INTERVIEW WITH DR. KIMFORD MEADOR – The Truth About Antiepileptic Drugs And Pregnancy


Dr. Kimford Meador is a Professor of Neurology and Neurosciences at Stanford University, and Clinical Director, Stanford Comprehensive Epilepsy Center. You can read Dr. Meador’s complete bio by clicking HERE.

Interview with Dr. Meador by: Simone L Graham – Executive Director HIS

This article addresses the following questions:

-What led to writing the proposal for the NEAD study?

-Is there sufficient research on this subject?

-Are you concerned about women with epilepsy having children, now that we know the drugs can cause cognitive damage before birth? What about surrogate mothers?

-I attended a conference by the Epilepsy Foundation of Greater Los Angeles in October 2017. The last part of the conference scheduled smaller group sessions for us to attend. I went to the small group on women’s issues. During the presentation, there was discussion on AEDs and their effect on pregnancy. One of the young women there using Depakote (valproate) was unaware that the drug can cause malformations and cognitive damage in a baby before birth.

-Do we know that any of these drugs are not going to harm the child? Better question, we have not studied all the drugs, have we?

-Do you know what part of the brain it affects?

-So, you are able to test to see what happens, you just do not know what exactly  is happening in there.

-How many do you estimate might have this cognitive effect?

What led to writing the proposal for the NEAD study?

Dr. Meador was interested in behavior and cognition studies and originally trained in behavioral neurology studying, for example, the effects of strokes. He became interested in epilepsy because it was one of the few diseases that could be recorded through brain wave activity during different behaviors. As he began reading adult epilepsy research studies, he became concerned that seizure drugs could be changing behaviors.

“I had conducted several adult studies and reviewed the literature to see if anticonvulsants have cognitive effects on adults.”

He found evidence that the older drugs had a substantial effect on adults, but some newer medications did not.

“The amount I was seeing was significant, but it was not huge. I thought that maybe it might affect children as they were growing.”

He began to look at the literature, but current child studies did not shed much light on his search. He says there still is not fantastic literature on the effects of seizure drugs on children. Dr. Meador did find animal studies fetal exposure to anticonvulsant drugs. The drugs were given at dosages that did not cause malformations, but the animals were not as smart when they matured to adulthood and put through rat maze memory tests. He wondered if it could happen in humans, as well.

He hesitated to pursue his own study for a few years because no one else was launching this type of behavioral study. His curiosity prevailed, and he began to assemble his research team.

“I went and found all the experts I needed. pediatricians, obstetricians, teratologists, geneticists, pharmacologists, and statisticians, and we proposed this study to look at the outcomes in the children. That was the NEAD study.”

Is there sufficient research on this subject?

I was so surprised by Dr. Meador’s answer revealing information I had never read or heard before, that I decided to include his responses exactly without attempting to paraphrase them. The comments about how long scientists, doctors, and pharmaceutical companies knew this drug was causing harm to babies born to women with epilepsy really caught me by surprise. I was under the impression that the medical world began to suspect in the late 1990’s that AED’s in general may be affecting the brains of newborns exposed before birth. The truth is much worse than that. I hope you find our conversation as interesting as I did. 

I have written about this, I wrote an article a couple of years ago, the way we are approaching this is inadequate, in my regard. We are getting slow answers.  In valproate we knew that it caused spina bifida early 1980’s. Then it took twenty something years to for us to get to the point that it causes more malformations and it causes more than any other drug. That is twenty something years of kids being exposed, about twenty-five years. So, that is pretty depressing. And the same thing, there was an inkling about maybe cognitive stuff and the differential of looking at them really took decades. We should be faster at that. We have new drugs coming up. If a woman, my patient, goes through the few drugs we know something about on the pregnancy outcomes, then I have to pick drugs where we have inadequate data. Most of the drugs we have inadequate data.

