Abstracts

Rationale: Surgical resection is considered as the best option for treating refractory focal epilepsy. Complete resection of the epileptogenic zone is known to be one of the determining factors for seizure outcomes. We hypothesized that completenss of resection varies according to the location of epileptogenic zone and that it may explain the lower success rate in patients with posterior quadrant onset.Methods: 25 patients (age range: 2-40; M 12 F 13) with refractory epilepsy who underwent intracranial EEG (iEEG) monitoring for further delineation of epileptogenic zone from Sep 2011 till April 2013 were reviewed in this study. This study was approved by the IRB at Florida Hospital for Children. After the implantation, a high resolution CT scan was carried out to localize the subdural electrodes. The surgery was performed by the neurosurgeon (JB) based on the resection plan reflecting epileptogenic zone identified in iEEG recordings and the results of other significant neuroimaging tests and electrical stimulation mapping. A post-operative MRI was acquired upon the completion of the resective surgery. By comparing pre-operative with post-resection pictures, and pre-operative with post-operative MRI co-registered to electrode CT images, a imaging analyst (PC) blinded to the resection plan determined the electrodes included in the actual resection margin. The results were compared with the resection plans by the epileptologists (KL and JS). Surgical outcomes (follow-up: 7-17 months) were obtained using Engel classification. We investigated the completeness rate of resection based on the lobar location and its relationship with surgical outcome. Results: The surgical outcome in the 25 patients was comparable to published series: Engel class I 41% (12/25), II 52% (15/25), III 7% (2/25) with relatively short follow-up duration (7-17 mo). The overall resection completeness rate was 71% (all electrodes resected/all electrodes planned). After excluding the electrodes overlapping with eloquent cortex found on intraoperative mapping, the completeness rate increased to 73%. Using this value as the threshold for considering the completeness of resection for each case (complete 73%; incomplete < 73%), the seizure freedom was related with completeness of resection (Fisher Exact p<0.01) as shown in Table 1. The resection location was a significant factor to resection completeness rate (ANOVA on Rank p < 0.01): temporal 88.8%, frontal 84.3%, parietal 66.4%, and parieto-occipital 40.2% respectively.Conclusions: Our study showed that completeness of resection is a significant prognostic factor for surgical outcome and that location of the epileptogenic zone affects the resection completeness rate. The low complete resection rate in the parieto-occipital lobe may be related to the limited accessibility to the region. Extensive use of MRI navigation with co-registered resection plan and post-resection intraoperative MRI may improve the outcome in these patients.