Abstracts

Rationale: The incidence of Sudden Unexpected Death In Epilepsy (SUDEP) in refractory epilepsy is 9.3/1000 patient-years (1). Post-ictal generalized EEG suppression (PGES) occurs after 40-66% of bilateral convulsive seizures in adults and is thought to be a potential biomarker of SUDEP risk (2). PGES is defined as the absence of electroencephalographic activity >10uV for longer than 1 second (2). PGES is more common in those with bilateral tonic clonic (BTC) seizures. We studied Electrocorticographic (EcoG) suppression in patients with the Responsive Neurostimulation (RNS) System. The RNS System long-term ECoG data gave us the opportunity to correlate the presence of postictal ECoG suppression (PES) records with electro-clinical features of the patient’s known seizure types.

Methods: All adult patients treated with RNS System at UC Davis Medical Center for at least 3 months were included. A retrospective chart review was performed to obtain clinical seizure characteristics. All ECoG records containing electrographic seizures stored since RNS placement were reviewed by a board-certified epileptologist (JK). Patients for whom electrographic seizures demonstrated PES in at least one ECoG record were identified and characterized by whether ECoG suppression was bilateral or regional/unilateral (for patients with unilateral lead placement).

Results: 15 patients received the RNS Systems and were followed for an average of 22 months (range: 5 - 36). A review of ECoG records showed post ictal ECoG suppression in one or more leads in 47% (7/15) of patients. All 7 patients had at least one temporal lead; lead placements for all 15 patients are shown in table 1. 86% (6/7) patients had ECoG records showing bilateral PES. 2 of the 6 patients with bilateral PES also had ECoG records showing focal ECoG suppression in at least one recording channel. 71% (5/7) of the patients with PES had a history of focal to bilateral tonic_x0002_clonic seizures. The other 2 (2/7) patients reported a history of focal impaired awareness seizures and one patient reported “drop attacks” which may have been BTC seizure semiology but had the EMU evaluations at an outside center. Interestingly, seizure types in 8/15 patients without suppression were predominantly focal seizures with rare or undocumented focal to bilateral tonic-clonic seizures, with 2 patients reporting no history of BTC events.

Conclusions: To our knowledge, this is the first review to assess the correlation of seizure semiology with post ical ECoG suppression in patients treated with the RNS System. Our findings suggest that patients with post ictal ECoG suppression tend to have BTC seizures (5/7) as compared to those without (8/8) and is concordant with prior studies (2). Long-term ECoG data may enable us to assess PES over time and help with SUDEP risk forecasting by identifying the frequency and duration of post ictal suppression. Further studies are needed to fully assess the impact that long-term ECoG data may have on SUDEP risk prediction and eventually reduction.

1) Dashieff, RM. J of Clin Neurophysiology, 1991
2) Kang, et al, Seizure, 2017

Funding: Please list any funding that was received in support of this abstract.: no funding.