Abstracts

Rationale: Responsive neurostimulation (RNS) for treatment-resistant epilepsy applies therapeutic stimulation in response to ongoing seizures or aberrant electrical activity detected from intracranial electrodes. Prior studies have found that RNS patients with highly successful outcomes undergo functional network changes whereas poor responders do not. We hypothesize that the timing of stimulation, specifically stimulation during non-epileptic brain states, is critical in driving long-term changes to restore healthy brain networks. This work aims to quantify how stimulation timing relative to low- vs. high-risk periods—i.e., stimulation occurring during or outside of periods of repetitive detections referred to as long episodes— correlates to RNS outcomes.

Methods: This study aims to quantify the number of stimulation episodes and time spent in low-risk epochs in a cohort of 39 patients at the University of Utah treated with the RNS System. All clinical outcomes and monthly seizure frequency rates were taken by board-certified epileptologists using electronic medical records from follow-up visits. Patients were categorized into three groups based on their seizure reduction values: super responders (>90% reduction, n=10), intermediate responders (≥ 50% reduction and ≤ 90% reduction), n=17, and poor responders (< 50% reduction, n=12). Low-risk periods were defined as days which had fewer long episode counts than the monthly average using stimulation episode histogram data extracted from the RNS device.

Results: Super-responder patients spent more time in low-risk periods since stimulation onset (Pearson’s correlation: p< 0.001), and seizure reduction correlated with more frequent stimulation during low-risk periods compared with high-risk periods (Pearson’s correlation: p< 0.01). RNS therapy statistically reduced seizure rates with a median reduction of 67% (one sample t-test: p< 0.0001) after a median follow-up time of 819 days. Baseline seizure frequency, epilepsy duration, and age of the patient was not a statistical predictor of outcome (Wilcoxon signed-rank test: p=0.110; Pearson’s correlation: p=0.547, p=0.414).We found that for both super-responder and poor/intermediate responder cohorts, daily stimulation significantly increased over time (Pearson’s correlation: p< 0.0001, p< 0.0001), though daily stimulation did not correlate with seizure reduction for the first 90 days of therapy or since stimulation started (Wilcoxon signed-rank test: p=0.262, p=0.0899).