Network coupling and surgical treatment response in temporal lobe epilepsy: A proof-of-concept study

Abstract

Objective: This proof-of-concept study aimed to examine the overlap between structural and functional activity (coupling) related to surgical response.

Methods: We studied intracranial rest and ictal stereoelectroencephalography (sEEG) recordings from 77 seizures in thirteen participants with temporal lobe epilepsy (TLE) who subsequently underwent resective/laser ablation surgery. We used the stereotactic coordinates of electrodes to construct functional (sEEG electrodes) and structural connectomes (diffusion tensor imaging). A Jaccard index was used to assess the similarity (coupling) between structural and functional connectivity at rest and at various intraictal timepoints.

Results: We observed that patients who did not become seizure free after surgery had higher connectome coupling recruitment than responders at rest and during early and mid seizure (and visa versa).

Significance: Structural networks provide a backbone for functional activity in TLE. The association between lack of seizure control after surgery and the strength of synchrony between these networks suggests that surgical intervention aimed to disrupt these networks may be ineffective in those that display strong synchrony. Our results, combined with findings of other groups, suggest a potential mechanism that explains why certain patients benefit from epilepsy surgery and why others do not. This insight has the potential to guide surgical planning (e.g., removal of high coupling nodes) following future research.

Keywords: Diffusion tensor imaging; Drug-resistant epilepsy; Resective surgery; Stereotactic EEG; Temporal lobe epilepsy.

Figure 1. Schematic of study design.

(A) Functional connectivity. Implanted sEEG electrodes record neural activity in drug-resistant temporal lobe epilepsy (TLE) patients. Functional connectivity can be measured by the statistical relationship between pairs of sEEG contact recordings. Beta coherence was used to create functional networks for each patient (red matrix). (B) Structural connectivity. Diffusion tensor imaging (DTI) was acquired in the same patients that underwent electrode implantation. Tractography was used to model the white matter tracts of the brain. Regions of interests (ROIs) were drawn around each contact, and fractional anisotropy (FS) was measured along the tracts between pairs of contact triplets (see text for further details on electrode groupings) to create a structural network for each patient (blue matrix). C. Jaccard index. The similarity between the structural and functional networks was measured using the Jaccard index. Both the structural and functional networks were matched into similar dimensions, and the upper triangular matrix of each network was transformed into a vector and then binarized. The binarized vectors were used to compute the Jaccard index.

Figure 2. Example of stereoelectroencephalography (sEEG) recording.

16-second ictal sample of an intracranial sEEG recording obtained from a patient undergoing Phase II investigation for presumed temporal lobe epilepsy. Depth electrodes for this patient included a left medial hippocampus (LMH), left heterotopia (LHET), right anterior hippocampus (RAH), right medial hippocampus (RMH), right posterior hippocampus (RPH), right heterotopia (RHET). Display conditions are LLF = 1 Hz, HFF = 70 Hz, Timescale = 35mm/sec (red line indicates seizure onset marked by neurophysiologist).

Figure 3. Group averages of responders and non-responders.

(A) All networks shown follow the same schematic represented here. Each matrix cell represents the connectivity (structural or functional) between pairs of contact triplets (deep, middle, or superficial; see Methods for details on contact groupings). Electrode LA is shown in the inset where the first row/column represents the deep (D) contacts, the second row/column represents the middle (M) contacts, and the third row represents the superficial (S) contacts. All electrode names are shown as abbreviations (LA to RPI) to the left of the example network. (B) The group average for fractional anisotropy (FA) is shown. (C) The group average for beta coherence is shown for rest, early, mid, and late seizure epochs. (D) The total number of seizure clips for each group is also shown, representing the number of samples acquired. R= right; L=left A = amygdala; AH = anterior hippocampus; PH = posterior hippocampus; AI = anterior insula; PI = posterior insula; LF = lateral frontal lobe; MF = medial frontal lobe

Figure 4. Non-responders to epilepsy surgery have higher structural and functional coupling (Jaccard index) in the rest and the early and mid-phase of a seizure than responders.

The Jaccard index represents the coupling, or the agreement, between structural and functional networks. A high Jaccard index indicates a large agreement between structural and functional networks. Conversely, a low Jaccard index indicates a small agreement between structural and functional networks. The non-responsive group had a significantly higher Jaccard compared to the responsive group at rest (p = 0.036, t = 2.386) and during early (adjusted p = 0.42) and mid-seizure (adjusted p = 0.47). Error bars represent standard error of mean. Venn diagrams (bottom) schematically represent the Jaccard index for each group at each epoch at the seizure clip level. Note while we performed statistical tests on the individual and not clip level, all seizure clips are plotted here.