The risk of having an autism spectrum disorder and childhood autism was significantly increased among children exposed prenatally to valproate, which persisted after researchers adjusted for parental psychiatric disease and maternal epilepsy in a Danish study that evaluated children up to age 14 years.
These risks were higher than the risks among the children whose mothers had stopped taking valproate before conceiving and also was elevated among the children of women who did not have epilepsy and took valproate, "indicating an association with valproate exposure," and that valproate may have a biological effect, said Jakob Christensen, Ph.D., of the department of neurology at Aarhus (Denmark) University Hospital, and his associates.
Dr. Jakob Christensen
"For women of childbearing potential who use antiepileptic medications, these findings must be balanced against the treatment benefits for women who require valproate for epilepsy control," they concluded (JAMA 2013;309:1696-703).
The population-based study used a birth registry of all children born alive in Denmark between January 1996 through Dec. 31, 2006, for a total of 655,107 children who had not been exposed to valproate and 508 exposed to valproate prenatally. At the end of follow-up, when the children were a mean age of almost 9 years (range, 4-14 years), 5,437 children had been diagnosed with autism spectrum disorder, including 2,067 with childhood autism. A national prescription registry was used to document exposure to antiepilepsy drugs (exposure during pregnancy was considered 30 days before conception to the day of birth) and a registry of psychiatric diagnoses was used to identify which children were diagnosed with autism.
Among the 508 children exposed to valproate, the incidence of autism spectrum disorder was 4.42% (hazard ratio, 2.9), and the incidence of childhood autism was 2.5% (HR, 5.2). The hazard ratios were adjusted for known autism risk factors, including age of the parents at conception, parental psychiatric history, and sex of the child.
Among the 432 children whose mothers had taken valproate for epilepsy (not another diagnosis treated with valproate), the incidence of autism spectrum disorder was 4.15% (HR, 1.7), and the incidence of childhood autism was 2.95% (HR, 2.9). Among the 6,152 children not exposed to valproate, the incidence of autism spectrum disorder was 2.44% and the incidence of childhood autism was 1.02%. The risks for both autism spectrum disorder and childhood autism were increased among women who took valproate during pregnancy, compared with the children whose mothers had taken valproate but had stopped taking it at least 30 days before they were conceived
The strengths of the study included the unlikely effect of selection bias because the investigators were able to follow-up most of the children born in the country over 14 years, and the high quality of autism diagnoses in the psychiatric registry. The possibility that some pregnancies exposed to valproate may have been missed was among the weaknesses of the study because the prescription registry only includes outpatient prescriptions.
The study was funded by grants from the European Research Council and the Danish Medical Research Council. Dr. Christensen received research support from the Danish Epilepsy Association and honoraria for serving on scientific advisory boards and lecture honoraria from UCB Nordic and Eisai AB, and received travel funding from UCB Nordic.
Dr. Meador received research support from the National Institutes of Health, the Patient-Centered Outcomes Research Institute, Pfizer, and UCB Pharma; Emory University received salary support from the Epilepsy Consortium, which gets funding from NeuroPace, Novartis, Upsher-Smith Laboratories, UCB Pharma, and Vivus Pharmaceuticals; and travel support from Sanofi Aventis. Dr. Loring received consulting fees from NeuroPace; grant support from NIH, the Patient-Centered Outcomes Research Institute, Pfizer, and UCB Pharma; and book royalties from Oxford University Press.