Short-Term Oral Diazepam Treatment during Pregnancy : A Population-Based Teratological Case-Control Study

To study the possible human teratogenicity of short-term (generally 3 weeks) oral diazepam treatment during pregnancy.

A matched case-population control pair analysis based on the total (maternal self-reported and medically recorded) or medically recorded diazepam treatments, in addition to a comparison between cases and patient controls in the population-based large data set of the Hungarian Case-Control Surveillance of Congenital Abnormalities from 1980 to 1996.

38 151 population-control neonates without any congenital abnormalities, 22 865 neonates or fetuses with congenital abnormalities (cases) and 812 neonates or fetuses with Down's syndrome (patient controls).

24 congenital abnormality groups.

2746 (12.0%) cases, 4130 (10.8%) population controls and 97 (11.9%) patient controls were born to mothers treated with diazepam during pregnancy. The matched case-population control pair analysis showed a higher rate of limb deficiencies, rectal-anal atresia/stenosis, cardiovascular malformations and multiple congenital abnormalities after diazepam use during the second and third months of gestation, i.e. in the critical period for most major congenital abnormalities, based on maternal self-reported and medically recorded information. However, the evaluation of only medically recorded diazepam use did not indicate a higher use of diazepam in any congenital abnormality group. Thus, the higher occurrence of diazepam treatment among cases in the primary analysis may be due to the lower proportion of maternal self-reported diazepam intake in the population control group, i.e. recall bias. The comparison of diazepam use between 24 congenital abnormality groups and patient controls as the referent group showed a difference only in the group of intestinal atresia/stenosis, probably due to chance error caused by multiple comparison.

Short-term diazepam treatment in usual therapeutic doses during pregnancy did not present any detectable teratogenic risk to the fetus.