Update on Reproductive Safety of Psychotropic Medications Part II of an Interview with

Cohen: Early reports suggested that benzodiazepines, specifically diazepam, were associated with higher risk of cleft lip and/or cleft palate in children exposed during the first trimester, but several subsequent reports did not replicate that finding. Over the next decade or so, a number of investigators studied effects of individual benzodiazepines and benzodiazepines as a class, but the methodologic problems in their studies were considerable: their patients came from very different settings, had been exposed to different benzodiazepines at different dosages, and often had taken other psychiatric medications or illicit substances. It is therefore difficult to decide which studies are most appropriate for inclusion in a meta-analysis of benzodiazepine exposure. Lori Altshuler and I published a paper in 1996, which included a meta-analysis of benzodiazepines as a class (including diazepam, alprazolam, clonazepam, and other benzodiazepines). We looked specifically at risk for cleft lip and/or cleft palate, because there has been little concern about other malformations associated with benzodiazepines. When we included studies that had great variability with respect to methodologic rigor, we noted a small, but nonetheless increased, risk for oral clefts associated with first-trimester benzodiazepine exposure on the order of 0.6 percent. That is a ten-fold increased risk relative to the general population, where the risk of oral clefts is six in ten thousand. An even larger meta-analysis was published in the British Medical Journal several years ago which examined the risk of malformations, and specifically oral clefts, across all benzodiazepines--and again, the results were mixed. In that study, the risk of oral clefts was increased, but not to the degree that we found in 1996. The increase in relative risk was more modest and differences were more discernible in the case-control than in the cohort studies. In fact, if one excluded from that meta-analysis the most profoundly flawed study (Laegrid et al), the risk estimate would be even smaller for clefts. I think we can summarize what we know about teratogenic risk of benzodiazepines this way: if there is a risk for clefts, it is probably very small. Accordingly, we do not consider benzodiazepines contraindicated during pregnancy, because, as with depression, the literature suggests an association between untreated anxiety during pregnancy and higher rates of perinatal complications--children with lower APGAR scores and higher rates of pre-term labor, for example. We even noted (and published several years ago) a case of placental abruption in a woman whose panic attacks at term were associated with lability of blood pressure. A short time before that, she had quickly tapered and discontinued treatment with anti-panic therapy, after which she had suffered severe recurrent symptoms. So again, consideration of risks and benefits suggests that benzodiazepines need not be uniformly avoided during pregnancy.