Abstracts

Rationale: Hemisheric and lobar concordance ratios between imaging, electrophysiologic and cognitive tests were calculated in subjects with intractable epilepsy being evaluated for epilepsy surgery. Methods: Data from 9 patients with intractable epilepsy being evaluated for surgery were analyzed. All patients had scalp video-EEG monitoring (svEEG), 1.5T or 3T MRI, PET) and formal neuropyschologic testing (NPT). Six patients had MEG with simultaneous EEG recordings (50 min) (CTF Omega system); a current density source reconstruction algorithm (sLORETA) was used to identify the spatial peaks of activity. Four patients had individually tailored intracranial video-EEG monitoring (ivEEG). Individual tests were assessed for hemispheric and lobar locations based on visual analysis. The data from each test were compared pair-wise (e.g., MEG & MR)to calculate concordance ratios. The ratio for hemispheric concordance was calculated as the sum of concordance values (0-nonconcordant, 0.5-partial concordance, 1-concordant) divided by the number of the patients. Lobar concordance rates were calculated identically. Results: Six of 9 patients had nonlesional MRI; two patients had left hippocampal atrophy and one bilateral parietooccipital cortical dysplasia. Hemispheric concordance rates were at best modest for the different modalities: the highest level was 0.75 between svEEG-ivEEG. NPT had concordance ratios of 0.67 with MEG and 0.61 with PET. MEG and svEEG had a concordance ratio of 0.58 (Table1). Among patients that had partial or full hemispheric concordance, lobar concordance ratio was 1.00 for scalp and invasive EEG monitoring. MEG and MRI had high concordance ratios, but there were only 2 patients. NPT had lobar concordance rates of 0.67 with svEEG- ivEEG and 0.62 with PET (Table2). Three patients had lesions on MRI. Only 1 had ivEEG. The hemispheric concordance ratios for these patients were 1.00 between MEG and NPT, MRI and NPT; 0.83 between PET and svEEG, MRI and svEEG, and NPT and svEEG;.0.75 between MRI and MEG and 0.67 between PET and MRI and PET and NPT; lobar concordance ratios were 0.83 between PET and svEEG and NPT and svEEG. Among for 4 patients that had ivEEG, the hemispheric concordance ratios were 1.00 between MRI and NPT, svEEG, ivEEG, PET; 0.75 between PET and svEEG, ivEEG, and MEG, There was no concordance between MRI and MEG or PET and MEG. Highest lobar concordance rate was 0.75 for svEEG and ivEEG, and PET and NPT. All the other tests had equal or less than 0.50 rate of concordance. Conclusions: The svEEG and ivEEG had as expected, the highest concordance ratios. Both hemispheric and lobar concordance ratios between MEG and other modalities were modest and lower than previously reported. NPT and PET had modest to high concordance ratios. Larger size studies are needed to assess the feasibility of non-invasive localization of epileptogenic tissue.