MONDAY, Dec. 1, 2008 (Health.com) — Pregnant women who take the epilepsy drug valproate may be more likely to have a child with autism than those who don't take the seizure-controlling drug, according to a study published Monday in the journal Neurology.
However, experts caution that the research is preliminary and needs to be confirmed.
"Women shouldn't suddenly stop taking their medication because they read this study," says Michael Goldstein, MD, a fellow of the American Academy of Neurology. "The consensus has always been that having convulsions is worse than the medication, so if a medication is needed to prevent seizures, it should be continued."
In the study, less than 1% of women without epilepsy had a child with autism, compared with 6.3% of women with epilepsy who had taken sodium valproate during pregnancy.
The study included 632 children; roughly half had a mother with epilepsy. Of those children, 64 were exposed to valproate in utero, 44 to lamotrigine, 76 to carbamazepine, and 65 to other epilepsy drugs or combinations of drugs. (There were 47 who were not exposed to any antiseizure medications.) The children were tested for autism at ages 1, 3, and 6.
Nine children of mothers with epilepsy developed autism, as did three children whose mothers did not have the seizure disorder. Compared to children whose mothers did not have epilepsy, the risk of autism was seven times higher in those with valproate exposure.
It has long been known that some antiseizure drugs can raise the risk of birth defects such as spina bifida (a defect of the spinal cord) and heart malformations. However, the drugs' impact on autism risk is less clear. Some animal research and case reports have hinted at the link, which prompted Gus Baker, PhD, of the University of Liverpool, and colleagues to undertake the study.
They say that their findings are preliminary and that the study is too small to conclude that other antiseizure medications are safe. (There was one case of autism each in children exposed to lamotrigine, phenytoin, and a combination of lamotrigine and valproate.)
Alycia Halladay, PhD, the director of research for environmental sciences at Autism Speaks, an advocacy group that raises money for biomedical research, says the study “adds to the growing literature that in utero exposure to [valproate] may be one possible risk factor for autism.”
The study results suggest that doctors may want to take additional precautions when prescribing medication to women who are pregnant or may become pregnant, says Dr. Goldstein. And some women considering pregnancy may want to taper their medication usage prior to conception, if possible.
"It's unlikely that stopping medication suddenly during pregnancy will be helpful. It's often already too late after the first trimester," he says. "The caution, though, is that this is a preliminary study, and a very small one. In the future, it may turn out to be nothing. We just don't know yet."
Autism Speaks is currently funding a number of studies that examine whether prenatal valproate exposure produces behavioral deficits in animals. The research focuses on whether there is a single critical period in development for susceptibility and how genetic backgrounds can influence the outcome.
"[Then] research can turn to the ways in which women who cannot stop or reduce [valproate] treatment while pregnant can ameliorate the risk of birth defects," Halladay says.