Abstracts

Rationale: Patients with medically resistant temporal lobe epilepsy (TLE) can frequently be rendered seizure free after temporal lobectomy. MRI findings consistent with unilateral medial temporal sclerosis (MTS) are known to be predictive of seizure free outcome in patients with TLE. Here we investigate non-invasive pre-surgical testing in patients with pathologic evidence of MTS. Methods: A retrospective analysis of consecutive patients with mesial temporal sclerosis as defined by pathology, who underwent prolonged video-EEG monitoring and subsequent temporal lobectomy for medically intractable seizures at the Mayo Clinic in Rochester Minnesota from 1997 to 2006 was conducted. In particular, the following data points were collected: preoperative MR imaging, preoperative interictal scalp EEG, preoperative ictal scalp EEG, and outcome according to Engel classification. Results: 127 (68 female) consecutive patients with pathologically proven mesial temporal sclerosis who underwent temporal lobectomy were identified. Thirteen patients were excluded from the analysis due to unavailability of outcome classification and inadequate follow-up. The median age was 37 years (range 16-74 years) and median follow-up 37 months (range 3-119 months). Of the 127 patient 105 (75%) were followed for more than a year. Of these 105 patients, 79% had excellent outcomes. Of the 22 patient with less than 12 months follow up, 72% had excellent outcomes. Mesial temporal sclerosis (MTS) was detected in MRI in 84% of patient. Unilateral anterior temporal discharges were observed among 62% of patients in interictal EEG. Ictal EEG was localized to the temporal lobe in 82% of patients. There does not appear to be a statistically significant relationship between Engel class outcome and MRI findings(p= 1), ictal EEG (p= 0.21), or interictal EEG (p=0.71) Conclusions: As previously reported in the literature, we demonstrate a high rate of excellent seizure outcome following temporal lobectomy in patients with mesial temporal sclerosis. The inability to find outcome predictive factors may be due to the low proportion of patients with non-excellent outcome, which reduces the power to detect significance in our statistical analysis.