Abstracts

Rationale: The degree of success of epilepsy surgery is influenced by the ability to localize the epileptogenic focus. Imaging studies such as MRI and 18[F]fluorodeoxyglucose Positron Emission Tomography (FDG-PET) have been used to identify or confirm ictal foci in preparation for surgery. To our knowledge, there is little information about the correlation between PET, MRI, intracranial EEG (icEEG), and pathological findings in the pediatric population. The main goal of this study is to compare these four modalities. Our second goal is to examine the hypothesis that PET scans are more sensitive than MRI in localizing epileptogenic areas. Methods: We reviewed medical records of thirty-eight patients, ages 1 to 17 years, with medically refractory epilepsy who underwent removal of the epileptogenic focus at the University of Chicago between May 2006 and March 2010. Patients with pre-operative MRI and PET, and post-operative pathology on file were included in the study (n=20). In all patients, epileptogenic foci were identified pre-operatively by icEEG monitoring. After removal, surgical specimens were analyzed by inspection of their gross anatomy, as well as by standard staining for cytological architecture, and immunohystochemistry. Pathological findings were then compared to the findings on pre-operative icEEG, MRI and PET. Results: Intracranial EEG was used as the gold standard method for localization of the epileptogenic foci. All the patients in our series (n=20) had focal icEEG findings. PET scans showed abnormal findings in 75% (n=15) of cases. One PET positive patient showed bilateral focal hypometabolism. 13 patients with lateralizing PET demonstrated regional cortical hypometabolism ipsilateral to the icEEG focus. One showed an ipsilateral hypermetabolic focus. PET provided correct lobar localization in 64% (n=9) of these patients. MRI was abnormal in 55% (n=11) of cases. Of these patients, one had a non-lateralizing abnormality (partial agenesis of the corpus callosum), and the other had a lesion contralateral to the epileptogenic focus. The rest 45% (n=9) patients had ipsilateral MRI abnormalities. All nine patients who had lesions on MRI also had abnormal findings on PET scan. Pathological examination of the ictal foci identified by icEEG revealed abnormal findings in 70% (n=14) of cases in our series. Of the patients with abnormal PET (n=15), 80% (n=12) had abnormal pathological findings. Conclusions: This study highlights the importance of including PET scans in the pre-operative workup for epilepsy surgery in the pediatric population. Our results suggest the following: 1. PET is more sensitive than MRI in identifying abnormal areas in the brain. 2. Although PET accurately lateralized the side of abnormal function, it was able to precisely localize epileptogenic foci in 64% of cases; therefore, icEEG monitoring must be used to further define these areas. 3. 80% of cases with abnormal PET demonstrated abnormal pathological findings these areas.