Abstracts

Rationale: Responsive neural stimulation (RNS® System, NeuroPace™) is a safe and effective therapy to reduce medically refractory focal onset seizures, with stimulation targeted to the epileptogenic region. Deep brain stimulation (DBS) of the anterior nucleus of the thalamus has also been proven effective for reducing focal onset seizures, potentially more effective in limbic epilepsy. We therefore combined the approaches in a patient for whom a characterized epileptogenic lesion involved more regions than could be adequately targeted with the current 2-lead NeuroPace system. We present examples of refractory seizures recorded in the region of a chronic lesion involving perisylvian/opercular, posterior frontal, and insular cortices. Recurrent seizures over 30 years began with tingling and numbness, either on the right or throughout the entire body, accompanied by anxiety and impairment of ability to speak with variable alteration of awareness. Episodes at times have been associated with jerking in his right arm and leg for approximately a minute, followed by impaired function for 15-30 minutes. Some episodes have consisted of isolated jerking of his right upper extremity, more common at night. A biopsy prior to presentation was nondiagnostic, and resulted in hemorrhage and transient right hemiparesis. Methods: The patient underwent a standard presurgical evaluation, including anatomical and functional brain MRI, phase I video EEG recording, vascular imaging, neuropsychological testing, and a phase II evaluation using 8 depth electrodes with multiple perisylvian and insular approaches, a left hippocampal electrode, and an electrode along the frontal prior biopsy tract. Because of the concern that two 4-contact RNS System leads would not adequately deliver stimulation to all potentially epileptogenic regions, the depth leads were placed stereotactically in the most active region, left opercular parietal cortex adjacent to the lesion and also in a trajectory aiming to traverse the anterior nucleus of the thalamus as well as the centromedian nucleus. Three perilesional back up depth leads were also placed but not connected to the neurostimulator. Results: Electrographic seizures as detected by the RNS neocortical depth lead correlated with changes in recordings in the thalamic lead, although not consistently. When ictal patterns were observed on the thalamic lead, they followed the neocortical depth lead onset pattern by several seconds, and ictal patterns could resolve sooner. Conclusions: The value of responsive brain stimulation with combined neocortical and thalamic depth leads requires further investigation. However, we demonstrate that seizure detection in thalamic leads is sometimes possible, although thalamic detection lags neocortical detection in this case. DBS has only recently been FDA approved; future incorporation of seizure detection technology into those systems might be helpful. Funding: None