Yield and Predictors of Epilepsy Surgery Candidacy in Children Admitted for Surgical Evaluation

Abstract Objective The aim of this study was to identify preadmission variables that are prognostic of epilepsy surgical candidacy for children admitted into an epilepsy-monitoring unit. Methods This study is a retrospective review of patients 0-18 years who were admitted into the pediatric epilepsy-monitoring unit at Mayo Clinic between November 2010 and December 2013 to assess for surgery candidacy for medically intractable epilepsy. Demographic data, epilepsy details, treatments, electroencephalograph and imaging results, and the consensus notes of the epilepsy surgery conferences were collected. Results One hundred and forty eight children underwent inpatient, prolonged video-electroencephalograph monitoring for presurgical evaluation, of which 136 had their typical events recorded. Five recommended for callosotomy alone were excluded and 131 composed the study group. Of these 131, the epilepsy surgery conference consensus deemed that 69 were surgical candidates, of which 45 (65%) underwent resective surgery. Chi-square analysis found seven preadmission predictors of surgical candidacy: single semiology at seizure onset (P < 0.001), structural etiology (P < 0.001), one or more interictal foci all in the same hemisphere (P < 0.004), focal background electroencephalograph slowing (P < 0.001), focal or hemispheric abnormality on magnetic resonance imaging (P < 0.001), male sex (P = 0.02), and normal development (P = 0.04). Conclusions The presence of fewer than four predictors was suggestive of low likelihood of candidacy (<31%), whereas if more than four factors were present, 91% were found to be surgical candidates. These findings facilitate clinical decision-making for providers in a cost-effective manner and provide realistic expectations for families.