Abstracts

The purpose of this study was to determine whether children and adolescents who undergo temporal lobe excisions for the treatment of intractable epilepsy are at risk for a decline in memory after surgery. Although this topic has had extensive consideration in the adult epilepsy surgery literature, much less is known in children; the existing findings are variable and are based on relatively small samples. We examined performance before and one year after surgery and evaluated a number of seizure-related and cognitive variables to determine whether it is possible to predict memory outcome after temporal lobectomy at either a group or individual level. This study was based on a retrospective chart review of patients who had undergone excision from the temporal lobe at the Hospital for Sick Children in Toronto. We identified 62 cases (5.5 to 19 years of age) who had intelligence and memory testing both before and after surgery. The memory tests included recall of a story, recall of a word list (learned over five trials), face recognition, and recall of a geometric design. The variables examined included: laterality of lesion, pathology (tumour, mesial temporal sclerosis, other), age at seizure onset, age at surgery, seizure outcome, pre-operative IQ, and pre-operative memory performance. Group performance was analysed in ANOVAs. Individual change was examined by calculating difference scores and identifying cases who had declined or improved after surgery. Age and cognitive variables were examined with correlations. Examination of laterality showed that decline was found after surgery on only one task, that of word list recall, and only for those with excisions from the left temporal lobe. There were main effects of pathology, in that those with tumours performed better on measures of verbal memory both before and after surgery. Those who were seizure-free after surgery did not differ in memory performance than those who continued to have seizures. Improvement after surgery was found in 4 to 30% of children, and decline in 0 to 47%, depending on the task. The percentage of children who showed improvement or decline did not differ by laterality or pathology. Older age of seizure onset was associated with greater improvement after surgery on word and design recall. Those with lower pre-operative memory scores were more likely to show improvement, whereas higher IQ was associated with positive change on design recall. Although change in memory can be documented after temporal-lobe excisions in children and adolescents, it is difficult to predict decline or improvement. Variables associated with the seizures, surgery, and underlying pathology showed limited capacity in prediction of change in memory after surgery, with only age at seizure onset demonstrating a relation to outcome. Pre-operative cognitive variable showed modest relations with change, but these relations varied across tasks.