Abstracts

Rationale: Epilepsy is refractory to medications in more than 20% of children. Some of these are candidates for focal surgical resection of the seizure onset zone with the goal of seizure-freedom. Intraoperative electrocorticography (ECoG) is a technique for monitoring cortical activity during the surgery with electrodes placed directly on the brain. It can be done pre- and post-resection (post-ECoG) and can evaluate resection margins for ongoing epileptiform discharges. Some studies have shown post-ECoG epileptiform discharges are predictive of seizure recurrence in adults, especially those undergoing temporal lobectomy, frontal lesionectomy, or low-grade glioma resection. Other studies have not shown post-ECoG discharges to be predictive of seizure recurrence. We hypothesized that discharges on post-ECoG predict seizure outcomes in children.Methods: Retrospective chart review of all patients who had surgery with post-ECoG 10/1/03-10/1/09 and had >1 year of follow-up. ECoG was recorded from 12 electrodes around the resection cavity at the direction of the surgeon.Results: Inclusion criteria were met in 52 patients. We found that 73% of the 37 patients with post-ECoG discharges were seizure-free, while 60% of the 15 patients with no post-ECoG discharges were seizure-free. The odds ratio of seizure freedom among those that had residual epileptiform discharges versus odds of seizure freedom among those that did not = (27/10) / (9/6) = 1.80 [95% Confidence interval for OR = (0.509, 6.361)], but was not statistically significant (?2, p = 0.36). Patients with gangliogliomas had better outcomes than other pathologies (Fisher s Exact test, p= 0.0102), despite 80% of them having post-ECoG spikes. There was a non-significant trend towards greater seizure freedom in lesional patients 74%, -vs- 33% in non-lesional patients (Fisher s Exact test, p= 0.06). Location was not associated with presence of post-resection spikes on ECoG (one-way ANOVA, p = 0.282), nor with seizure freedom (one-way ANOVA, p = 0.123).Conclusions: Absence of epileptiform discharges on post-ECoG does not appear to predict seizure freedom in all pediatric patients referred for epilepsy surgery, although our sample size was too small to detect a subtle difference. Interestingly, the trend of better seizure freedom was in the opposite direction of the hypothesis, as patients with post-ECoG discharges did slightly better than those without, although this was not statistically significant. Our outcomes in patients with gangliogliomas were better than those with other pathologies, irrespective of the frequent presence of post-ECoG spikes. There is a paucity of data about post-ECoG and thus disagreement about its use. Our data indicate that it is safe to leave epileptiform discharges unresected in many pediatric patients. More research is needed to confirm this and inform ongoing debates about the use of ECoG.