Post Resection Electro-corticography Predicts Seizure Outcome - A Multivariate Logistic Analysis (P6.339)

OBJECTIVE: To evaluate post-resection ECoG spikes with respect to location of the epilepsy, presence of an obvious lesion, and duration of the epilepsy and determine the predictive assessment after epilepsy surgery. BACKGROUND: Electrocorticogophy(ECoG) elucidates the characteristics and the extent of the epileptogenic zone or eloquent cortex. Utility of post resection ECoG recordings in predicting outcome or guiding additional resection in epilepsy surgery is unclear. METHODS: From a prospectively compiled database of 195 patients undergoing epilepsy surgery by a single surgeon at the University of Texas Comprehensive Epilepsy Program, six-month postoperative data was compiled for patients 18 years and older. Independent variables included post-resection ECoG (0=no discharges, 1=subtle or rare discharges 2=significant discharges), age of onset of epilepsy, duration of epilepsy, resection type (1=mesial temporal, 2=temporal neocortical and 3=extra-temporal) and pathology type (1=cavernoma/tumor, 2=mesial temporal sclerosis (MTS)/focal cortical dysplasia(FCD), 3=gliosis/heterotopic neurons/encephalomalacia). An ordered logistic regression analysis was used to predict seizure outcome as measured by the ILAE(International League Against Epilepsy) scale was performed. RESULTS: Of the 75 patients included in the analysis, there were 52 (69.3%) mesial temporal lobe surgeries, 7 (9.3%) temporal neocortical only surgeries and 16 (21.3%) extra temporal neocortical cases. Pathology identified 15 cases of cavernoma/tumor, 40 cases of MTS/ FCD, and 20 cases of other non-specific pathologies such as gliosis/encephalomalacia. Logistic regression model incorporating these variables revealed that nonspecific pathologies categorized in group 3 (gliosis/ heterotopias/ encephalomalacias) > MTS/FCD > defined lesions categorized in group 1 (such as tumor/cavernoma) predicted a higher i.e. worse ILAE outcome score (OR=1.65(1.29; 5.46). Post resection spikes significant > rare >none predicted higher ILAE outcome score (OR=1.91(1.08; 3.38). CONCLUSIONS: Pathology predicts outcome after epilepsy surgery. Nonspecific pathologies such as gliosis and post-resection spike discharges predicted worse outcome after epilepsy surgery. The location of surgery did not predict outcomes after epilepsy surgery.