Abstracts

RATIONALE: Unilateral anterior temporal lobectomy (ATL) for the treatment of medically refractory partial epilepsy is known to be associated with risks of post-operative cognitive impairment, particularly affecting memory. Recent retrospective studies, identifying hippocampal sclerosis (HS) in the excised tissue, suggest that the presence or absence of HS is the primary determinant of post-operative memory disability. Volumetric MRI provides an opportunity to identify HS prior to surgery with a high degree of accuracy. If used as the primary criterion for surgery, volumetric MRI may lead to a reduction in cognitive disability after surgery.
METHODS: We report results of neuropsychological investigation in a consecutive series of patients undergoing video-telemetry (n=167) and, in a subsample undergoing unilateral ATL guided by volumetric MRI (n=88) at St. Vincent[ssquote]s Hospital, Melbourne. Using a previously validated measurement model comprising five core cognitive abilities, we tested the hypothesis that memory function prior to surgery could predict side of HS identified by MRI and confirmed by histology after surgery. We also examined cognitive outcome in the operated group, contrasting pre- with post-operative memory and other ability scores with retest scores from the community control sample in a series of 2 x 2 ANOVA[ssquote]s.
RESULTS: Logistic regression modelling failed to show any significant relationship between preoperative memory scores and side of HS. In addition to being nonsignificant, all statistical effects were small, for example, the double dissociation of memory function prior to surgery and side of HS was associated with an eta-squared of .01. Comparison of cognitive abilities before and after surgery suggested no change or modest improvement in material specific memory function and more significant improvement in fluid intelligence when the left and right ATL groups were examined separately.
CONCLUSIONS: Overall the data suggest that memory assessment prior to surgery is not a good predictor of side of pathologically confirmed HS, and that the risk of cognitive disability is low after unilateral ATL guided by volumetric MRI. Finally results are discussed in terms of implications for use of the intra-carotid amytal (Wada) procedure. Calculation of a range of positive predictive power values based on optimistic estimates of sensitivity and specificity suggest that, even with high sensitivity and specificity, the likely range of positive predictive power does not justify the use of the Wada in patients selected for surgery using volumetric MRI.
Support: Australian Research Council