Prenatal exposure to valproate and the risk of autism spectrum disorder: A systematic review

Abstract

ABSTRACT Objective: To determine whether children prenatally exposed to valproate (VPA) have increased risk of autism spectrum disorders (ASD). Methods: The literature search was conducted in PubMed, Web of Science, Scopus, and PsycINFO with the following keywords combined with the Boolean operators 'AND' and 'OR': "valproate in pregnancy", "prenatal exposure to valproate" and "offspring autism", "autism spectrum disorder" or "autism". Inclusion criteria were research studies reporting results on the use of VPA in pregnant women and its relationship with ASD, resulting in a total of 260 articles from which 9 were included in this review. Results: Of the nine articles included in this review, eight were population-based cohort studies and one was a population-based case-control study. Prenatal exposure to VPA was associated with ASD development in six studies, whereas three reported no association between VPA exposure and ASD, possibly due to sample size and ASD assessment method. Among five studies examining VPA dose effects, two reported a dose-response relationship with ASD, whereas one reported that the risk of ASD was similarly increased with both high and low VPA doses, and another two showed no association, possibly due to a limited range of doses in the samples. No difference was found between exposure to VPA monotherapy and polytherapy on ASD in one study, whereas another three reported differences, but in opposite directions. Two studies were concordant in that there was no difference in the risk of ASD between offspring of women who took VPA early and later in pregnancy. Results by sex (three studies) were inconsistent, with one study concluding that ASD was more frequent in VPA-exposed boys than girls, one showing no evidence of sex differences, and the other showing that ASD is more frequent in females than in males VPA-exposed children. Significance: The results suggests that the use of VPA during any pregnancy trimester, alone or in combination with other antiepileptic drug, can be associated with a significantly greater risk of ASD in the offspring, comparing with unexposed children.