Abstracts

Rationale: There are known to be several normal variants on MEG (Magnetoencephalography). Some of them have an epileptic appearance [1] Although some investigators have focused on a single type of MEG normal variants, there have been no studies systematically investigating them in epilepsy patients [1] [2] [3] [4]. It is important to know the characteristics of normal variants on MEG to avoid misinterpretation of epileptiform-appearing transients. The aim of this study is to localize sharp transients on MEG in localization-related epilepsy. Methods: We retrospectively reviewed the patients with medically intractable localization-related epilepsy who underwent MEG during presurgical evaluation from November 2012 to December 2013. We included the patients who 1) showed a single population of interictal dipoles during the MEG recording, 2) had epilepsy surgery after the MEG recording, and 3) became seizure free postoperatively for more than 6 months. We excluded the patients whose resection was distant from the area within which MEG interictal dipoles were distributed. Only signals recorded by gradiometers were used for MEG analysis. After environmental noise reduction by temporally Signal Space Separation (tSSS) [5] and a band-pass filter between 6 and 50 Hz, the waveforms were rectified and the peaks whose amplitude exceeded 10 S.D. of the background were detected. Any peak which was within 400ms from the last selected peak was not selected. We manually removed any activities due to artifacts. Single equivalent current dipoles (ECDs) were calculated at the timepoint of the peak by the least square method using a single-sphere head model fitted to the individual patient. Individual dipoles were accepted for the next phase of the analysis if a) the goodness-of-fit was more than 70 % and b) the confidence volume was less than 1000 mm3. These selected activities were classified into two groups based on the location of the source. The activities whose source was localized within or close to the resection cavity were recognized as epileptic spikes. Those whose source was localized distant from the resection cavity were recognized as sharp transients (benign variants). The locations of the dipoles of the sharp transients were transformed to the standard Montreal Neurological Institute (MNI) 152 space with Functional MRI of the Brain Software Library (FSL). The dipole location was investigated using the Harvard?"Oxford cortical structural atlas (http://www.cma.mgh.harvard.edu/). Results: A total of 10 patients met the criteria. The number of the dipoles localized within the cerebrum was 564. 30% of them (169) were recognized as sharp transients. The dipoles of the sharp transients were localized mainly in the central region (Postcentral Gyrus, Precentral Gyrus. 10 of 10 patients), the antero-medial region (Cingulate Gyrus, anterior division, Juxtapositional Lobule Cortex (supplementary motor cortex), Paracingulate Cortex. 7 of 10), the medial aspect of occipital lobe (Cuneal Cortex, Intracalcarine Cortex, Lingual Cortex, Supracalcarine Cortex. 6 of 10). Conclusions: Patients with localization-related epilepsy tended to have sharp transients in the central region, the antero-medial region and the medial aspect of occipital lobe. In the case where the epileptic focus is located around these areas, special attention should be paid to distinguish between epileptic spikes and benign variants. References 1. Kakisaka Y et al. Journal of clinical neurophysiology 2013;30(3):235 2. Dehghani N et al. Human brain mapping 2011;32(12):2217-27 3. Lin Y-Y et al. Seizure 2003;12(4):220-25 4. Nakasato N et al. Electroencephalography and clinical neurophysiology 1990;76(2):123-30 5. Taulu S et al. Physics in medicine and biology 2006;51(7):1759 Funding: none