Understanding MEG Source Profiles in Tuberous Sclerosis

PURPOSE: Patients with tuberous sclerosis complex (TSC) and medically refractory epilepsy are increasingly referred for epilepsy surgery, but presurgical evaluation is challenging due to multiple tubers and complex seizure onset zones. The aim of this retrospective study was to correlate key characteristics of magnetoencephalography (MEG) sources with intracranial EEG and surgical outcomes in patients with TSC. METHODS: Patients with TSC and MRE who underwent MEG, stereo-EEG (sEEG) monitoring, and have at least 1-year postsurgical follow-up were included. MEG source number, laterality, location, and type (interictal or ictal) were compared with the seizure onset zones, defined by sEEG and postoperative seizure outcomes. RESULTS: Thirty-three patients (15 female patients and 18 male patients) (mean age 8.0 ± 8.2 years) were included. No significant correlation between the laterality of MEG sources (unilateral or bilateral) and surgical outcomes (χ2(1, n = 33) = 0.11, P = 0.75) was found, and there was no significant correlation between the number of MEG sources and surgical outcomes (P = 0.68). MEG-sEEG concordance scores showed a high specificity (86.7%) and positive predictive value (75%) as a predictor of postoperative seizure outcomes, and the concordance score was significantly higher in patients with an ictal MEG (4.4 (ictal) versus 3.0 [interictal]) (P = 0.04). CONCLUSIONS: This study demonstrates a localization concordance between MEG and sEEG that is related to postsurgical seizure outcomes in patients with TSC. However, no correlation between MEG source number, laterality (unilateral vs. bilateral), and surgical outcomes was found, supporting the continued use of MEG as a presurgical epilepsy tool in this patient population, no matter the number of potential source locations.