British Society for Clinical Neurophysiology promote and encourage for the public benefit the science and practice of clinical neurophysiology and related sciences

Abstract Details

A service development project of the detection of NCSE in comatose ITU patients

Nonconvulsive status epilepticus (NCSE) can occur in critically ill patients and contribute to adverse outcomes. The exclusion of NCSE is a common reason for urgent EEG requests in comatose patients. Previous studies have demonstrated detection rates of 8-21% in such patients, depending on recording duration.

We retrospectively reviewed 145 patients who were comatose on Kings College Hospital ITU and underwent an urgent inpatient EEG to identify NCSE or subclinical seizures between October 2018-April 2019. Data were extracted from the referral request, EEG report and supplemented from medical records.
The presence or absence of electric seizures, and further characteristics were documented on EEG reports, assessed by at least two neurophysiologists. Expert review determined the clinical significance of seizures on the conscious level of patients in whom seizures were identified. Inclusion criteria included patients aged over 1 year, who were ventilated/breathing spontaneously, and had at least 20 minutes of continuous EEG recording.



The age of participants ranged between 1–89 years (mean 61) and 90/145 (62%) were male. Only four cases out of 145 were reported to have NCSE. All four cases of NCSE had a history of epilepsy or presented acutely with clinical seizures/ status epilepticus (4/36 = 11% of all such patients). The cases with confirmed NCSE had classic features including eyelid fluttering, facial twitching, rhythmic eye movement and fluctuation of blood pressure.

Other clinical information in the remaining patients: 4/23 patients that had suffered cardiac arrest had hypoxic myoclonus or short electrographic seizures; 3/20 patients with traumatic brain injury had either a possible electrographic seizure, liability to seizures or alpha coma; and 2/27 patients that suffered a stroke had either an electrographic or clinical motor seizure. 20 patients had ‘another’ diagnosis including metabolic and infective causes, and only one showed liability to seizures. None of 19 neurosurgical patients that were included had NCSE or seizures. One case was initially diagnosed with NCSE, but this was subsequently revised to CJD.

NCSE is a diagnostic challenge in critically ill patients and should not be overlooked. Patients with known epilepsy and those admitted with seizures should be considered a high-risk group for NCSE. The yield of EEG may have been improved by continuous monitoring, but in the setting of stroke, trauma or neurosurgery patients, the yield was very low. Failure to prioritise has cost implications and limits access to the patients who would benefit most.

TitleForenamesSurnameInstitutionLead AuthorPresenter
DrShahrzadHadaviKings College Hospital
DrRobertElwesKings College Hospital
No references