Resident Physician Icahn School of Medicine at Mount Sinai New York, New York
Rationale: Electroencephalography (EEG) provides crucial information in the management of patients with suspected non-convulsive status epilepticus (NCSE). However, availability of long-term EEG monitoring (LTM) and interpretation infrastructure is limited, particularly at community hospitals, often necessitating transfer of patients to quaternary medical centers. Novel portable EEG systems may help expand EEG access to community hospitals, thereby enabling initial management of NCSE on-site and appropriate transfer for LTM. Here, we describe the clinical impact of deploying Rapid Response EEG System (Rapid-EEG; Ceribell Inc., Mountain View, CA) at a community hospital with remote interpretation by specialists based at an academic medical center. Methods: We retrospectively identified a cohort of patients who underwent Rapid-EEG monitoring at our community hospital as part of routine care. We abstracted patients’ medical records to describe detection and treatment of seizures or status epilepticus, disposition and resource utilization, duration of EEG monitoring, hospital length of stay, and transfer to affiliated flagship hospital for conventional EEG monitoring. Results: Seventy-five patients underwent Rapid-EEG monitoring, of whom 34 (45%) presented with a clinical event concerning for a seizure and 39 (52%) were intubated. EEG monitoring occurred in the intensive care unit (ICU; 69%) and emergency department (ED; 24%); three patients’ EEG monitoring started in the ED and continued during admission to the ICU. Rapid-EEG revealed seizures (n=7, 9%), highly epileptiform patterns (HEP; n=8, 11%), burst suppression (n=3, 4%) and slow/normal activity (n=56, 75%). Twenty-five studies (33%), including five that detected seizure, were performed after-hours (overnight or weekend) when routine EEG would ordinarily be unavailable. Among the seven patients with seizures detected on Rapid-EEG, detection resulted in early treatment with antiseizure medications in 4 (57%), avoided overtreatment in 4 (57%), and confirmed seizure cessation in 6 (86%). Mean duration of EEG monitoring was 1.5±1.0 days (2.6±1.5 days for seizure cases; 2.1±1.5 days for HEP cases; 1.3±0.7 days for slow/normal cases), and mean hospital length of stay was 10.0±9.9 days (9.1±5.6 days for seizure cases; 15.4±4.8 days for HEP cases; 9.4±10.6 days for slow/normal cases). Transfer occurred in four cases (5%), of which Rapid-EEG revealed seizure in one case, HEP in one case, and slow/normal activity in two cases. Rapid-EEG findings mitigated the need for transfer in 23 cases (31%), including five of the seven cases with seizures on Rapid-EEG. Conclusions: Rapid-EEG increased access to EEG for patients in a community hospital setting. This resulted in earlier detection and treatment of seizures, even after-hours, and often mitigated the need for inter-hospital transfer for EEG monitoring. Portable EEG systems may optimize NCSE treatment by supporting a hub-and-spoke model (analogous to existing acute stroke treatment infrastructures), with increased EEG capacity at community hospitals and tele-EEG interpretation by specialists at academic hospitals that can accept transfers for LTM. Funding: Please list any funding that was received in support of this abstract.: No study funding was received. Dr. Gururangan serves as a consultant for Ceribell, Inc., who is commercializing the Rapid Response EEG System but did not provide any study funding. Drs. Madill and Krishnamohan have no relevant disclosures.