Abstracts

Rationale: Responsive neurostimulation (RNS) has been described as a safe and effective treatment for seizures in patients with mesial temporal lobe epilepsy (MTLE). For patients with mesial temporal sclerosis (MTS) that have been treated with RNS, seizure reduction ranged from 30.5 to 89.2% (Geller EB et al, 2017). The patients in this cohort had variable demographics including duration of epilepsy, frequency of seizures, unilateral and bilateral seizure onsets, and treatment history. We present two MTS patients with similar epilepsy histories yet significantly disparate response in seizure control from RNS. Methods: Retrospective case and electrocorticography reviews were conducted for two patients at our institution. Results: We compare two female patients, each with over thirty years of focal impaired awareness seizures. Continuous video EEG for both revealed left temporal focal slowing, left anterior temporal sharp waves, and left temporal electrographic seizures. MRI brain also revealed left MTS in both patients. Each patient was offered a standard left temporal lobectomy, but each declined resection due to potential compromise of memory or musical processing. Both patients underwent surgery for placement of a RNS system, which was titrated over 24 months. The first patient has one sub-temporal strip and one temporal depth placed in the left hippocampus. Her seizure frequency remains unchanged with mild reduction in her subjective seizure severity. The second patient has two sub-temporal strips placed and is currently seizure-free. Interestingly, the patient with the longer epilepsy history and higher frequency of seizures, as well as more trialed medications, has the better seizure response to RNS. Conclusions: RNS is a potential therapeutic option for patients with medically refractory MTS who are not resective surgery candidates. RNS efficacy can vary greatly, and predicting RNS responsiveness based on patients' clinical features and imaging results remains a challenging process. Additional studies are needed to better predict RNS outcomes. Funding: No funding