Abstracts

Rationale: Intracranial electroencephalographic (iEEG) monitoring and neurostimulation in patients undergoing epilepsy surgery evaluations have been used extensively to identify epileptic tissue and map functional anatomy of the brain. Behavioral observations during stimulation trials have been used to identify brain regions associated with emotion, e.g., yielding the cingulate cortex as a stimulation target for psychiatric indications. It remains unclear how stimulation modulates neural networks and to what extent these behavioral changes are attributable to changes in epileptiform activity (1,2). This study explores network dynamics in one patient who demonstrated marked changes in anxiety during a therapuetic stimulation trial during pre-surgical iEEG evaluation. References: 1. B. N. Lundstrom et al., “Chronic subthreshold cortical stimulation to treat focal epilepsy,” JAMA Neurol., 2016. 2. K. Starnes et al., “Two cases of beneficial side effects from chronic electrical stimulation for treatment of focal epilepsy,” Brain Stimul., 2019. Methods: The patient was retrospectively identified. Clinical information was extracted from the medical record. iEEG was recorded from electrodes placed in the left hemisphere including the middle and posterior cingulate cortex, precuneus, postcentral gyrus, and middle frontal and supramarginal gyri (margin of prior resection). EEG were parsed into pre-stimulation (hour preceding stimulation onset), early post-stimulation (one hour following stimulation end), and late post-stimulation (20 hours following stimulation end) periods. We calculated spectral power, zero-phase lag coherence, and average variance of coherence (20 minute moving window) for bipolar contact pairs using 10 second segments and frequency bands of 1-3 Hz, 3-7 Hz, 7-13 Hz, and 13-30 Hz for the above periods. EEG from the 19 hour stimulation trial was excluded from analysis. Ratio of delta to beta power was calculated as proxy for sleep/wake state. Results: One day prior to stimulation, patient had 17 clinical seizures. During stimulation of seizure onset zone electrodes in the left supramarginal gyrus, patient had one clinical seizure and decreased inter-ictal discharges. Patient had a history of depression and reported a decrease in anxiety and improved sleep quality with stimulation. Over the day following cessation of stimulation, inter-ictal discharges, anxiety, and tearfulness increased. In the early post-stimulation period, there was a significant increase in average coherence (p < 0.01 for all bands tested) for contact pairs between the middle cingulate, posterior cingulate, precuneus, middle frontal gyrus, and postcentral gyrus as compared with the pre-stimulation period. Variance of coherence increased significantly (p < 0.01, all bands) as well. Coherence was significantly greater (p < 0.01, all bands) in the late post-stimulation period than in the early post-stimulation period. This was true for coherence between electrodes that had frequent epileptiform activity and between electrodes that did not. Delta-beta ratio did not differ between these periods suggesting similar sleep/wake state in all conditions. Conclusions: During a trial of therapeutic electrical brain stimulation for this patient, we observed post-stimulation changes in spectral coherence between cortical electrodes. Results suggest that stimulation decreased patient anxiety and led to increased coherence and decreased epileptiform activity observable in the EEG after cessation of stimulation. It remains unclear to what extent these changes are related. Increased spectral coherence and decreased spike rates have been demonstrated during cortical stimulation (3). Our overall finding suggests that changes in network dynamics can be identified following cessation of stimulation and suggest the utility of prospectively tracking behavioral states during and after brain stimulation. References: 3. K. Westin et al., “Neurophysiological effects of continuous cortical stimulation in epilepsy – Spike and spontaneous ECoG activity,” Clin Neurophysiol., 2019. Funding: This work was supported in part by the National Institutes of Health (UH2/UH3 - NS95495).