Abstracts

Rationale: Lateralized, material-specific memory problems are commonly observed in adults with temporal lobe epilepsy but this cognitive finding has been less commonly reported for younger patients. The inconsistent findings for the two age groups may be due to differences in the incidence of mesial temporal involvement. Furthermore, when increased memory deficits are reported in younger patients, it is often unclear whether the memory deficits are specific to temporal lobe seizures. We evaluated the hypothesis that temporal lobe seizures are more commonly associated with memory problems than seizures emanating from other sites. We also assessed the specific vulnerability that might be associated with mesial temporal structural abnormalities. Methods: Eighty-one patients were administered a neuropsychological battery that included tests of memory for semantically related verbal material (WRAML Story Memory and Story Memory Delayed),list learning of semantically unrelated words (WRAML Verbal Learning and Verbal Learning Delayed), learning of visual-spatial material (TOMAL Visual Learning and Visual Learning Delayed) and receptive vocabulary (PPVT-IV). The average age of the patients was 12.26 (3.32) years. Full Scale IQ for each patient was >69, with an average IQ of 90.59 (14.50). The patients were divided into three groups on the basis of seizure focus--temporal (n=45), frontal (n=22) and other (n=14). The temporal group was further divided into those who had a mesial structural abnormality (n=28) and those with either a lateral structural abnormality or normal MRI findings (n=17). Multivariate Analyses of Variance (MANOVA) were computed for each type of memory test to test group differences.Results: Overall F values for MANOVAs computed for WRAML Verbal Learning and TOMAL Visual Learning were nonsignificant. The overall F for the MANOVA with WRAML Story Memory was significant, p<.001. Neither the main effect for hemisphere nor the interaction of hemisphere and lobe was significant. The effect for lobe was significant, p<.01. Post-hoc analyses revealed that the temporal and frontal focus groups had significantly lower scores for both Story Memory and Delayed Story Memory than did the other seizure focus group but the temporal and frontal groups were not different from each other. Because the frontal and temporal groups had lower scores than the other seizure focus group on the PPVT-IV, the analyses were repeated with the PPVT-IV as a covariate but the results were unchanged. Analyses of memory test performance for the two temporal lobe groups did not reveal any differences for the Verbal Learning and Visual Learning subtests. The mesial group performed more poorly on Delayed Story Memory, p<.03, but not on immediate recall.Conclusions: Memory problems were as common with frontal lobe seizures as they were with temporal lobe seizures. When memory problems did occur with temporal lobe seizures, they tended to be associated with mesial temporal abnormalities. The findings from this study do not provide any evidence for lateralized, material-specific memory deficits in pediatric epilepsy.