Abstracts

Rationale: Epilepsy surgery is currently the preferred treatment modality in selected patients with drug-resistant epilepsy. Favorable outcome of epilepsy surgery depends critically upon selection of appropriate surgical candidates and accurate delineation of the epileptogenic zone. Magnetoencephalography (MEG) has been shown to provide clinically valuable information to these ends. However, the extent of its contribution is not established and varies across MEG centers. It is likely determined by the overall approach to the presurgical evaluation and MEG analysis as well as the diagnostic and prognostic accuracies attained in each center.

Methods: In this retrospective cohort study, we assess the diagnostic accuracy of the MEG interictal dipole localization approach used in our center in identifying the epileptogenic zone and prognosticating the postsurgical seizure outcome.

Results: We studied 45 surgical patients, with a minimum postsurgical follow-up of 12 months. Interictal epileptiform discharges were detected in 34 (76%) patients. Among them, MEG localization was monofocal (one tight cluster of dipoles) in 24 (71%) cases. Monofocal localization predicted postsurgical seizure freedom with an accuracy of 74% and statistically significant diagnostic odds ratio (DOR) of 5.7. MEG localization was concordant with the surgical resection site (i.e. MEG cluster was completely resected) in 18 patients, 17 (94%) of whom achieved seizure freedom. Localization was discordant (i.e. nonlocalizing, diffuse localization, or MEG cluster was partially or not resected) in 16 patients, only six (38%) of whom became seizure-free. Resection of MEG focus predicted seizure freedom with an accuracy of 79% and statistically significant DOR of 28.3. Patients with concordant MEG localization had a significantly higher chance of seizure freedom than patients with discordant findings (P = 0.0004). The diagnostic accuracies were overall higher for extratemporal than temporal lobe epilepsy (DORs for predicting seizure freedom: monofocal localization, 10.7 vs. 3.7; MEG focus resection, 45 vs. 7.5).

Conclusions: These results demonstrate the good accuracy of our interictal MEG analysis approach in localizing the epileptogenic zone and prognosticating seizure-free outcome. They confirm that MEG is a clinically valuable noninvasive diagnostic modality, and add to the growing evidence supporting its regular utilization in guiding the selection of epilepsy surgery candidates and delineation of the epileptogenic zone.

Funding: Please list any funding that was received in support of this abstract.: No Funding.