Structural Connectivity Underpinning of MEG Clusters Outside of the Epileptogenic Zone
Abstract number :
1.245
Submission category :
5. Neuro Imaging / 5A. Structural Imaging
Year :
2023
Submission ID :
214
Source :
www.aesnet.org
Presentation date :
12/2/2023 12:00:00 AM
Published date :
Authors :
Presenting Author: Ryuzaburo Kochi, MD, PhD – Cleveland Clinic
Spencer Morris, MS – Cleveland Clinic; Imad Najm, MD – Cleveland Clinic; Hiroatsu Murakami, MD, PhD – Cleveland Clinic; Irene Wang, PhD – Cleveland Clinic
Rationale:
Magnetoencephalography (MEG) is an effective clinical tool for epilepsy. MEG interictal spikes clusters show high spatial concordance with the epileptogenic zone (EZ). However, in some cases, MEG clusters can be detected outside of the EZ. Propagation of interictal spikes, which can be facilitated by white matter tracks, is a plausible reason for this mismatch. Here, we investigated the relationship between the location of “mismatched” MEG clusters and structural connectivity by diffusion tensor imaging (DTI) analysis in patients with medically intractable focal epilepsy. We hypothesize that a MEG cluster showing spatial mismatch may have increased structural connectivity to the EZ.
Methods:
We retrospectively identified patients with medically intractable focal epilepsy from 2009 to 2017 with the following criteria: 1) underwent epilepsy surgery and achieved seizure freedom for > 1 year, 2) had a MEG cluster (tight/loose) outside of the EZ, and 3) had a pre-operative DTI scan of sufficient quality. DTI data was also collected from age-and-gender-matched health controls (HCs). MEG data were recorded from a 306-channel whole-head MEG system. Information of MEG interictal spike clusters was collected from official medical records, and the clusters’ anatomical locations were further categorized by sublobar regions using the Virtual Epileptic Patient (VEP) atlas. The DTI acquisition was performed in axial plane: 60 slices, TR = 6900 ms, TE = 71 ms, thickness = 2 mm without gap, voxel size = 1.9 x 1.9 x 2mm3, 30 directions, b = 1000 s/mm2. DTI analysis was performed with MRtrix which uses fiber bundle element-based analysis of the apparent fiber density. Ten million whole-brain streamlines were generated and weighed using SIFT2, and further used to create a weighted-undirected structural connectome. We normalized the raw scores of the structural connectome by performing percentile ranking within each patient, and used the values as a measure for connectivity strength. We compared each patient’s connectivity strength between the location of the MEG cluster and the EZ with homotopic regions in HCs. For patients whose mismatched MEG clusters were ipsilateral to the EZ, we also compared the connectivity strength with homotopic regions on the contralateral side.
Results:
Ten patients and 23 HCs were included. In six patients, the mismatched MEG clusters were on the contralateral side, and in four patients ipsilateral. Six patients showed higher connectivity strength between the location of the MEG cluster and the EZ, compared to the homotopic regions in HCs; two of them reached significance (p< 0.05). An illustrative case is shown in Figure 1. In the patients with ipsilateral pattern, one showed significantly higher connectivity than homotopic regions on contralateral side (p< 0.05).
Neuro Imaging