Seizure Identification on the Intensive Care Unit (ICU)

The aim of this project is to assess the ability of different groups of National Heath Service (NHS) professionals to correctly identify clinical seizures, and distinguish them from other movements commonly seen in the ICU environment, when shown digital video recordings only.

Patients on the ICU are at risk of having seizures, however also commonly make other movements, including shivering, jerking, tics and tremors. An Electroencephalogram (EEG) records the brain wave activity and can help distinguish epileptic seizures from other movements. In a study by Bendadis et al (2010), 52 video-EEGs were reviewed containing "possible seizures" on the ICU. They found only 27% recorded actual epileptic events, with the other 73% having a range of other movements. Malone et al (2009) studied accuracy of diagnosis of 20 video recordings of clinical episodes on the neonatal unit, comparing different staff groups. They found no significant difference between Doctors and Nurses in correctly identifying seizures, however found that accuracy of diagnosis was generally poor.

Clinical scientists are currently expanding their roles and responsibilities across Neurophysiology, including giving consultant-level advice on EEG investigations. EEG recordings on the ICU are often obscured by excessive, unavoidable electrical/movement artefacts caused by equipment such ventilators and pumps, and patient factors such as position, breathing artefact and suctioning. These make the EEG difficult to interpret (Boggs 2021). Assessing the clinical signs and symptoms which we may see in ICU patients, in the absence of interpretable EEG, is an essential skill.

This study aims to assess Clinical Scientists skills at clinical interpretation, in comparison with other staff groups in the ICU setting. Staff will be asked to watch video clips of events captured in the ICU, and tell us whether they think they are seizures or not, and explain their thought process behind the decision.