Abstracts

Rationale: The determination of whether a patient is a candidate for epilepsy surgery is highly individualized; however, there still is ambiguity regarding what factors constitute a surgical candidate and the chances of an epilepsy free surgical outcome. Thus, the aim of this study was to review the pre-admission variables that are potentially prognostic of surgical candidacy for children admitted into an epilepsy-monitoring unit (EMU). In addition, we assessed the likelihood that a child is a surgical candidate based on the accumulation of different pre-admission factors with the goal of aiding clinical decision making for both physicians and families. Methods: This study is a retrospective review that looked at patients between the ages of 0-18 years that were admitted into the pediatric EMU at Mayo Clinic between November 2010 and December 2013 to assess for possible surgery candidacy for medically intractable epilepsy. Demographic data, epilepsy details (seizure type, electroclinical syndrome, frequency, etiology, and neonatal seizures), cognitive status, prior and current treatments, outpatient EEG results, neuroimaging results, video-EEG (VEEG) monitoring, and the consensus notes of the Epilepsy Surgical Conferences were collected. Results: 148 children underwent VEEG for surgical workup and 136 of them had their typical events recorded. From the 136 children, the Epilepsy Surgery Conference consensus deemed that 74 were surgical candidates (54.4%): 4 corpus callosotomy alone, 1 resection plus callosotomy, 27 single resection without intracranial monitoring, and 42 with intracranial monitoring followed by resection. Out of the 62 patients that were denied as surgical candidates, the most common reason (47%) was multifocal ictal onset. From the 69 resective surgical candidates, 23 had no record of undergoing surgery, 4 were palliative, and 42 underwent surgery with the goal of seizure freedom. Of these 42, 94% (16/17) of the cases with resection without intracranial monitoring and 76% (19/25) of those with intracranial monitoring followed by resection were seizure free postoperatively. Finally, Chi-square analysis found five statistically significant predictors of surgical candidacy: single semiology at seizure onset, structural etiology, either single interictal focus or multiple foci but limited to one hemisphere, focal background EEG slowing, and focal/hemispheric abnormality on MRI. Conclusions: The likelihood of surgical candidacy was low if two or fewer predictors (out of the five) were present (16/64, 25%), moderate if three predictors were present (17/27, 63%), and very high if four or more predictors were present (37/41, 90%). These findings can be very useful in clinical practice to give physicians and families insight into the likelihood that a child will be a surgical candidate prior to admission into the EMU.