Abstracts

Rationale: ICEEG recording of seizures is used widely in patients undergoing evaluation for the surgical treatment of epilepsy. Optimal placement of ICEEG electrodes is critical for the success of ICEEG recordings to prevent incomplete or misleading localization of the epileptogenic zone. In addition ICEEG is costly and associated with considerable morbidity. Noninvasive approaches are necessary to optimize and ideally to obviate placement of intracranial electrodes. The objective of this study was to retrospectively examine the concordance (or lack thereof) between preoperative MEG and ICEEG findings. Methods: Eighty five invasive evaluations were performed after our MEG became operational in aug/2008 until may/2009. Preoperative MEG recordings were performed in 19% these patients (15 patients total, ages 12 to 57 years, 10 females and 5 males). All patients had a diagnosis of refractory epilepsy and underwent ICEEG electrode implantation after review of all their preoperative information (including their MEG data) during a multidisciplinary patient management conference. Concordance between MEG and ICEEG with respect to sublobar localization was defined based on the criteria used in the previously published prospective study by Knowlton et al. (Ann Neurol. 2006 May;59(5):835-42). Results: The majority of these patients had extratemporal epilepsies, and only five were suspected to have a diagnosis of temporal lobe epilepsy. MEG did not show interictal epileptiform discharges in four patients (26.7%). In two of the patients (13.3%)ICEEG had to be aborted after less than 24hours to prevent postoperative complications. In the remaining 10 patients, where both MEG and ICEEG information were available, excellent sublobar concordance was found in 4 out of 10 (40%), 2 temporal and 2 extratemporal. With respect to localization within the same lobe, MEG and ICEEG were concordant in an additional 3 patients (30%). Neither MEG nor ICEEG were able to localize the possible epileptogenic zone in one patient. Finally, there was no concordance between the two modalities in the remaining 2 patients (20%). Conclusions: This study suggests that concordance between MEG and ICEEG at the lobar level can be found in almost 50% of patients (7 out of 15) with intractable focal epilepsy undergoing a presurgical evaluation with both modalities. The degree of concordance is higher when considering only patients in whom both MEG and ICEEG data were available. Furthermore, MEG may provide unique information in patients when ICEEG is contraindicated or has to be aborted. Although this preliminary study is limited by the relatively small number of patients and retrospective design, it suggests that MEG can help to better delineate the epileptogenic zone and/or tailor the ICEEG evaluation in epilepsy surgery candidates referred for ICEEG.