Abstracts

Rationale: In presurgical evaluation of patients with drug resistant epilepsy, stereo-electroencephalography (SEEG) can elucidate seizure onset zones (SOZs) and propagative zones (PZs), important predictors of surgical success. Nondirected and directed functional connectivity (FC) analyses can distinguish SOZs via brief interictal recordings (Goodale et al, 2020). Spatial relationships of recording contacts may lead to increased coherence of signal and should be taken into consideration (Wang et al, 2020). Our analysis evaluated how the relationship of FC by distance from recorded brain regions could distinguish SOZs, PZs and non-involved (Non) regions.

Methods: Eighty-one patients with medically refractory epilepsy who underwent resting state SEEG recordings at Vanderbilt University Medical Center were included in this retrospective analysis. Brain regions were labeled a priori as SOZs, PZs or Non by neurologists’ interpretation of ictal recordings. Z-scored nondirected FC (Imaginary Coherence) and directed FC (Partial Directed Coherence) were calculated in SOZs, PZs, and Non regions and evaluated over Euclidean distance between SEEG recording contacts. Post-operative Engel outcomes at most recent follow up were determined from review of clinical notes.

Results: For both nondirected as well as inward and outward directed connectivity, FC is higher for shorter Euclidean edge distances and tapers exponentially as distance increases, with differences between SOZs, PZs and non-involved regions being significant at local distances < 20 mm and decreased at distances >20 mm (Figures 1A-C). Notably, the difference of inward and outward connectivity is stable across distance with SOZ FC being strongest, PZ FC being intermediate, and Non FC being lowest (Figure 1D). This relationship remains apparent in Engel I patients (Figure 2A), but in Engel II-IV patients, PZs more closely represent FC patterns of Non regions (Figure 2B).

Conclusions: Coherence of FC measures is higher in recordings of proximal contacts and diminished over distance, as expected; however, differences in inward and outward connectivity are stable over distance and electrophysiologically distinct in SOZs, PZs and Non regions. The PZ connectivity difference observed for Engel II-IV outcome could suggest that there is a fundamental difference between the PZ characteristics in surgical responders and non-responders, or this difference could be an indication that SEEG sampling of the epileptic network is insufficient for epileptic network characterization. Overall, this distance-invariant electrophysiologic pattern may be useful in identifying SOZs and PZs via brief interictal recordings to target during surgery. Finally, future work aimed at understanding PZ-specific directed connectivity over distance for surgical responders and non-responders could improve preoperative surgical outcome prognostication.
_x000D_ References:_x000D_ Goodale et al. Resting-Stage SEEG may localize epileptogenic brain regions. Neurosurgery. 2020;86(6):792-801. doi: 10.1093/neuros/nyz351. _x000D_ _x000D_ Wang et al. Interictal intracranial electroencephalography for predicting surgical success: the importance of space and time. Epilepsia. 2020;61(7):1417-1426. doi: 10.1111/epi.16580._x000D_
Funding: None