Epileptogenic networks in extra temporal lobe epilepsy

Abstract

Extra temporal lobe epilepsy (eTLE) may involve heterogenous widespread cerebral networks. We investigated the structural network of an eTLE cohort, at the postulated epileptogenic zone later surgically removed, as a network node: the resection zone (RZ). We hypothesized patients with an abnormal connection to/from the RZ to have proportionally increased abnormalities based on topological proximity to the RZ, in addition to poorer post-operative seizure outcome. Structural and diffusion MRI were collected for 22 eTLE patients pre- and post-surgery, and for 29 healthy controls. The structural connectivity of the RZ prior to surgery, measured via generalized fractional anisotropy (gFA), was compared with healthy controls. Abnormal connections were identified as those with substantially reduced gFA (z < -1.96). For patients with one or more abnormal connections to/from the RZ, connections with closer topological distance to the RZ had higher proportion of abnormalities. The minority of the seizure-free patients (3/11) had one or more abnormal connections, while most non-seizure-free patients (8/11) had abnormal connections to the RZ. Our data suggest that eTLE patients with one or more abnormal structural connections to/from the RZ had more proportional abnormal connections based on topological distance to the RZ and associated with reduced chance of seizure freedom post-surgery.

Keywords: Connectome; Diffusion MRI; Epileptogenic zone; Extra temporal lobe epilepsy; Network; Resection; Seizure; Structural connectivity; Tractography.

Figure 1.

Methods to compare abnormality of the RZ prior to surgery, in a heterogenous eTLE cohort. (Step 1) Resection zone (RZ) delineated in the post-operative T1_w scan. (Step 2) The mask is manually drawn in the pre-operative T1_w space using the post-operative T1_w scan as an anatomical reference. (Steps 3 and 4) Resection mask is imported into the atlas as a new region, removing previous areas in that location. (Step 5) Connectivity matrix is built for each patient and controls to the corresponding specific patient atlas created in Steps 1–4. Tractography streamlines are imported from the HCP842 population average. GFA is used as a measure inferring connectivity strength. (Step 6) Connectivity strengths for connections to the specific resection location for an individual patient are compared with the same RZ area for the control group. The z-score is then generated by comparing the individual patient with the controls as a measure of abnormality. (Step 7) Z-scores are compared between patients to perform group analyses (e.g., of seizure outcome, proportion of abnormal connections).

Figure 2.

Proportion of abnormal connections decreases with greater topological distance from the resection zone (RZ). (A) Illustration displaying reduced proportional abnormality further from the RZ. In this schematic illustration, three of seven direct RZ connections are abnormal, representing 42.8% of all primary connections. There are five secondary connections (i.e., connections one step removed from the RZ), of which two (40%) are abnormal. Of the three tertiary connections, only one (33.3%) is abnormal. (B) Example violin plot of all z-score connections between differing nodal distances for a single patient highlighted in panel C. Abnormal threshold set at z < −1.96. (C) Proportion of abnormal connections for each patient (n = 11 had at least one abnormal primary connection) at each nodal distance to RZ. Greater network distance from RZ is associated with fewer abnormalities. The purple line indicates the example patient from panel B. Patients without any abnormal primary connections are omitted for clarity. (D) Hierarchical modeling of the patients in panel C. Connections to primary nodes had significantly more abnormal values compared with subsequent node connections, when accounting for patient as a random intercept, thus confirming the visual impression of panel C (p < .05).

Figure 3.

Nodes connected to the RZ have more abnormalities in downstream connections if the upstream connection was abnormal. (A) Illustrative example. Nodes are displayed as circles and lines as connections, abnormal connections and nodes are highlighted in red. A node was defined as abnormal if its upstream connection to the RZ was abnormal. (B) Results from patients with at least one abnormal connection with RZ. There was a higher proportion of abnormal connections if the upstream node had an abnormal connection to the RZ (p < .05, paired t test).

Figure 4.

Group difference in resection zone connectivity and seizure outcome. A majority of seizure-free patients did not have a single abnormal connection to/from the RZ (8/11); in contrast, a majority of non-seizure-free patients had one or more abnormal connections (8/11). An example patient (left and right panel) from each group is highlighted to illustrate the difference between groups. The top part of each example displays a sagittal view of the RZ in red, connected tracts in yellow, and a fractional anisotropy map overlaid on the background. The bottom part illustrates on a 3D glass brain surface the same example tracts highlighted as either normal (yellow) or abnormal (red). * p < .05.