Abstracts

Rationale: Patients with medically refractory epilepsy and MRI findings of mesial temporal sclerosis often undergo single-stage anterior temporal lobectomy when presurgical evaluations are consistent with localization to the corresponding temporal lobe. In the absence of lesions in the temporal lobe the surgical target is not as obvious. We reviewed intracranial EEG localization of epileptogenic zones in patients with Video-EEG findings suggestive of temporal lobe localization and no MRI lesions in the corresponding temporal lobe. Methods: We retrospectively searched the adult epilepsy surgery database at the Medical College of Wisconsin from 2002 to 2011 and identified 33 patients with Video-EEG findings that suggested temporal lobe origin of seizures, but with no ipsilateral temporal-lobe lesions on high-resolution MR imaging, who had undergone ictal intracranial EEG recordings as part of their pre-surgical evaluation. We reviewed the intracranial EEG data from these patients to determine whether a standard anterior temporal lobectomy would have included the epileptogenic zone that was identified by the invasive recordings. Results: Mean age at the time of surgery for our patients was 34.03 ( 10.9) years, with mean age of seizure onset 15.30 ( 9.9) years, and mean epilepsy duration of 18.72 ( 13.07) years. In 14 (42.4%) of our patients, invasive EEG demonstrated ictal onset zones that included extratemporal regions. Of these epileptogenic zones were temporo-frontal in 4 (12.12%), temporo-parietal in 3 (9.09%), extratemporal in 3 (9.09%: 2 frontal and 1 supplementary motor area), temporo-parieto-occipital in 2 (6.06%), and temporo-fronto-parietal and posterior temporal in 1 patient each in 1 (3.03%). PET results were available for 12 of these patients, but was normal in 5 (41.66%) patients, and concordant with the intracranial EEG only in 2/12 (16.66%) patients. Of the 33 patients 3 did not undergo a resective procedure after the invasive EEG study: 2 due to bilateral independent seizures from the mesial temporal structures and one due to overlap of the ictal onset zone with language areas. Surgical outcome was Engel class I in 25 patients (83%), class II in 4 (12.12%), class III and IV in 2 (6.06%) patients each, with mean follow-up period of 3.69 ( 2.48) years.Conclusions: Our retrospective review of invasive EEG findings in patient with non-lesional epilepsy with apparent temporal lobe localization on scalp Video-EEG suggests that well over a third of these patients have epileptogenic networks that include areas outside the antero-mesial temporal regions. Single-stage anterior temporal lobectomies cannot be expected to ameliorate seizures in these cases. Localization of seizures using ictal intracranial EEG recordings should be the norm in this patient group before formulating a plan for resective epilepsy surgery.