Abstracts

Epilepsy is the most frequent manifestation of congenital mass lesions including cavernous angiomas (CAs) and benign brain tumors (BTs). Lesionectomy in combination with removal of the surrounding epileptogenic areas is usually required in achieving seizure cessation. To investigate epileptogenecity of patients with CAs and BTs, we analyzed C-11 flumazenil (FMZ) and F-18 fluorodeoxyglucose (FDG) PET abnormalities in those patients. We studied FMZ/FDG PET abnormalities of 8 patients (4 CAs and 4 BTs) with intractable temporal lobe epilepsy. Using image analysis software Dr. View (Asahi Kasei Information Systems, Tokyo, Japan) working on a personal computer, the region of interests (ROIs) were placed on the lesion and contralateral brain over the MRI and then transferred to the FDG/FMZ-PET images co-registered with MRI. Areas with peri-lesional decrease in FMZ/FDG uptake were visually defined and their volumes were analyzed on the individualized co-registered images. A voxel-based analysis was also performed using statistical parametric mapping (SPM) as compared with normal control database (FDG: n=21, FMZ: n=19). Invasive monitoring was performed in one patient with left posterior temporal lesion. All the 8 patients underwent lesionectomy along with the resection of surrounding epileptogenic areas under electrocorticography guidance. 1)The extent of FMZ-PET abnormalities was significantly larger than the corresponding structural lesions, but it was significantly smaller than the areas with decreased FDG-PET uptake. SPM analysis indicated a significantly decreased peri-lesional FDG uptake in all of the 8 patients. Decreased FMZ binding was found in 3 patients and, in all of them, the area was smaller than that with decreased FDG-PET uptake. 2) Patients with CAs had smaller areas with decreased uptake both of FMZ/FDG than those with BTs. SPM analysis showed significantly decreased peri-lesional FMZ binding in 3 of 4 patients with BTs and none of 4 patients with CAs. 3) In the area remote from the structural lesion, SPM analysis detected the area with decreased FDG uptake in all of the 8 patients and decreased FMZ uptake in 2 patients. 4) All patients except for one with incomplete resection of epileptogenic areas achieved seizure-free outcome. The ROI and SPM analysis of FMZ/FDG PET images indicated that TLE patients with BTs and CAs had epileptogenic areas surrounding the lesion. The present study also suggested that peri-lesional epileptogenic areas were larger in BTs than in CAs.