Abstracts

Rationale: Intracranial brain stimulation is used for pre-surgical functional mapping, localization of epileptogenic tissue, and treatment of drug resistant epilepsy via chronic implanted devices. Cortico-cortical evoked potentials (CCEPs) measure the brain's response to a single stimulation pulse, and serve as a foundational unit for understanding perturbations to neural networks._x000D_  _x000D_ Here, we test the hypotheses that (1) cortico-cortical evoked potentials (CCEPs) demonstrate systematic variability across trials and (2) CCEP amplitudes depend on the timing of stimulation with respect to endogenous low-frequency electrophysiologic oscillations. In this study, we quantify trial-by-trial variability in CCEPs that would be removed by standard approaches relying on average waveforms, and we aim to utilize plasticity in stimulation response as a marker of epileptogenic tissue.

Methods: We studied 11 patients who underwent CCEP mapping while admitted to the epilepsy monitoring unit at the Hospital of the University of Pennsylvania after intracranial depth electrode implantation for surgical evaluation of drug-resistant epilepsy. Evoked potentials were measured from all electrodes after each pulse of a 30 second, 1 Hz bipolar stimulation train applied to adjacent electrode contacts. We computed the Spearman rank correlation between trial index and N1 (15-50 ms post-stimulation; Figure 1A) and N2 (50-300 ms post-stimulation; Figure 1B) amplitudes to quantify monotonic trends (Figure 1C-D). To identify low frequency oscillations, we used the fooof package to identify peaks of power spectral density exceeding 1/f background in each recording electrode (Figure 2C). We computed the circular-linear correlation between estimated signal phase in significant peaks from 3 to 14 Hz and N1 or N2 amplitude as a measure of intermittent phase synchronization of stimulation response.

Results: We found that evoked N1 and N2 waveforms exhibited both positive and negative monotonic trends in amplitude over the course of a 30 second, 1 Hz stimulation train (Figure 2A). N1 and N2 amplitude tended to be lower in electrodes belonging to the clinically determined seizure onset zone (SOZ). Monotonic trends were more common when stimulating or recording from the SOZ, an effect that appeared to be driven by SOZ recording electrodes (Figure 2B). We did not find a statistically significant relationship between N1 or N2 amplitude and pre-stimulation phase of low frequency oscillations in the hippocampus (Figure 2D), anterior cingulate cortex, or recording electrodes.

Conclusions: These findings suggest that standard approaches for CCEPs mapping, which involve computing a trial-averaged response over a 0.2 to 1 Hz stimulation train, may be discounting monotonic trends in amplitude that appear to localize to epileptogenic tissue. We did not find evidence for intermittent synchronization of CCEP amplitudes with ongoing low-frequency oscillations. Further targeted experiments are needed to determine whether phase-locked stimulation could have a role in localizing epileptogenic tissue.

Funding: R01NS116504 (KAD), 1K23NS121401-01A1, Burroughs Wellcome Fund Career Award for Medical Scientists (EC)