Abstracts

Rationale: Retrospective studies have highlighted an association between temporal lobe epilepsy and a childhood history of prolonged febrile seizures (PFS). Moreover, imaging studies that have investigated this issue prospectively have found evidence of hippocampal injury following PFS in the form of oedema and increased asymmetry. Given the well established role of the hippocampus in memory, we were interested in investigating whether recognition memory processes were disrupted in a group of children following a PFS. To that end we used the visual paired comparison (VPC) task, which has been commonly used to investigate recognition memory in children. Previous animal work with this task has shown that the tendency to look longer at novel items following a delay relies on the hippocampus. Given the association of hippocampal abnormalities with PFS we predicted that children in the PFS group would show impaired novelty preferences relative to healthy controls. Methods: Twenty-three children underwent neuropsychological and MRI investigations a mean of 48 days after their PFS episode. The Bayley Scales of Infant Development provided measures of cognition. Intracranial volumes were calculated on the T1 weighted image by using an automatic skull-stripping technique. Two independent researchers traced hippocampal volumes to ensure measurement reliability. For the VPC task, each child was familiarized to one face in five-10 second trials. Following a 5-minute delay, the child saw two memory trials of the familiar face alongside a novel face with the position of the novel face counterbalanced. Looking times were scored off-line from a digital video. Novelty preference was computed as the mean proportion looking to the novel face across the memory trials. Performance was compared to that of 13 normally developing children of similar age. Results: A univariate analysis of variance test controlling for cognition found that the two groups were different in their novelty preferences (p=0.01). Namely, the PFS group spent less time looking at the novel face (0.48) than the control group (0.59) across both memory trials. An independent samples t-test revealed that the PFS group performed worse than controls on cognitive measures (p=0.05), yet there was no effect of cognition on their novelty preferences (p= 0. 57). A partial correlation within the PFS group controlling for age found that mean hippocampal volumes were positively correlated with novelty preferences (r=0.46, p=0.04). Conclusions: In this study we found differences between a group of healthy controls and children following a PFS on a hippocampally dependent task. Namely, patients following a PFS spent less time looking at a novel face following a delay when compared to their normal counterparts. The absence of any relationship between cognitive level and performance on this task despite the group differences in cognition attests to the specificity of the memory impairment exposed by the VPC task. Finally, the relationship between mean hippocampal volumes in the PFS group and their degree of novelty preference corroborates the important role of the hippocampus in these processes.