Abstracts

Utility of MEG in Evaluations for Epilepsy Surgery: A 10-year Review from a Single Epilepsy Center

Abstract number : 3.148
Submission category : 3. Neurophysiology / 3D. MEG
Year : 2019
Submission ID : 2422046
Source : www.aesnet.org
Presentation date : 12/9/2019 1:55:12 PM
Published date : Nov 25, 2019, 12:14 PM

Authors :
Daniel Champeau, Medical College of Wisconsin; Chad Carlson, Medical College of Wisconsin; Christopher T. Anderson, Medical College of Wisconsin; Serena Thompson, Medical College of Wisconsin; Patrick Bauer, Medical College of Wisconsin; Linda Allen, Medi

Rationale: To evaluate the clinical utility of magnetoencephalography (MEG) for identifying likely seizure-onset-zones (SOZs) during the pre-surgical evaluation of patients with drug-resistant epilepsy (DRE). Methods: We retrospectively identified all patients (N=107) with DRE who completed an invasive EEG and/or ablative epilepsy surgery over a ten-year period at the Medical College of Wisconsin Adult Epilepsy Center and underwent a MEG study during presurgical evaluation. We retrieved demographic and clinical variables including gender, age at seizure onset, seizure lateralization/localization from video-EEG studies, MR imaging findings, localization of irritative zones (IZ) from MEG, SOZs identified on iEEG, and resection zone (RZ). The relationship between MEG and iEEG findings were classified as concordant when the IZ localized by MEG was concordant with SOZs identified by iEEG at the lobar level, partially concordant in cases of a partial lobar overlap between findings of the two modalities, or discordant in cases with no lobar overlap between the two. Results: Of 107 patients who had a pre-surgical MEG study, 84 (79%) also underwent iEEG, while 23 (21%) underwent single-stage resections without an iEEG study. Of the 84 patients (79%) who underwent iEEG studies, 33 (39%) had concordant MEG and iEEG findings, 30 (36%) were partially concordant, and 4 (5%) were discordant, while 20% of cases had an uninformative MEG study. In all, 75 % of patients who had iEEG studies showed full or partial concordance between MEG and iEEG. Of the 107 patients, 90 (84%) went on to a surgical resection. Of these, 43 (47%) showed concordance between MEG and the resection zone (RZ), 24 (27%) partial concordance, 2 (2%) discordance, while MEG was uninformative in 23%. In all, 74% of cases who underwent resections had MEG findings which at least partially implicated the RZ. In 17/107 cases (16%), a resectable surgical target was not identified after iEEG and no ablative procedure was performed. Conclusions: Our clinical experience supports the conclusions of prior studies demonstrating the value of interictal MEG source localization in pre-surgical planning for DRE.  In our series, lobar concordance or partial concordance was seen in ~75% of cases with respect to SOZs identified with iEEG and ~74% with respect to RZs.  When the nondiagnostic MEG studies are excluded, these numbers improve to 94% and 97%, respectively. Funding: No funding
Neurophysiology