Abstracts

Rationale: Non-lesional focal epilepsy is a challenging clinical problem. A lesion is not present in ~50% of cases of focal epilepsy, which is associated with a less favorable prognosis for surgical treatment. Given the interest in studying seizure propagation networks, we have used resting state functional MRI (rsfMRI), an approach that allows determination of how different brain regions are functionally connected (FC) or co-activated. FC describes the temporal correlation between spatially remote neurophysiologic events, and has been measured using different modalities including rsfMRI. We hypothesize that fMRI will identify aberrant, more bilaterally distributed, functional connectivity between the seizure onset zone (SOZ) and other brain areas.Methods: Two patient populations were used: 1. Patients with medically refractory focal epilepsy and non-lesional structural MRI who subsequently undergo intracranial EEG (icEEG) evaluation. 2. Age and health-matched normal controls. FC maps derived from the SOZ in epilepsy patients (n=6) were compared with maps generated in matched controls. FC maps in patients were derived by first placing a 5mm ROI over the location of the most active electrode within the SOZ, as determined from icEEG, and in a homologous region in a matched control . These maps consist of all voxels whose BOLD time signal is significantly correlated with the time course of the SOZ ROI.Results: After the generation of the maps, different aspects of the connectivity of the SOZ with the rest of the cerebral cortex were calculated including: total number of voxels in maps (reflection of total connectivity), average map Z score (reflection of connectivity strength), and laterality index of the maps (LI) (measure of the laterality of the connectivity). For the total number of connected voxels, a group comparison was performed of the number of significantly connected voxels from six patients. There was no significant difference from controls (p=0.25; two-tailed t-test, paired). The average map Z score was calculated and normalized to each patient’s control. There was no significant difference in the mean Z score of the maps (p=0.21; two-tailed t-test, paired). Finally, a laterality index (LI) of the maps was calculated (a lower number indicates a more distributed, bilateral map). There was a tendency towards lower LI in the epilepsy group, however no significant difference, in this small sample (p=0.1; two-tailed t-test, type 2) (see figure).Conclusions: Results from six patients suggest tendency towards more bilaterally distributed functional connectivity of the seizure onset zone in epilepsy patients compared to controls, despite no significant differences in the total connected brain volume and average connectivity strength. This may reflect aberrant structural and/or functional connections of the SOZ with other brain regions. Furthermore, the SOZ maps may coincide with seizure propagation networks, a hypothesis that we plan to test by overlaying sites of epileptiform activity, derived from icEEG, on the fMRI functional connectivity maps.