Abstracts

Rationale: We present the results of long-term ambulatory electrocorticography (ECoG) from 22 subjects with localization related epilepsy of temporal lobe onset (TLE) who were studied with bilateral hippocampal (Hc) depth electrodes while participating in a trial of the RNS System (NeuroPace) cranially implanted responsive neurostimulator. The intent was to determine the distribution of left and right Hc onsets, as well as the length of time that seizures were detected from one Hc before being detected in the other. Methods: All subjects had been diagnosed with TLE of Hc origin and were participating in an investigational trial of the RNS System at 4 centers. Four-contact depth leads were stereotactically implanted along the longitudinal axis of each Hc. ECoGs were stored by the neurostimulator based on detected electrographic seizures and then uploaded to a secure data management system. Electrographic seizures were defined as episodes of low-voltage fast-activity or rhythmic sharp activity, distinct from background, detected for longer than 25 seconds. The data was collected post-implant, prior to receiving neurostimulation. Medications remained stable during the data collection period. An analysis was performed of the side of ictal onset as well as the longest interval before onsets switched from one Hc to the other (longest interval before switching (LIBS)). Results: Subjects were between 21 and 52 years of age. All had failed at least 5 antiepileptic drugs and 7 had failed VNS therapy. Fourteen (14) subjects had presumed bilateral mesial temporal onsets from either scalp or intracranial monitoring. Eight (8) subjects had unilateral mesial temporal onsets on scalp or intracranial EEG but were not candidates for temporal lobectomy because there was bilateral Hc atrophy (2), onsets contralateral to the Hc atrophy (1), failed WADA testing (2) or onsets from the dominant temporal lobe with PET abnormality from the contralateral side (1) or failed dominant lobe resection (2). Temporal lobe lateralization, MRI results, length ECoG data collection, number and lateralization of Hc seizure onsets recorded by the RNS System and number of days before onsets switched from one Hc to the other (LIBS) are presented in the table. Conclusions: Twenty-two (22) subjects with TLE of Hc origin underwent long-term intracranial ambulatory ECoG monitoring while participating in the RNS System investigational trial. Three (3) subjects with presumed left lateralization had bilateral onsets. In the 15 subjects with bilateral independent Hc onsets, seizures were lateralized solely to one Hc for an average of 16 days (4 - 39 days) before onsets were recorded in the contralateral Hc. This suggests that many days of EEG monitoring with scalp or intracranial electrodes may be required in order to confidently lateralize epilepsy of Hc origin.