EEG in the critical care setting

Abstract

Continuous EEG (cEEG) is increasingly used in the intensive care unit (ICU), but implementation and interpretation remain variable. To better understand its clinical impact, this review evaluates its role in diagnosis, prognostication, and treatment. To do so, we conducted a narrative review of observational studies, randomized controlled trials, consensus recommendations, and illustrative cases across adult and pediatric critical care populations. In these studies, nonconvulsive seizures occur in ∼10–20 % of ICU patients undergoing cEEG, most without clinical signs, and risk is highest in coma, pediatrics, with prior clinical seizures (recent or remote), and acute structural brain injury. Seizure burden > 12 min/hour (20 % burden) qualifies as status epilepticus (SE). Furthermore, higher seizure burden consistently predicts poor outcomes, including hippocampal injury, functional and cognitive impairment, epilepsy, and increased mortality. Additionally, rapid-response EEG and AI-based tools improve diagnostic efficiency, expand access, and may improve outcomes and length of stay. Quantitative EEG can reliably detect other acute brain events, such as delayed cerebral ischemia after subarachnoid hemorrhage, usually sooner than any other practical method. In summary, cEEG has transformed seizure detection and outcome prediction in the ICU, though gaps persist in access, workforce capacity, interpretation, patient selection, and clinical implementation. Broader standardization, expansion of point-of-care rapid EEG devices, and AI integration are critical to scaling cEEG and improving patient outcomes.