Electroencephalography (EEG) is an essential tool in the monitoring and management of hospitalized patients with encephalopathy [1]. While generalized seizures are usually obvious on history or physical exam, the occurrence of non-convulsive status epilepticus (NCSE) has been increasingly recognized as a cause of encephalopathy. With NCSE, the clinical features can be subtle, ranging from twitching of an eyelid or extremity, to changes in behavior or even coma.

Rapid diagnosis of NCSE is essential as persistent seizures can become refractory to treatment and contribute to permanent brain injury [2]. The American Clinical Neurophysiology Society recommends continuous EEG monitoring in patients with altered mentation following generalized status epilepticus (SE), or those with unexplained altered mental status [3]. The American Heart Association and the Neurocritical Care Society recommend quickly screening for seizures in patients with cardiac arrest who fail to recover consciousness [4]. However, EEG monitoring is often not available at all hours in resource- and personnel-limited community hospitals. Formal EEG testing typically requires a specially-trained technician and placement of 25 or more recording probes, followed by close monitoring at the bedside using a dedicated workstation. This requires constant staffing and investment, and may not be feasible for many hospitals, especially to be able to provide around-the-clock coverage. A recent cross-sectional study identified that only 27.3 % of hospitals were EEG capable, and only 22.5 % had continuous EEG (cEEG) capability [5]. Transfer to EEG-capable centers has been the typical practice, however this has been increasing difficult in the post-COVID era and can be associated with increased costs and delayed care.

The Ceribell® (Sunnyvale, CA) point-of-care rapid EEG device is a small, 10‑lead headband with a compact portable monitor, and utilizes artificial intelligence to monitor and identify EEG waveforms for seizure activity (Fig. 1). It has been FDA-approved and available since 2017, and has gained traction in the diagnosis of NCSE. It can be placed and operated by bedside staff with minimal training, be remotely monitored by a neurologist, and kept on for prolonged periods of time. The device utilizes artificial intelligence to display a “seizure burden” value and alarms the treating team when seizures are detected, allowing prompt diagnosis and treatment.

The purpose of this study was to review the implementation of a rapid EEG program in a community hospital, identify its performance when compared with formal EEG testing, and identify the frequency at which treatment decisions were influenced by rapid EEG results.