Abstracts

Rationale: Corticocortical evoked potentials (CCEPs) can probe network connectivity and indicate input and output hubs. The goal of this study is to determine the impact of the seizure onset zone (SOZ) on surrounding as well as distant brain networks using CCEPs to investigate the spatial extent, network involvement, and magnitude of input to as well as output responses from this region. The ability to better map functional interconnectivity of the seizure focus with other brain regions, including individualized asymmetries, can help guide surgical resection plans or ablative treatments as well as targeting by neuromodulation protocols. Methods: Bipolar cathodic 1 Hz electrical stimulation for 50s epochs was applied to healthy tissue; epileptic tissue within the SOZ; and tissue near the SOZ in 19 patients at the Beijing Tiantan Hospital with refractory epilepsy with depth and grid intracranial electrodes consisting of approximately 100 contacts during Phase II EEG monitoring.  Evoked response N1 and N2 amplitudes as well as latencies were averaged across trials from all contacts.  Network activation based on epileptogenicity and brain location, including investigation of grey vs. white matter stimulation, was determined.  Results: N1 mean +- SEM amplitudes are reported with units of µV. Comparison of anatomic differences in brain tissue/network recruitment by stimulation of non-epileptic brain regions while recording evoked responses in all other contacts revealed that locations within the mesial temporal lobe and white matter activate the largest responses with the broadest reach (88 +- 1.8 and 107 +- 7.6, respectively). Comparison of location-matched electrical stimulation of epileptic versus non-epileptic tissue showed larger local as well as distant responses with stimulation of epileptic tissue. For the hippocampus, local evoked responses were 820 +- 4 when generated from epileptic compared to 464 +- 2.8 when from non-epileptic tissue. Distant evoked responses were 106 +- 2.4, compared to 88 +- 1.8 (p<0.01 for all comparisons). Stimulating the seizure onset zone or nearby tissue broadly affects distant CCEP amplitude in a graded manner. Evoked response amplitude averaged across all contacts is significantly larger with stimulation of epileptic tissue (120 +- 9.7) than of non-epileptic tissue (42 +- 3.4) using two-tailed Student's t-test (p<0.01). Epileptic tissue also responds to stimulation of non-epileptic tissue with larger amplitude potentials than surrounding healthy tissue. Responses in the SOZ are, on average 1 SD above normalized baseline of evoked response amplitude across contacts. In patients with multiple SOZ, responses in non-stimulated SOZ to stimulation in another SOZ are 3.4 SD larger than normalized baseline of evoked response amplitude across contacts.  Conclusions: CCEPs are useful in identifying a seizure focus, but results need to be related to the intrinsic excitability of that part of the brain. CCEPs generate larger local and remote potential when an epileptic focus is stimulated, compared to anatomically-matched non-epileptic tissue. CCEP responses recorded in the SOZ from stimulation outside the SOZ may provide clues regarding the most excitable SOZ. Hence, CCEP interrogation during Phase II trials may be helpful for guiding resection surgery or targeting neuromodulation to epileptic foci and related networks. Funding: No funding