepsy, itself, is an adverse factor for pregnancy outcomes, in comparison to healthy controls, is provocative and needs further investigation and clarification.
This study provides additional evidence that can be used to counsel patients that it is best to try to maintain seizure control during pregnancy. These data are sorely needed, given that clinicians still struggle to communicate effectively with pregnant women with epilepsy to impress upon them exactly why preventing seizures during pregnancy is important, as prospective mothers usually are instead focused primarily on the risks to the fetus associated with AED use.