ARTICLE IN BRIEF
Investigators reviewed birth defects among infants whose mothers took lamotrigine, which was approved in 1994 for partial seizures, or oxcarbazepine, topiramate, gabapentin, or levetiracetam, in the first trimester of their pregnancy. They found no increase in risk of major birth defects among children of mothers taking newer AEDs, and in infants of nonepileptic mothers.
A Danish population-based study has found little difference in birth defects among infants born to mothers who took second-generation antiepileptic drugs (AEDs) during their first trimester of pregnancy compared to those not exposed. Not all researchers are comfortable with the findings, however, and the US FDA raised the risk level of one such drug, topiramate, in mid-March.
Several epileptologists, who were not involved with the study, said the new findings should be read with caution, since the study design itself was problematic and the data conflict with earlier analyses from other registries, which led to the FDA warning.
The new study, published in the May 11 issue of the Journal of the American Medical Association, reviewed birth defects among infants whose mothers took lamotrigine, which was approved in 1994 for partial seizures; oxcarbazepine; topiramate; gabapentin; or levetiracetam, in the first trimester of their pregnancy. The investigators found no increase in risk of major birth defects among children of mothers taking newer AEDs, and in infants of nonepileptic mothers. The study only compared defects associated with the use of the older medication lamotrigine and the newer drugs as a group, and not differences in risk between specific drugs.
Ditte Mølgaard-Nielsen, MSc, and Anders Hviid, MD, MSc, both epidemiology researchers at the Statens Serum Institute in Copenhagen, Denmark, compared birth defects and AED use in a population of 837,795 infants born between 1996 and September 2008, using data from nationwide health registries. (See “Danish Study Data.”)
“Our study was designed to evaluate the risk of major birth defects following the use of a number of newer generation antiepileptics in early pregnancy, and we found no support for an increased risk of birth defects,” wrote co-author Dr. Hviid in an e-mail to Neurology Today.
In response to the FDA warning about topiramate, Dr. Hviid told Neurology Today that the group did not specifically look at topiramate and clefts, but the researchers plan to revisit the data to further investigate any specific data regarding topiramate.
“The topiramate and clefts signal is new, and as such our study was not designed to evaluate this association. We only evaluated specific groups of birth defects in the context of any newer generation antiepileptics and lamotrigine alone, where we had sufficient statistical power,” he explained.
“This is an interesting study, the main advantage of which is the use of a large population-based cohort,” commented Emilio Perucca, MD, PhD, professor and chair of medical pharmacology at the Institute of Neurology of the University of Pavia in Italy.
But there were also many study limitations, he told Neurology Today in an e-mail, including the retrospective nature of the analysis and very low sample sizes except for lamotrigine and oxcarbazepine.
Some of the variables assessed may not be accurate, he said, adding that malformation rates may be underestimated due to various factors, “such as the exclusion of cases diagnosed in the primary care setting,” he said. “Although I am not a statistician, I also have concerns about the efficiency of the adjustment procedures, particularly adjustments for concomitant old generation AEDs.”
These cannot be treated as a uniform confounder, he noted. “For example, co-medication with valproate is not the same as co-medication with carbamazepine.”
His greatest concern, he said, was the lack of a positive control group.
“Although the authors claim that their study indicates lower risk with newer AEDs than with older AEDs, none of the old AEDs were investigated in this population,” he wrote.
“To me, this is a major weakness,” he continued. “I would have liked to know how carbamazepine and valproate performed in the same population. Making inferences based on comparisons with data from other studies can be dangerous, particularly in an area where methodological differences in selection methods, ascertainment and classification can impact tremendously on results. While the data send a reassuring signal, I would like to base clinical decisions on evidence stronger than this.”
STUDY STRENGTHS, WEAKNESSES
Kimford J. Meador, MD, professor of neurology and director of the epilepsy center at Emory University in Atlanta, told Neurology Today that the study is nonetheless unique for being the first to use national registry data for the newer AEDs.
“Overall, the findings add to our knowledge base,” he said. “The sample size for lamotrigine is large, and is another strength, however the percentage of birth defects listed for lamotrigine is likely higher than it would be for monotherapy as both polytherapy and monotherapy appear to be included in their estimate.”
Another strength is that the study explored multiple confounding factors, he added. “However, there are also several limitations. The information for the other AEDs is encouraging, but the sample sizes are too small to be definitive, and it is not clear that the association of maternal epilepsy with birth defects is not confounded by AED dose, since the treatment of epilepsy usually employs higher dosages overall than migraine, pain and psychiatric indications,” he told Neurology Today.
“Also, similar to almost all prior studies, there are no blood-levels reported. Moreover, the marked individual variability in metabolic changes during pregnancy could obscure teratogenic dose effects.”
FDA WARNING ON AEDS
On March 4, the FDA issued a drug safety communication instructing physicians and other health care professionals to warn women of childbearing age that fetal exposure to the AED topiramate increases the risk of oral birth defects such as cleft lips and palates.
The announcement was based on a review of data from two major national registries: the North American Antiepileptic Drug Pregnancy Registry and the UK Epilepsy and Pregnancy Register. (See “FDA Warns That Topiramate Could Cause Birth Defects: What to Tell Patients,” in the Apr. 7 Neurology Today: http://bit.ly/jbaN4Q.)
The US data showed the prevalence of oral clefts was 1.4 percent in those whose mothers took topiramate, versus 0.38 to 0.55 percent among those exposed to other AEDs, and 0.7 percent in mothers without epilepsy and those not taking any antiepileptic.
The U.K. study found the prevalence was 3.2 percent in infants whose mothers were undergoing topiramate monotherapy in the first trimester compared to 0.2 percent among the general population.
Based on these risk data, the agency reclassified topiramate from a “Category C” to a “Category D” drug, meaning that data now confirm an elevated risk, but that using the drug when pregnant might still be acceptable under certain circumstances.
DANISH STUDY DATA
* A total of 1,532 infants had been exposed to lamotrigine, oxcarbazepine, topiramate, gabapentin, or levetiracetam during their first trimester. Among them, only 49 had a major defect, compared with 19,911 defects in children whose mothers had not taken the drugs, 3.2 percent versus 2.4 percent, respectively, for an adjusted prevalence odds risk of 0.99.
* Lamotrigine use in the first trimester was associated with a major birth defect in 38 of 1,019 infants. 2.8 percent of infants whose mothers had taken oxcarbazepine also had major birth defects.
* The birth defect rate was 4.6 percent in infants exposed to topiramate. Exposure to gabapentin and levetiracetam was uncommon, with only one infant in the gabapentin group and none in the levetiracetam group having a major defect.