Abstracts

Rationale: Selected patients with continued medically refractory seizures despite an initial surgery may be considered for further surgery. Prior to the reoperation, additional tests such as imaging studies, video EEG, and repeat MRI are ordered. We retrospectively reviewed children who have undergone reoperation and examined the pattern of ordering additional test, its utility and outcomes. Methods: We identified patients from our epilepsy surgery database who had cortical resection for medically intractable focal epilepsy with a minimum of 6 months follow up from the last surgical resection. Patients who had palliative procedures, corpus callosotomy, VNS, or and an initial hemispherectomy were excluded. The medical records were queried for demographics, video EEG and MRI findings, tests done, seizure focus and etiology, number and type of surgical procedures that resulted in cortical resection, and outcome (Engle). Results: Twenty two of 153 patients (13%) had two or more surgical resections. Five of the 22 had more than 3 or more. Etiologies included: cortical dysplasia (8), tumor (6), encephalitis/Rasmussens (3), mesial temporal sclerosis (1) and nonlesional (4). Fourteen had an initial extratemporal and 8 were temporal surgeries. Eleven patients out of 17 (64%) with a repeat resection had a Class I Engle surgical outcome, while 3 (17%) had Class II. Only 2 (11%) had Class IV outcome. Of the five who had 3 resections, 3 (60%) had Class I outcome and 2 were Class IV (1 became seizure free when a new antiepileptic drug was started). Of those who had Class IV outcomes, 2 were nonlesional case, 1 had dysplasia and 1 had dual pathology on the contralateral hemisphere with mixed seizures and multifocal spikes. All had an initial extratemporal resection. For those with a 1st reoperation, all patients had a video EEG and MRI of the brain done. In two of the patients, additional test specifically MEG were done other than video EEG and MRI. For 20 out of the 22 patients, the video EEG findings were concordant to the previous video EEG or the seizures were coming from the area where the previous resection site was. In one patient, a MEG study was done but the results were concordant with the video EEG. One patient had a video EEG done with no clear focus.The MEG study results however was focal and guided in the placement of the grids and resection after, showed an Engel Class IA outcome. Conclusions: Selected patients with continued seizures may be candidates for further surgery. Good outcome were seen in cases were extension of the previous resection site was done and a lesionectomy was completed. An initial extratemporal resection, dual pathology and nonlesional cases were predictors of poor outcome. The study shows that at a minimum, concordant imaging and video EEG findings arising from the previous resection site are adequate to be considered for reoperation and continue to be predictors of good outcome. This is the same principle considered prior to an initial resection. In a few selected cases however, additional tests are needed.