Abstracts

Rationale: Seizure onset in insular cortex is underrepresented in epilepsy surgery literature, possibly due to deep anatomic location and rich vascular network. Seizure semiology in insular epilepsy can mimic frontal, temporal, or parietal lobe semiology, resulting in incomplete resections and poor seizure outcome. Three case series have described surgery in adult and pediatric epilepsy cases with insular lesions on MRI. We report an additional six pediatric cases that emphasize the role of surgery in intractable insular epilepsy, including MRI-negative cases.Methods: Of 97 epilepsy surgery patients at Miami Children s Hospital between May 2009 and May 2011, six pediatric patients (6.2%; M/F:1/5; age range 5-13 years at time of surgery) underwent insular resection. All patients had intractable focal epilepsy with presurgical evaluation (long-term video EEG, MRI, and PET/SPECT). Five patients had chronic recording with subdural electrodes over cortical convexity and depth electrodes into insula; intraoperative ECoG was performed in one patient.Results: All preoperative EEGs revealed widespread electrographic field across opercular region [frontotemporal (4), parietotemporal (1), or frontoparietotemporal (1)]. Prior to insular resection, one patient had undergone orbitofrontal/anterior temporal resection with poor outcome. Five patients were MRI-negative for insula (one demonstrated a region of frontal FCD), and one evidenced an insular tuber. Subsequent review after full evaluation revealed insular FCD in one patient initially thought to be MRI-negative. PET (1) or SPECT (3) evidenced insular abnormalities. Five patients evidenced frequent spikes arising from the insular depth electrodes interictally. Ictally, EEG revealed seizure onset in the electrodes covering the frontal (1), temporal (2), and parietal (1) regions. One patient monitored only with ECoG evidenced spikes from electrodes covering insula and frontal lobe but no seizures were recorded. In addition to insular resection, all patients had cortical resections involving frontal (6) or temporal (2) regions. Resection was complicated by ischemic stroke in the immediate postoperative period in one patient, resulting in transient hemiplegia. None of the patients experienced permanent deficits. Four patients were Engel class I, one Engel class III, and one Engel class IV at last follow-up (mean 8 months). Pathology revealed cortical dysplasia (3) and cortical tuber (1); two reports are pending.Conclusions: Seizures arising from insular cortex are not rare and may be underrecognized. Suspicion for insular cortex involvement should be considered in children who present with atypical seizure semiology for suspected lobe of involvement, broad epileptogenic zone on scalp EEG involving opercular region, or abnormality in the insula on anatomic/functional neuroimaging. Failure to identify insular seizures may result in incomplete excisional procedures and poor outcomes. Depth electrodes implanted directly into the insula are useful to evaluate seizure activity. Resection with inclusion of insular cortex in carefully selected pediatric cases can be effective and safe.