Dose selection is as critical as is the choice of antiepileptic drug for avoiding birth defects in women with epilepsy, according to a large observational cohort study published online June 6 in The Lancet Neurology.
In a comparison of the four most commonly prescribed antiepileptic drugs, lamotrigine at a dose of less than 300 mg/day at the time of conception was the least likely to be associated with birth defects. Valproic acid at 1,500 mg/day or more was 16 times more likely to be associated with birth defects, according to the results of the multivariate analysis. Carbamazepine and phenobarbital carried intermediate levels of risk, depending on the dose.
The results came from an observational cohort study of 3,909 pregnancies representing 3,521 women in 42 countries between 1999 and 2010. All the women had epilepsy and were taking one of the four drugs at conception.
The investigators, led by Dr. Torbjörn Tomson of the Karolinska Institute, Stockholm, excluded pregnancies in women whose antiepilepsy prescription was changed during the first trimester, those who were exposed to antiepileptic polytherapy or to other potentially teratogenic medications, and women with comorbidities such as diabetes, toxoplasmosis, and HIV that are known to increase the risk of congenital malformations (Lancet Neurology 2011 June 6 [doi:10.1016/S1474-4422(11)70107-7]).
In their multivariate analysis, the investigators controlled for a host of potential confounders. These included maternal age, sex of child, parental history of major congenital malformations, geographical region, parity, type of epilepsy, generalized tonic-clonic seizures during the first trimester, parental education, and the use of folic acid.
The overall rate of birth defects among women taking one of the four drugs was 6%. The lowest doses of lamotrigine carried the lowest absolute risk of birth defects 17 of 836 pregnancies (2%). The highest doses of valproic acid carried the highest absolute risk; 24 of 99 pregnancies (24%) resulted in major congenital malformations.
The investigators compared all drugs and doses to lamotrigine at less than 300 mg/day. Lamotrigine at more than 300 mg/day carried 2.2 times the risk of birth defects.
The lowest dose of carbamazepine (less than 400 mg/day) was not statistically more risky than was low-dose lamotrigine. An intermediate dose of carbamazepine was associated with a 2.5-fold increase in risk, and carbamazepine at 1,000 mg/day or more was associated with a 4.6-fold increase in risk.
Phenobarbital at less than 150 mg/day was 2.5 times as risky as was low-dose lamotrigine, and phenobarbital at 150 mg/day or greater was 8.2 times as risky.
Valproic acid at less than 700 mg/day was 2.8 times as risky as was low-dose lamotrigine, intermediate doses of valproic acid were 5.8 times as risky, and doses of 1,500 mg/day or more were 16.1 times as risky.
Except for the lowest dose of carbamazepine, all increases in risk associated with the drugs were statistically significant when compared to low-dose lamotrigine.
The only nondrug covariates that proved to carry a significant increase in risk were folate supplementation and a parental history of major congenital malformations. Inappropriate folicate use increased the risk of birth defects by 40%, compared with appropriate use. A parental history of birth defects carried a 4.4-fold increase in risk compared to pregnancies without that characteristic.
"Our findings show that the risk of major congenital malformations increases dose-dependently with all assessed antiepileptic drugs," the investigators wrote. "The approach generally accepted by physicians so far has been to identify, before conception, the lowest effective dose of the drug that is most appropriate for the woman’s epilepsy. Our study gives the prescriber the possibility of assessing how teratogenic risks at that dose compare with the risks associated with alternative treatments at various doses. Our data suggest that many women can enter pregnancy at low doses and maintain seizure control, although in many such cases dose adjustment might be needed later in pregnancy."
The study was funded by Eisai, GlaxoSmithKline, Janssen-Cilag, Novartis, Pfizer, Sanofi-Aventis, UCB, Netherlands Epilepsy Foundation, Stockholm County Council, and ALF. All but one of the eight investigators reported relationships with numerous pharmaceutical companies. The authors stated that the sponsors had no role in study design, data collection, data analysis, data interpretation, or writing the report.