I feel strongly that there should be a reporting requirement for malformations in babies. Then we figure out what drugs the mother is taking at time of conception. (They aren’t required to report malformations?) No, it is not required at all! They have some things they do voluntarily, but we are nowhere near getting the population data that will really give you numbers to the point where you will figure this out. It takes numbers to look at malformations to see if they are related to the drug. We don’t do that in the United States. There are only a few countries that actually do it. That is Scandinavia and most of the advance has been by voluntary registry. So, they depend on people, the women, in the registries signing up and following up, that kind of thing, and it means that less women do that, and we are not getting the whole sample, so we miss things. We require reporting for all kind of diseases, like TB and such, which probably have less of an impact than malformations in the United States right now. We should have better information on that. Also, I think you can see in our study, which is probably the most successful in the world looking at cognitive outcomes in these kids, how tenuous our funding is to do this.

There are good animal models for some of the effects of the drugs and only a small handful of the drugs have been tested in those animal models. I see people put in grants for basic science that do not get funded in the NH system. It should be part of the regular regulatory requirements of the testing in looking at these things so that we understand it, so we don’t have to go generations down the road to find out ‘oh my God’ this drug really messes up the babies. Nobody wants that.

When I met women that had a problem with their child with valproate exposure, the interesting kind of thing I find out is that they feel guilty. I ask them, why do you feel guilty? How were you supposed to know? You Know? How are you supposed to know? But it is a common response that they feel guilty, very common. So, you can look back a couple of years ago I wrote it in Neurology about this with specific recommendations about how things can be made better, in my opinion, (than) where we stand right now.”

Simone: 3A. Are you concerned about women with epilepsy having children, now that we know the drugs can cause cognitive damage before birth? What about surrogate mothers?

I have rarely made that recommendation and I have never had a woman who I was following have a surrogate mother. I might make the recommendation of surrogate pregnancy if they were on multiple drugs and especially if we know some of the drugs are bad in terms of outcomes. Then I would be really worried about the outcome. Of course, it is not the exposure the woman had any time before the pregnancy, so the egg is not affected. It is when the fetus starts growing. However, I think the majority of women who have epilepsy can have their own children. There needs to be pre-pregnancy planning to reduce the risk as much as possible. Remember, historically, we had several states, I think nine states that had sterilization laws in the 1950’s for women with epilepsy. The last law forbidding marriage in people with epilepsy ended in the 1970’s in England and in the 1980 in the last state in the United States. So, I am cautious not to tell these women they cannot, because some of them really want to have their children. When I was very young as a medical student, I met a woman, they did a hysterectomy on her because she had epilepsy. She was the best aunt ever because she loved children she would go and do everything with her sister’s children. She always wanted to have children herself. So, that is horrible when the majority of kids are being born normal. The vast majority of the kids are born normal to women with epilepsy. So, they should be able to, in most cases, without problem. Now our new studies show there is not an increased risk for OB or neonatal problems.

So, I encourage women if they want to have babies, “Let’s do it, but let’s plan ahead.” They have to have good information. Too often people put it off until they get pregnant. Well, you cannot do that! You have already missed a big window to do alterations to make sure they are taking folic acid, to make sure you are on the best drug possible to make sure you have good control. If you have good control of your seizures going in, you have not had seizures for nine months, 85-90% of women are not going to have any seizures during pregnancy. So, you really want to try to get good seizure control before you go in to the pregnancy to make sure you are on the best drug and get on folic acid, all these things, and if they understand once they get pregnant for some of the drugs it is very important that they tell the doctor and to follow-up to get the levels checked so they can adjust them to prevent them from falling too low to prevent a seizure.

I attended a conference by the Epilepsy Foundation of Greater Los Angeles in October 2017. The last part of the conference scheduled smaller group sessions for us to attend. I went to the small group on women’s issues. During the presentation, there was discussion on AEDs and their effect on pregnancy. One of the young women there using Depakote (valproate) was unaware that the drug can cause malformations and cognitive damage in a baby before birth.

There are a lot (of women) that do not know. We did a study a few years ago in the Florida Medicaid population because we could get those records and the use of valproate for women who were taking it for epilepsy had fallen. But the use of valproate for women who were taking it for psychiatric indications had gone up. It actually went up! I think most of these scripts are written as an outpatient and mostly not written by neurologists and they aren’t aware of this data. They are writing a drug for which they do not know the risks, the major risks. So, that is really a shame. That is, basically, that woman being put on the drug and not knowing about this. That is basically malpractice. By the time she gets pregnant it is already too late. The kid probably already took the hit for the malformation.

