Abstracts

Rationale: The purpose of this study was to assess if in cases in which the traditional presurgical diagnostic evaluation of continuous video-EEG monitoring, neuropsychological testing, and high resolution MRI does not provide a clear surgical target, completion of both MEG and SISCOM adds complementary localizing information to aid in epilepsy surgery planning compared to either of these modalities alone. Methods: All consecutive presurgical cases at our epilepsy center between 2007 and 2009 undergoing both MEG and SISCOM were included in this study. A total of twelve cases had complete datasets for analysis. Spontaneous MEG data was acquired and analyzed using a Neuromag system. SISCOM data was evaluated with Analyze software which was also used to coregister MEG and SISCOM. The coregistered data were visually inspected for degree of overlap and classified as either concordant or discordant. Concordant results were further classified as lobar or sublobar. Results: Of the twelve cases, five were concordant in terms of MEG and SISCOM results. Four of the five of these had underlying cortical dysplasia. The fifth case was nonlesional. Of the four concordant dysplasia cases, two were initially felt to be nonlesional after completion of the traditional evaluation. After the subsequent finding of concordant MEG and SISCOM at a sublobar level however, further analysis of MRI data revealed findings consistent with cortical dysplasia that was not initially identified. In one of these cases re-analysis of the original MRI showed a highly focal area of abnormal cortical thickening by visual inspection as well as cortical thickness analysis with Freesurfer software and later confirmed by pathology. In the other case, a high field, high resolution MRI was subsequently completed and revealed findings consistent with cortical dysplasia that was not seen with prior conventional MRI. Analysis of the single, nonlesional concordant case also showed agreement at a sublobar level. In this case however, a second more subtle area of abnormality was seen on SISCOM that did not have a MEG correlate. Because this more subtle region seen only on SISCOM correlated with the semiology of the patient s seizures, and SISCOM is a measure of ictal activity, while MEG is typically interictal, it was felt that the more obvious area of MEG and SISCOM concordance in this case represented a false positive. Conclusions: While previous studies have shown that both MEG and SISCOM can independently aid in the placement of intracranial electrodes in epilepsy surgery, this case series suggests that the use of both modalities together may further improve presurgical planning by the identification of structural lesions in select cases that may otherwise be considered nonlesional. As the single false positive case illustrates however, careful analysis of traditional diagnostic test results along with both MEG and SISCOM data and an understanding of the limitations of each is required to avoid mistaking an irritative lesion for an epileptogenic lesion.