Use of four of the most commonly prescribed seizure-control drugs at the beginning of pregnancy is associated with a dose-dependent increased risk of major birth defects. The findings, from 33 countries worldwide published Online First in The Lancet Neurology, are the first to provide a multivariable analysis of the risks associated with individual drugs and their doses, and will be crucial in helping doctors identify the safest effective treatment for women with epilepsy considering pregnancy.
Between 0•3% and 0•7% of all pregnancies are in women with epilepsy. Most of these women need to use antiepileptic drugs during pregnancy because uncontrolled seizures can harm the mother and fetus. However, previous research suggests that epilepsy drugs (particularly valproic acid) can increase the risk of birth defects. But until now, no adequately powered studies have compared the risks associated with different doses or assessed the influence of potential confounders (other factors that can affect outcomes) such as a family history of birth defects or severity of epilepsy.
"Present guidelines caution on the use of valproic acid during pregnancy, but offer little guidance on alternative options and how to manage women whose seizures cannot be controlled by other drugs", explains Torbjörn Tomson from the Karolinska Institute, Stockholm, Sweden and international colleagues.
To provide more evidence, the researchers investigated the association between the use of carbamazepine, lamotrigine, valproic acid (valproate), and phenobarbital at different doses and the risk of major birth defects detected up to the end of the first year after birth. The study used data from the International Registry of Antiepileptic Drugs and Pregnancy (EURAP) and included 3909 pregnancies.
A total of 230 pregnancies associated with major birth defects were identified by the end of the first year after birth. An increase in the rate of birth defects was noted with increasing dose for all drugs. The rate was lowest for low doses of the drugs lamotrigine (less than 300 mg per day) and carbamazepine (less than 400 mg per day).
The highest doses of valproic acid (1500 mg per day or greater) and phenobarbital (150 mg per day or day or greater) posed the highest risk to the fetus, with particularly high rates of birth defects recorded in pregnancies exposed to valproic acid 1500 mg per day or greater. The authors caution: "It should be emphasised, however, that, irrespective of which of the four investigated drugs was prescribed, the vast majority of women gave birth to perfectly healthy children."*
Additionally, the risk of defects was four times greater for offspring with a parental history of major congenital malformations.
The authors say: "Our results show that dose selection is as crucial as the choice of drug…[and] gives the prescriber the possibility of assessing how teratogenic [ability to cause birth defects] risks at that dose compare with the risk associated with alternative treatments at various doses."
In a Comment, W Allen Hauser from Columbia University, New York, USA says: "The findings are important to the clinician treating people with epilepsy because they provide specific information not only on the drug but also on the dose. It is easy to recommend against use of a specific drug (valproic acid, for instance) because of a higher risk of malformations, but if seizure control is not possible with alternative therapeutic regimens, such recommendations are difficult to implement. The data provide another reason for use of the lowest dose of a drug associated with optimum seizure control. Incidence of major congenital malformations associated with a low dose of a higher-risk drug might be lower than that associated with a high dose of a lower-risk drug."
Dr Emilio Perucca, University of Pavia, Pavia, Italy (until Saturday 4 June, noon CET) T) +39 338 610 3392 E) firstname.lastname@example.org
Or Dr Torbjörn Tomson, Karolinska Institute, Stockholm, Sweden (after Saturday 4 June, noon CET). T) +46 70 7907788 E) email@example.com
Professor W Allen Hauser, Columbia University, New York, USA. T) +1 212 305 2447 or +1 914 760 3144 (mobile) E) firstname.lastname@example.org
The Lancet Neurology