Where the cognitive autistic stuff happens is not completely clear. There is some data that suggests that it happens in the third trimester, but it is not completely clear. There hasn’t even been a systematic study in rats to see which trimester it causes the cognitive problem.

Epilepsy is underfunded. The North American registry is not funded by government. It is funded by a conglomeration of drug companies. They kind of run it on a shoestring overall trying to run these things You don’t see any national advertisements saying ‘Join this’ because they do not have the money to do that. Plus, the IRB* doesn’t want them doing that; oh, you are not being fair to these women. They slowed it down, the IRB that runs that thing slowed it down, the enrollment, by saying the woman has got to call on her own. The doctor cannot call. She cannot call from the doctor’s office because that would be coercion. That would be coercion to get this study, this no risk study. The worst thing that could possibly happen is to give up her personal information. Other than that, there is no risk. They are just collecting data. They are not doing anything to the woman. They are not changing the drugs, not doing anything, just collecting data.

Because they made the women go home and call, I sent several women home telling them you need to call this is so important. They get home, the baby is crying, they have other kids and they do not call. So, they do not enroll. Much less do we have a national requirement for reporting. We do not have that. So, we have got cobbled together funding out of the registry, they are hampered by the IRB that think they are protecting these women when, in fact, in the long-term they have hurt these women and other women because they have slowed the access of this information that would help people make good choices about their medication. So, that is the kind of irony of their intent, the effect ended up doing the opposite of what their intent was. That is too bad, I think. I am saying that with the complete understanding of how important it is to not coerce people into this. At this point, if you have to go through all those hoops it really slows it down. EURAP has done a better job because they set it up differently. They have the doctor just get the information and send it to them. Then they had all the doctors join so they can do the informed consent locally. Again, most of it is not a complete population-based study, there is a complete population in there, but they are across 80 countries. So, they get big numbers and they are exceeding the U.S. registry by far, in part because they have larger catchment, but also in part because they have a different recruitment method. Their IRB is set up in, I think, Sweden and Italy are the two main countries that started this.

There are better ways to do it. We should be thinking about better ways to do it.

I didn’t realize it was that slow.

It seems fast because so much has happened since 2004 or so after decades of nothing happening. Still, when you are in an office and you have run out of all the drugs that you know are safe and you are trying to avoid the ones that you know that are pretty bad, then you have a whole clump of the drugs that you do not really know about. It would be really nice to know more about those faster.

*NOTE: IRB = Institutional Review Board. Research projects must be approved by each institution’s IRB. This is not related to funding. The national registry for malformations related to antiseizure drugs (North American AED Pregnancy Registry) is funded by a consortium of pharmaceutical companies.

Do we know that any of these drugs are not going to harm the child? Better question, we have not studied all the drugs, have we?

No, we have not. We are not even close to studying them. The two or three we have the best information on, that includes both malformations and cognitive outcomes include lamotrigine, levetiracetam, and carbamazepine. We have malformation data on oxcarbazepine and no cognition data. It has never even been tested in the animal model that I have talked about. The alcohol fetal apoptotic model I have talked about. That is a good predictor, in general, about these things. There are some lingering concerns that maybe lamotrigine has some effects, but there is one animal study, not apoptosis, it does not do that, but it does affect physiology of the developing fetus in some way. We had a little data in ours concerned it might affect language to some extent. We are going to find out about that in this new study. We will know for sure about that. I don’t know that you can ever say that they are completely safe, but there are drugs that have very low risks. The malformation risks for lamotrigine and for levetiracetam are either at or just above the general population risk 2%. They are at around 2 – 3%. It is dose dependent, so bigger dosages, more risk. The EURAPS study actually showed that because they had large numbers and could show that even lamotrigine has a dose dependent effect even though it is the best for the ones they have tested. It has a dose dependent effect. So, there is always some risk. We are always balancing those risks back and forth between the two and we’re doing that also through a lot of it we don’t know the risks. We have certain risks we know, and we react to those things, then we have this big conglomerate of unknown risks at this point.

