Abstracts

Rationale: Traditionally epilepsy that arises from a region of brain where an MRI evident abnormality is observed is considered lesional epilepsy. The most common anomalies characterized as lesional cases have included benign tumors and low flow vascular anomalies such as cavernomas. Focal resection of the abnormality (lesionectomy) has been the conventional initial approach however removal of just the imaging abnormality may fail in up 30%. Advances in MRI technology have improved the capability to image more subtle epileptogenic lesions of multiple etiologies including malformations of cortical development (MCD), tubers and subtle areas of encephalomalacia or gliosis that may be actively epileptogenic. As such the definition of lesional epilepsy is expanding and the approach to lesional epilepsy is continuing to evolve. The aim of this study was characterize the contribution of magnetoencephalography (MEG) imaging to the management of a contemporary series of surgical lesional epilepsy patients.Methods: Following IRB approval a retrospective review of epilepsy center databases (UAB or Children s Hospital of Alabama between 2003 and 2010) was performed to identify patients (adult and pediatric) who had epilepsy arising from a region of brain with an MRI evident abnormality or lesion. Clinical data including MRI findings, video-EEG, operation performed, use of frameless navigation, histopathology and seizure outcome was recorded. Imaging data included MRI, MEG, ictal SPECT/SISCOM and PET findings.Results: We identified 99 patients (55 children and 44 adults) who had lesional epilepsy surgery. There were 32 tumors, 34 cases of MRI evident FCD, 18 cases of gliosis or encephalomalacia, 8 phakamatoses and 6 vascular lesions. Thirty five patients underwent MEG studies during their evaluation. In 9 cases MEG utilization extraoperative alone(epileptogenic region identification) wheras in 25 patients MSI data was co-registered to frameless navigation systems. MEG contributed in the tumor cases by directing/expanding electrode placement, defining adjacent eloquent cortex, and in aiding pre-operative localization of dipoles. MEG was highly useful in cases of post infarct and post hemorrhagic epilepsy arising from areas of gliosis and encephalomalacia. In these situations MEG demonstrated dipoles in regions that could not be studied by other functional imaging modalities (IS, PET) (n=4) and expanded the implicated region of ictal onset to prompt more widespread electrode coverage (n=2). In seven patients the MEG studies were minimally contributory. Regression analysis studies are ongoing.Conclusions: MEG studies can significantly contribute to the assessment of lesional epilepsy. Extraoperative utilization can include localization or to implicate areas beyond the lesion in epileptogenesis, to map adjacent eloquent function and to allow study of tissue that is difficult or to study with other functional imaging modalities. Fusion with frameless navigation systems allows intraoperative use to guide IC-EEG electrode placement.