By Shawna Cutting

I recently spoke with Dr. Kimford Meador, Chairman of the Department of Neurology at Georgetown University. He is running a national study on how epilepsy medication affects the unborn child, both in the womb and after birth. This study is known as NEAD (Neurodevelopmental Effects of Antiepileptic Drugs). It began in the fall of 2000 and is continuing to enroll patients.

What got you interested in studying the effects of seizure medicines?

That's a long, winding story, actually. I started working in epilepsy because epilepsy and epilepsy surgery can tell us a lot about behavioral issues. I became concerned with the effects of antiepileptic drugs on adults. The research literature on that topic at that time was lousy. And so we began a series of studies looking at those effects. We found that the old drugs all have cognitive effects [such as effects on attention, speed, and learning] that are generally modest, but they can be significant for some patients. Over the years I began to think that these effects might be dramatic in children while their brains are developing, because they could add up over many years. That made me think that the effect might be even greater in a fetus because brain development there is so rapid. Studies of animals showed clearly that some antiepileptic drugs could affect behavior of the offspring if the mother was exposed during pregnancy. That concerned me and I began to work on this project because it didn't seem that anybody else was studying this subject. I got the help of other epileptologists and pediatric neurologists; we put together a grant and received funding to try to sort out this issue.

So there is no agreement on the effects of these medications on unborn children?

There is not even agreement about the anatomical effects of antiepileptic drugs on the unborn child—what do they do to the structure of the body? In animals, all the older seizure medicines produce anatomical defects, and some also produce behavioral effects. But you can't apply results to different species because they may not apply in the same way. The effect could be less in humans, or it could be worse in humans. We don't know. It really puts pregnant women in a bad spot and we physicians taking care of these women are also in a bad spot because we don't know how to advise them. They can't quit taking the drug—the risk of seizures is worse than the risk of the drug. I should point out that overall the majority of babies are normal, but we want to reduce that risk even further. The risk of anatomical defects may be much smaller than the risk of behavioral effects.

What is your study looking at?

Our study, the NEAD study, focuses on the major medications being used right now with epilepsy and the pregnant woman. We want to know if there is a difference between drugs. We're not testing polytherapy [taking more than one medication at the same time] versus monotherapy [taking only one medication] because it is clear even from studies in humans that polytherapy has a higher risk of anatomical defects. Our study is looking for something more subtle—a behavioral change.

A person's behavior changes throughout their lifetime. How long are you following the mothers and children?

Very good question. The original grant only follows the children until they are 2 or 3 years of age. But it's clear that we need to follow them to at least 6 years of age. This age is so important because this is when measures such as IQ begin to match up with adult measures. If you measure a child's IQ at 3 years of age, it may not predict the child's development. But a measurement at 6 years of age statistically will predict what will happen when this kid is an adult. It's also a very important point because that's when kids enter school. Whatever is going on at that point is obviously going to affect their learning.

How do you look at behavior and learning capabilities?

For behavior, we mainly use questionnaires. We ask parents and teachers to complete a set of standard questionnaires. This gives us a sense of whether the kids are having behaviors that are abnormal or even mildly abnormal, such as hyperactivity, or whether they have more severe behavioral problems. So that's how we get at the behavioral component.

For the cognitive aspect, one of the major measures is an IQ measure, because it's predictive of what school performance is going to be. It's not the only measure we use, because it doesn't cover all cognitive abilities. So we have other measures to look at the broader scope of what's going on with the child. For example, IQ and memory may be separate. A child can have a high IQ and have a really horrible memory, or just a mild memory problem. If a child has bad memory, though, over time it'll probably affect the IQ. The child won't continue to learn and eventually it will lower the rest of the cognitive abilities. Another time when IQ and functioning may be separate is in patients who have frontal lobe damage. Their IQ may remain normal, but their frontal lobe functions may be horribly impaired in terms of how they direct themselves. A lot of these effects come out in behavior. But there are certain tests that can capture components of that too.

Are you also looking at the amount of medication the mother takes and other things that might go along with pregnancy?

Yes. It's hard to get all that information once the child is grown, or even when the child is 6 or 8 years old. So we are collecting information on seizure frequency, on the mother's seizure type, nutrition information, what medicines and dosages she has taken, if she is taking the medicine regularly, blood levels, and what effects she's experiencing at various blood levels. So there are a variety of measures that we are trying to look at.

What medications are you looking at?

We are studying four drugs. These drugs are the four most commonly used by women with epilepsy who are pregnant. They are carbamazepine (Tegretol or Carbatrol), phenytoin (Dilantin or Phenytek), valproate (Depakote or Depakene), and lamotrigine (Lamictal). There are generic versions of several of these, in addition to the brand names here. Phenytoin is the oldest of the four—it's been around for over 70 years. Carbamazepine also has been around for quite a while, and it's the one most commonly prescribed for pregnant women with epilepsy. Of the new drugs, lamotrigine is the one that's used most often by pregnant women.

We have a few reports of pregnancies with other new antiepileptic drugs, but the number is not as high as for the other four drugs so we haven't added those. If we try to look at every drug and there aren't many people taking some of them, we won't have much information when we get done. When the sample size is so small, we can't make good conclusions. We'd like to expand the study in the future and put in additional new drugs, but right now other new drugs are not being used at the same rates. They may be used a lot, but not so much in women who are pregnant.

What are the incentives for a woman to enroll in the NEAD study?

I think the major incentive is the increased attention to the child. The women also get a small amount of money to help out with travel, babysitters for other children, and things like that. It's not a huge amount, but it is several hundred dollars over the course of the study. And the woman is also helping out women in the future, so we can give them real answers. Right now, we can't even give them good guesses, because we don't know.

Are you running any other studies?

We are doing another study involving older kids. We saw a disturbing report from England about a study that suggested that the drug valproate (Depakote, Depakene) was producing worse effects. This study was retrospective, meaning that they were testing kids between 6 and 16 years of age who were exposed to an antiepileptic medication sometime in the past. Retrospective studies can have some problems, partly from patient selection and partly from missing information. So we don't know if the results of this English study were real or if they were the result of problems with the study itself. But we are concerned because the results were fairly dramatic. We are going to do a small retrospective study of our own to see if we can reproduce those results. If we see the same pattern that they saw in England, we would be more concerned that the findings weren't the result of a sample selection problem. So some of the centers involved in the NEAD study also are enrolling moms and their kids between 6 and 16 who were exposed to carbamazepine (Tegretol), phenytoin (Dilantin), or valproate (Depakote, Depakene) during the mother's pregnancy.

How can a woman find out whether she and her child are eligible to participate in the retrospective study?

If she goes to www.neuro.mcg.edu/np/NEAD.htm and looks at "Independent Retrospective Study" she will find more information about eligibility. Then she can talk with a study coordinator in her region or at the main site to find out if the nearest center is participating.