Wouldn’t it be great if we could get the information out sooner as we start to use these drugs? Typically, a drug will come out and it gets used in patients that have failed other drugs and moved to that drug to get a lot of experience with it. The same thing is true for pregnancy. But with a lot of women with epilepsy, you burn right through the first drugs because they don not work or the cause other side effects. Now what do you do? When you get to the next level, you are guessing next. So, I spend a lot of time kind of guessing in the clinic what might be the next best thing to do. It would be nice to have more information, or at least to be on a steady track to get more information at a rapid pace so you would know as soon as possible and not have these long delays where babies get exposed to something that we don not know for sure about the risks. We do not put women on these drugs for nothing. They have some other problem and we are trying to treat that. We have to balance that against the risks the drug gives also to the baby.

Do you know what part of the brain it affects?

We are guessing from the animal data that it is a diffuse effect. It does not affect all the nerves; the apoptosis is diffuse and scattered around. There are certain regions that are more susceptible, but it is not the whole brain. But what is most important is not the cells that die, which is not that huge; it is what happens to the remaining cells that do not work right. So, in the case of alcohol, and in the case of valproate, that is more than likely what is going on causing these problems. We do not have good research looking at brain anatomical structure, white matter connection, proximal synaptogenesis in humans. We know that they suffer these cognitive deficits for some drugs that have been tested. They do not for some other drugs, but there are still huge gaps in how we understand exactly what is going on here.

So, you are able to test to see what happens, you just don’t know what’s exactly happening in there.  

Right. Structurally, we do not know what is happening in there. And also, it would be good to be doing research to see if there are ways we can block that or treatments we can do to reverse some of the symptoms when we get to that point.

How many do you estimate might have this cognitive effect?

Well, we can come up with a number of how many were exposed, but how many suffered the effects is a whole different thing. Even for a drug like valproate where there is substantial risk, it depends how you define what the risk is. Are you going to say, how many are mentally retarded? Well, that is a small subset. Let’s say the kid’s parents have an IQ of 120 and the kid has an IQ of 100. That is very normal, but it is not normal for him given his parent’s IQ. And then you have these behavioral deficits that are just as devastating as the cognitive deficits on top of that. So, it is hard to estimate a number. Although, I would on the optimistic side say that most of the babies born to women with epilepsy are normal although they increase risks for these things; they increase risk for cognitive stuff, increase risk for behavioral problems, increased risk for attention deficit disorder. But if we understood which drugs are safer, then we can reduce this number by trying to use the drugs that are safer, on one hand.

We also should have a high suspicion if a kid is having any kind of delay if you have had any exposure like this in pregnancy. We should be looking at those kids and interceding early for any kind of cognitive or behavioral delay kind of thing, because when we intercede early the kids do better. I have had mothers come and ask me, “I’m worried about my baby talking, he is two and a half or three years old, my physician says let’s watch it”. I tell them no. Go back and tell them you want a referral to get an evaluation because the baby is at a higher risk. So that is where we are at this point with that.”

Remember that a lot of these babies exposed turn out completely normal. I had a lady come up to me a few years ago at a conference and say, “I was exposed. My mother took phenobarbital when I was a fetus. I’ve done OK.” She had a PhD and was working at the university. She did very OK and so there is this variance that we do not understand. The teratogen whether it is causing a birth defect, or a behavioral cognitive problem is dose dependent and on a lower dose less likely. But it also depends on the genetics of the baby. So, some people have genes that make them more susceptible to the effects if the drug exposure than other people. When I look at what we call a scatter plot, it is not like everybody is in one spot for one dose. They are scattered around. Some do very well, some do God awful. And overall you average that together that is lower for a drug like valproate, but some of the valproate kids do well all because their genes are different and so is their susceptibility. We are nowhere near getting to the point where we go like, OK, these genes are possible, so we can stay away from that drug. We try to do preventive kind of thinking. We try to pick the drugs with the lower risk and try to keep getting more information, so we can have more picks to choose from.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.