Abstracts

Rationale: TSC patients present a challenge in determining the epileptogenic zone(s) for planning epilepsy surgery.  We studied localizing value of MEG by comparing it with the two EEG studies (EEGs), a ‘scalp EEG done simultaneously during the MEG (MEEG)’ and the long-term video EEG monitoring (VEEG) done for epilepsy surgery evaluation.  Methods: 21 consecutive TSC patients who had MEG were analyzed. Demographics, MEEG, VEEG, MRI and imaging data were also collected. For each MEG source dipole cluster, the degree of concordance to EEGs (MEEG as well as VEEG) was categorized as concordant (MEG dipoles occupied the same EEGs region over no more than two lobes), concordant+ (dipoles in the same region but MEG occupied a larger area than EEGs), Concordant- (dipoles in the same area but MEG occupied a smaller area than EEGs), partially concordant (partial overlap between MEG and EEGs), and discordant (no overlap between MEG and EEGs). Since localization of scalp EEG by conventional visual assessment is necessarily less precise than MEG’s equivalent current dipole modeling approach, these rough categories were considered adequate. MEG dipoles were also analyzed after co-registration with each patient’s brain MRI.  Results: 21 patients had 46 MEG dipole sources; 8 patients had 1, 5 had 2, 5 had 3, 2 had 4, and 1 patient had 5 MEG sources.  MEG dipole sources were localized to fronto-central (25%), perisylvian-opercular (16%), parieto-occipital (17%), temporo-occipital (15%), temporal (11%) and other (16%) regions. On comparing 46 MEG dipoles to MEEG, 50% were concordant, 3% were partially concordant, 30% were concordant-, and 17% were discordant. On MEEG, 54% patients had focal, while 39% had generalized or multiregional spikes.  Of patients (18) with MEEG multiregional spikes, MEG was successful in 77% (14) in showing a focal dipolar pattern. Three dipole groups were MEG unique with no corresponding spike on MEEG. On MEG comparison with VEEG, 28% were concordant, 14% were concordant+, 48% were concordant-, and 10% were discordant. 43% VEEG patients had focal and 57% had generalized and multiregional spikes.  Of 12 with multiregional VEEG spikes, MEG refined the localization in 10 (83%) patients leading to a more precise hypothesis. Overall, 80% of VEEG were localized as a focal dipolar pattern on MEG. On MEG dipole co-registration to the tuberous regions on each patient’s brain MRI, 14% patients were concordant, 10% were concordant+, 52% were concordant-, and 24% were disconcordant. Of 16 patients with multiple bilateral tubers, MEG dipoles showed a more restricted region of epileptogenicity in 11 (69%).  Conclusions: All TSC patients showed positive MEG results during a one hour MEG recording. In 80% of the VEEG and 77% of the MEEG patients, MEG dipoles were more focal than the EEGs.  MEG can help identify areas of epileptogenicity from the more widespread picture on EEGs as well as help in MRI co-registeration for epilepsy surgery planning. Funding: None