Abstracts

RATIONALE: To analyse the long-term results of first 10 years of temporal lobe epilepsy (TLE) surgery in a new epilepsy surgery centre, and evaluate the value of different preoperative investigations with respect to outcome.
METHODS: Kuopio University Hospital epilepsy surgery program was started in 1988. Between 1988-1999 140 patients (67 females,73 males) with drug-resistant TLE were operated on at our hospital. Preoperative investigations included neurologic evaluation, MRI, ictal video EEG recording, neuropsychologic and psychiatric evaluation, and Wada test. Since 1993 a standardized MRI protocol with a 1.5T imager was introduced, and temporal lobes were imaged with a tilted coronal MP-RAGE sequence with parameters 10/4/1 (TR/TE/excitations), inversion time 250ms, flip angle 12[degree]; FOV 250 mm, matrix 256x192. Hippocampal and amygdaloid volumes were available for 67 patients. The median age of the patients was 32y, the median duration of epilepsy 19y, and preoperative seizure frequency 10-1655 seizures/y. The aim of surgery was preoperatively assessed as curative (n=103) or palliative (n=37) based on the expexted outcome.
RESULTS: When all patients were evaluated together, 62% achieved good postoperative outcome (Engel I-II) after surgery (median follow-up 5.2y). The outcome was expectedly better in the curative subgroup, where 55% of patients were completely seizure-free and 15% had rare seizures after surgery. Only 35% of palliative patients became seizure-free and 5% had rare seizures (p[lt]0.05 between subgroups). After the introduction of a standardized MRI protocol in the middle of the observation period, the results improved significantly: 59% of curative patients became completely seizure-free and additionally 17% had good outcome with rare seizures (Engel I-II, 76%)(p[lt]0.05). A similar trend was seen in the palliative group where 71% of patients achieved at least an 80% seizure reduction in the postoperative follow-up. In a logistic regression analyses quantitative hippocampal volume at the side of the epileptic focus, and unilateral hippocampal and/or temporal cortical atrophy in the qualitative MRI predicted good postoperative outcome (p[lt]0.05 in both). Life table analyses showed that 74% of patients with original good outcome (Engel I-II) remained in good seizure control in the long term follow-up, and 80% of seizure relapses occurred within 2y from operation.
CONCLUSIONS: The results confirm the prognostic value of both qualitative and quantitative MRI in the preoperative evaluation of patients with drug-resistant TLE. After the introduction of a standardized MRI protocol 76% of patients with unilateral TLE achieved good seizure control in the long-term follow up. Also well selected palliative patients clearly benefit from surgery.
Support: This study was supported by Academy of Finland, Finnish Cultural Foundation, Kuopio University Hospital Research Fund, University of Kuopio, and Vaajasalo Foundation.