Major congenital abnormalities may be more common in children whose mothers received valproate monotherapy during pregnancy as compared to other medications, according to a study, Neurodevelopmental Effects of Antiepileptic Drugs, which was reported in the August 8, 2006 issue of Neurology.

In a prospective, observational trial, 323 pregnant women were enrolled in the study in which the authors note "to determine if fetal outcomes vary as a function of different in utero antiepileptic drug, AED, exposures." The women were taking 1 of 4 AEDs: valproate, carbamazepine, lamotrigine, or phenytoin monotherapy. Of these women, 81 percent were seizure free and only 3 percent experienced convulsions during pregnancy.

The incidence of serious outcomes, including fetal death and congenital malformations, was assessed in 333 children. While there were serious adverse outcomes in all groups, the incidence in the valproate group was 20.3% whereas the incidences for the other groups were: lamotrigine, 1%; carbamazepine, 8.2; and phenytoin,10.7%.

Even when the risk of individual adverse outcomes was evaluated separately, congenital malformations were significantly greater among infants exposed to valproate relative to those exposed to the other AEDs.

The study took place across 25 epilepsy centers in the United States and the United Kingdom which enrolled pregnant women with epilepsy from October 1999 to February 2004. Kimford J. Meador, MD, Chief of Service for the Department of Neurology at Georgetown University Hospital was the project leader. Dr. Meador is now with the University of Florida.

Based on their findings, the research team concluded: "Although valproate will continue to be an important treatment option in women who fail other AEDs, we advise that valproate not be used as the AED of first choice for women of childbearing potential, and, when used, its dose should be limited, if possible." Neurology 2006; 67:407 – 12.


Orrin Devinsky, MD, is Professor of Neurology, Neurosurgery, and Psychiatry at NYU School of Medicine. His epilepsy research interests include: quality-of-life, cognitive and behavioral issues in epilepsy, surgical therapy, and new medications. He has published widely in epilepsy and behavioral neurology, with more than 250 articles and chapters and more than 15 books. He has chaired several committees of the American Epilepsy Society and has served as a Board member.


What does this study mean for pregnant women and how will the study findings change your approach to the field?

The major finding of this study was in the higher rate of congenital malformations for valproate. This study supports a growing body of literature that shows that valproate appears to have the highest rate of congenital malformations among the commonly taken AEDs. It was also found that the malformations were dose related; that is, the higher the dose, the greater the chances for an adverse outcome.

For some women valproate is an excellent medication. For others it may be the only way to maintain seizure control. However, it does force neurologists to reconsider prescribing this to women in their child-bearing years. Each woman must be made aware that if she chooses to stay on valproate, she may be putting her baby at a higher risk.

What does this suggest for a woman who is currently on valproate and who wants to conceive?

I personally will not change my practice based on this study. If a woman is on valproate and comes to me and says, "I am getting married and I would like to have a child," I will talk to her about minimizing the risk to the baby. I would certainly want to talk about other medications or a lower dose. However, we need to keep in mind that a lower dose or different drug poses a potential risk of a seizure occurring. We need to balance risk and benefit. Even a 5 percent risk of a breakthrough seizure, if a woman has a long drive to work for example or spends much of her week driving around children, may be too great."

What about women who feel that they want to take the risk with valproate, become pregnant, and suggest being followed carefully with ultrasound?

The risks of birth defects with valproate extend beyond the widely publicized neural tube defects (such as spina bifida) that can often be detected on ultrasound. Valproate is also associated with other birth defects that cannot be identified by ultrasound and also with developmental delays that are not identified until months or years after birth.

Cynthia L. Harden, MD, an associate professor of neurology and neuroscience at the Weill Medical College of Cornell University in New York City, has been affiliated with the Comprehensive Epilepsy Center for over a decade. She has broad interests in epilepsy-related issues that include women's and children's concerns as well as the uses of new antiepileptic drugs. Dr. Harden is an ad hoc reviewer for Neurology and Epilepsy. She has also authored or collaborated on many book chapters and more than 35 articles for peer-reviewed journals.



What does this study mean for pregnant women and how will the study findings change your approach to the field?

This study reinforces what has been shown in other larger studies, particularly about the risk associated with valproate use in pregnancy.

If a woman on valproate wanted to become pregnant, I would ask her to discuss with me medication alternatives that would be reasonable for her. If she has never explored the possibility of trying different drugs, this might be the time. However, we need to take all factors into account. What if she had tried other medications but valproate was the medicine that stopped her seizures and she was feeling well on it? There are situations where we don't want to risk taking her off that medication.

What does this suggest for a woman who is currently on valproate and who wants to conceive?

With regard to medication -- if a woman wants to stay on valproate during pregnancy because she is doing well, what would I do? I would have to say, "I will support your decision, and follow your pregnancy with you very carefully. But we need to continue talking about whether or not this is the best decision for you."

What about women who feel that want to take the risk with valproate, become pregnant, and suggest being followed carefully with ultrasound?

With regard to prenatal testing, ultrasound is very good today. However, even if all of the bodily structures of the baby seem normal, it does not guarantee that the child will be free of learning disabilities. Ultrasound results do not guarantee that the baby will be perfect.