The effect of fetal valproate exposure on memory function and hippocampal structure in school-aged children

Women with epilepsy are usually advised to continue taking anti-epileptic drugs (AEDs) throughout pregnancy due to the dangers that uncontrolled seizures can pose for both mother and fetus. Compared to other AEDs, sodium valproate (VPA) has been associated with greater risk of birth defects and cognitive impairment. Previous research suggests that VPA impacts on the development of intellectual abilities, however the long-term effect of fetal VPA exposure on other cognitive skills has been poorly characterised. Further, it is unknown whether abnormalities in brain structure underlie these cognitive deficits. This research aimed to investigate memory functioning and hippocampal structure in school-aged children who were exposed to VPA in the intrauterine environment. One hundred and five prospectively recruited school-aged children (aged six to eight years) exposed to AEDs in utero participated in the study. The sample included 26 children exposed to VPA monotherapy and 15 exposed to VPA polytherapy. All children underwent a neuropsychological assessment that included measures of memory. MRI scans were also collected from 14 children exposed to VPA monotherapy. Results from neuropsychological measures showed that verbal memory skills were negatively affected by exposure to VPA, in a dose-dependent manner. Conversely, non-verbal memory abilities were developmentally appropriate. Language impairment appeared to underlie the reduced performance on some verbal memory measures, except for on a measure of retroactive interference. Analyses of the imaging data did not reveal differences between left and right hippocampal volumes (after controlling for head size) in children exposed to VPA monotherapy. VPA dose was not correlated with hippocampal volume; however there was a significant relationship between hippocampal volume and associative learning ability. These data provide evidence of memory impairment in school-age children exposed to VPA, adding additional weight to the argument that VPA places children at risk of poorer developmental outcomes. The results suggest that these deficits in memory functioning may be partly mediated by abnormal hippocampal development; however studies with larger sample sizes and the inclusion of controls are warranted. The findings suggest that the prescription of VPA to women of childbearing age should be carefully considered in terms of risks and benefits. Women taking VPA should be supported throughout pregnancy and their children monitored from an early age. Future research is required to improve detection of children at risk and develop effective intervention strategies.