ARTICLE IN BRIEF
An updated analysis of outcomes from the North American Pregnancy AED registry found that the highest risk of malformations associated with AEDS was for valproate, at 9.3 percent (30/323), followed by phenobarbital at 5.5 percent (11/199), topiramate at 4.2 percent (15/359), carbamazepine at 3.0 percent (31/1033), phenytoin at 2.9 percent (12/416), levetiracetam at 2.4 percent (11/450), and 2.0 percent (31/1562) for lamotrigine.
Is it safe for women of childbearing age to take antiepileptic drugs (AEDs)? This question has received increased scrutiny in recent years, as new drugs and reports of associations with fetal defects come to light with the help of pregnancy registries.
In a new analysis of outcomes from the North American AED Pregnancy Registry, however, investigators offer reassurance that drugs are becoming safer: The newer generation AEDs were associated with a lower risk of birth defects than older, more traditional AEDs, they reported in a May 2 online paper in Neurology.
The study, led by Sonia Hernández-Díaz, MD, DrPH, associate professor of epidemiology at the Harvard School of Public Health, updated the registry's initial findings reported in Archives of Neurology in 2004.
Commenting on the findings, Jennifer L. Hopp, MD, assistant professor of neurology at the University of Maryland School of Medicine, who was not involved in the study, said that the new data were important because they included more women, additional and statistically stronger samples for some of the older AEDs, and reports on newer anticonvulsants. This study, she added, will help physicians “make more informed decisions about risk-benefit ratios for our patients, and allows us to counsel patients on the specific kinds of malformations and problems. It also lets patients make more informed decisions in working with their physicians.”
Between 1997 and 2010, this cohort of nearly 5,000 eligible women (out of 7,370 in total) self-enrolled early in their pregnancy and were followed through the duration and then up to 12 weeks post-partum. The women responded to a computer-assisted interview at enrollment, at seven months' gestation, and at eight to 12 weeks after the expected date of delivery.
Women were asked to provide information on dosing and frequency of medication, number and type of seizures during pregnancy for those with epilepsy; demographic characteristics; habits, such as smoking, alcohol and drug use; medical conditions; use of other medications; family history; and the results of any prenatal testing. The authors also recruited for comparison a reference group of friends and relatives of the AED-exposed women, who were pregnant but did not have epilepsy and were not taking any AEDS.
The findings showed the highest risk of malformations associated with AEDS was for valproate, at 9.3 percent (30/323), followed by phenobarbital at 5.5 percent (11/199), topiramate at 4.2 percent (15/359), carbamazepine at 3.0 percent (31/1033), phenytoin at 2.9 percent (12/416), levetiracetam at 2.4 percent (11/450), and 2.0 percent (31/1562) for lamotrigine. For the unexposed control group, the risk of malformations was 1.1 percent (5/442). [See “AEDs and Risks of Malformations.”]
“Even with the small sample sizes for specific AEDs, the risks associated with use of the older AEDS valproate and phenobarbital were so high that the difference with newer AEDs was visible,” Dr. Hernández-Díaz said.
“While the lower risk associated with newer drugs is reassuring, the increased risk of clefts observed in newborns exposed to topiramate early in pregnancy has raised some concern,” she added. “This was an unexpected finding, but to keep things in perspective, even a ten-fold increased risk in oral clefts would translate into a risk in the order of 10 cases per 1,000 topiramate-exposed pregnancies.”
Since there were only 359 women taking topiramate in the current registry, the authors said the association found with oral clefts needs to be replicated in further studies.
Dr. Hernández-Díaz said the investigators hope to expand the number of women who are enrolled in the registry. She added that knowing more about the mechanism behind the associations would help in developing potentially safer medications and also in understanding the origins of malformations and specifically oral clefts. “Perhaps if we find the underlying mechanism, we can work towards preventing some of these malformations,” she said.
“This registry, along with others, provides remarkably good information to help direct the care of women with epilepsy at childbearing age,” Kimford J. Meador, MD, professor of neurology at Emory University and director of university's Emory Epilepsy Center in Atlanta, GA, told Neurology Today.
Since it is impossible to do a randomized trial of this population of pregnant women, Dr. Meador said, more data from registries should be gathered. More numbers are needed to both gain certainty about individual malformation risks and, specifically, in the drugs where the samples were too small to measure an effect — such as, for example, clonazepam, zonisamide, gabapentin, and oxcarbazepine. The idea is to establish accuracy with repeat results, he said.
Dr. Meador noted that the percentages for specific defects are smaller in the North American registry than they were for a European registry, EURO-CAT; the study authors said this likely had to do with the strict outcome inclusion criteria. “Also, registries rely on volunteers to participate; a population that might have lower risk of malformations,” they wrote. There is also some speculation, that self-enrolled registries would have a bias towards better-informed women, experts noted.
In the registry, there were also women who take AEDs but don't have epilepsy, Dr. Hopp pointed out, around 6 percent with mood disorders and 1 percent with migraine. “I think that one area for future research may be to get more data about AEDs in pregnant women with these other disorders.”
Dr. Hopp said patients are coming in informed about malformations and cognitive risks — and they ask good questions. “Getting updated data and in larger numbers of women from this study allows me to be more confident in quoting the risk of malformations to my patients and guiding them — whether it's in continuing on the drugs that they are currently already taking or potentially considering a switch that might allow us to reduce malformation risk,” Dr. Hopp told Neurology Today.
Listen as lead study author Sonia Hernández-Díaz, MD, DrPH, associate professor of epidemiology at the Harvard School of Public Health, discusses how neurologists can use these findings in practice, as well as the future research goals of the North American AED Pregnancy Registry. http://bit.ly/JcBDJc
Jennifer L. Hopp, MD, assistant professor of neurology at the University of Maryland School of Medicine, explains why the data from the North American AED Pregnancy Registry are so important for neurologists, and how she counsels her female patients with epilepsy before and during pregnancy. http://bit.ly/Kp4ar4
HOW TO COUNSEL FEMALE PATIENTS?
It's important to keep in mind that at least in the US, nearly half of all pregnancies are unplanned, said Sonia Hernández-Díaz, MD, DrPH, associate professor of epidemiology at the Harvard School of Public Health. “That means that whatever we recommend for pregnant women based on the safety of these medications, we have to apply to all women of childbearing age,” she said.
“I tell patients that women with epilepsy in general do have healthy pregnancies, but there are things that we can potentially do to help minimize risk of seizures as well to reduce risk of malformations for their babies,” said Jennifer L. Hopp, MD, assistant professor of neurology at the University of Maryland School of Medicine. She provides her patients with past and updated data from the North American and worldwide registries, and encourages pregnant women with epilepsy to join the North American registry.
Kimford J. Meador, MD, director of the Emory Epilepsy Center and professor of neurology at Emory University in Atlanta, GA, said he informs his patients about relative risks associated with different antiepileptic drugs and also provides them with registry data. “It's clear that valproate poses the greatest risk to the unborn child, but this study has given us a better feeling about some of the newer drugs because the risks appear to be fairly low.
Although patients should have access to all information on the risks and benefits of antiepileptic drugs, Dr. Hernández-Díaz said neurologists do need to sit down with the individual patient and go through the relevant data in a comprehensive manner in order to avoid raising any unnecessary alarm or sudden medication termination.