Abstracts

Rationale: Temporal lobe epilepsy (TLE) has increasingly proven to be a disease of brain networks, with mounting evidence hinting at the importance of connections between the affected medial temporal lobe and contralateral and/or extratemporal brain regions. Such connections may have either functional/compensatory or pathologic significance. Resting state BOLD fMRI has reliably demonstrated specific, reproducible brain networks in healthy controls. Here we used this technique to investigate whether the medial temporal regions in TLE patients show abnormal connectivity both intrinsically (within the temporal regions) as well as extrinsically with contralateral temporal brain regions. Methods: Twelve patients with TLE (6 with left TLE, 6 with right TLE) were consecutively enrolled. Each patient had confirmed medial temporal epilepsy with hippocampal atrophy and EEG localization of seizures. Subjects underwent a resting-state BOLD fMRI session including 2-5 BOLD acquisition runs, each lasting approximately 6 minutes, acquired with a 3.0 T MRI scanner. During each BOLD run subjects were instructed to rest with their eyes closed. High-resolution structural T1 and T2-weighted images were also acquired for each subject for registration and atlas transformation. A group of healthy control subjects (n = 31) acquired under similar conditions was used for comparison. Eight regions of interest (ROIs) were defined a priori based on anatomical landmarks, including the entorhinal, hippocampal head, hippocampal body, and parahippocampal region in each hemisphere. Each ROI was used as a seed in a functional connectivity analysis using typical methods (e.g. Fox et al., J. Neurophysiol., 2009, 101:3270-83). Voxelwise connectivity maps for each seed were used to compute connectivity strength between regions. These measures were entered in a random effects model. Results: Results showed significantly decreased connectivity (p <.05) in the anterior medial temporal ROIs (entorhinal and hippocampal head) of the affected hemisphere compared to the contralateral hemisphere. Interestingly, the affected hippocampal head showed stronger connectivity with the contralateral hippocampal head than with the ipsilateral medial temporal regions (entorhinal, hippocampal body, and parahippocampal regions). This suggests both reduced functional cross-talk between the diseased hippocampal head and neighboring medial temporal regions, and increased functional communication with the contralateral hippocampal head, possibly in compensatory fashion. Conclusions: Preliminary results argue for a potential marker of focal medial temporal pathology in the form of locally reduced functional connectivity, which may aid in localization for diagnosis and presurgical planning, as well as for assessment of therapy effects. Furthermore the noted strong interhemispheric connectivity suggests the existence of partially preserved function in the affected hippocampal head. Tracking the amount of preserved connectivity in TLE-affected regions at the individual subject level may help predict which patients are at greatest risk of postsurgical functional decline.