Abstracts

Rationale: Frontal lobe epilepsy (FLE) remains an area in need of further review to establish long term outcome following frontal lobe resection. The purpose is to assess the value of frontal lobe surgical resection and subsequent seizure recurrence rates in the pediatric population with intractable epilepsy.Methods: Retrospective epilepsy data review of 388 pediatric patients who underwent epilepsy surgery from 1992 to 2009 at Georgia Health Sciences University. A total of 56/388 patients (14.4%) with medically refractory FLE were selected for the study. Patients underwent preoperative prolonged video EEG recordings (56 patients) and intracranial EEG (53/56 patients) for ictal localization and semiology characterization. Resection sites for the study were divided into: precentral (3), premotor (14), prefrontal (18), frontal lobe (2), supplementary motor area (12), and lateral dorsal (16). Data for Magnetic Resonance Imaging (MRI) (3 patients not available for review), PET/SPECT (41 patients), and neuropathological examinations (47 patients) were included. Resection of the frontal lobe epileptogenic foci was determined as complete or incomplete. Seizure outcome was graded according to Engel s criteria at postoperative year 1, 3, and 5. Results: There were 38 males, 18 females, and the mean age of seizure onset was 4.8 years. The mean age at surgery was 11.2 years and seizure duration prior to surgery was a mean of 6.4 years. There were 49% with cryptogenic and 51% with symptomatic epilepsy. MRI abnormalities were identified in 21/53 patients (40%, three-not available to review). Pathological diagnosis (47/56) was classified into gliosis 53% (25), cortical dysplasia 38% (18), tumor 6% (3), and inflammation 2% (1). Engel class I (seizure freedom) was achieved in 54% (30/56) at post-operative year 1, 47% (22/47) at year 3, and 44% (17/39) at year 5. By the analysis of contingency tables, the following features were found to be significantly predictive of earlier seizure recurrence: preoperative EEG interictal contralateral spikes (p=0.02), desynchronous EEG pattern (p= 0.02), preoperative seizure clusters (p= 0.03), cryptogenic seizures (p= 0.01), presence of non-lesional MRI (p= 0.01), and younger age at onset of epilepsy (p= 0.02). Postoperatively, early recurrence was more likely if there were postoperative EEG spikes (p= 0.01). Conclusions: The study adds to our understanding of outcomes after frontal lobe resection for intractable pediatric FLE. More than half of children with intractable FLE achieved seizure freedom after epilepsy surgery. This is not sustained over time however. In the future, the predictive use of contralateral spikes, desynchronous EEG and multiple seizure types should be employed to assist with patient selection to sustain and/or improve seizure freedom rates.