Abstracts

Rationale: New and innovative neuro-imaging/neurophysiology studies have led to an increase in knowledge of seizure generation of focal epilepsies; however, there is paucity of research in bitemporal lobe epilepsy (BTLE). The aim of this study was to assess if seizure generation in BTLE is variable over time occurring over either temporal lobes independently or if one side remains predominantly epileptogenic, leading to propagation of ictal patterns to the other temporal lobe. Methods: Following IRB approval, 4 patients with a clinical diagnosis of BTLE were studied with MEG. A 10 minute resting-state neuroimaging study was performed: 148 channel MEG combined with 32 EEG. Two of the 4 patients returned for a repeat MEG scan. MEG and EEG were reviewed for interictal and ictal patterns of epilepsy. Equivalent current dipoles (ECD) and Coherence source imaging (CSI) analysis were performed on the MEG data, results were coregistered and displayed on the patient's MRI. Coherence is a measure of oscillating neuronal activity and their synchronicity across the cortex, which defines the neuronal network connectivity. Granger causality was applied to the MEG data to assess the direction of the information flow in the epileptic network. Results: Four patients (3 females) with a mean age of 40 years (range 26-50y) and mean age of epilepsy onset of 13.6 years (range 0.5-24 y) with video-EEG findings consistent with BTLE were included. MRI brain showed unitemporal pathology (temporal atrophy and/or HS) in 1, bitemporal in another and unremarkable in the other 2 patients. Epileptiform activity was detected by MEG in only two of the four patients. ECD was used to map these interictal spikes to the temporal gyri and to the fusiform gyri in both cases. MEG-CSI showed consistent high coherent activity in both the right and left fusiform gyrus of all 4 patients. Three of these patients displayed the right fusiform as the sending site and the left as a receiver of information area using granger causality. The other patient had the left fusiform as the sending site and the right as the receiver. This same activity was found to have reversed its direction of information flow in the repeated MEG scans performed on 2 of the patients. Conclusions: This study reveals that the dominant network in BTLE may include areas localized over the fusiform gyrus that can be detected by MEG in the absence of interictal epileptic activity using coherence source imaging. Over time these fusiform areas of activity may lead to seizure generation in either temporal lobe independently. This information provides further clues to understanding the patho-physiologic substrate of BTLE. Detection of this coherent activity in the fusiform gyri may help in exploring further treatment options for better seizure control in BTLE.