Electroencephalography in the Intensive Care Unit: Experience of the Department of Functional Neurosensory Explorations at Gonesse Hospital, France.
Idrissa Doumbouya, Karinka Diawara, Elian Hapca, Djénabou Négué Barry, Mohamed Traoré, Gabriela Carelli, Oumar Sylla, Andreea Stanciulescu, Fodé Abass Cissé
Abstract
Open AccessINTRODUCTION: The aim of this study was to evaluate the indications and outcomes of electroencephalography (EEG) in the intensive care unit (ICU), as well as to assess their prognostic values in adults. MATERIALS AND METHODS: We conducted a two-year retrospective study from November 2022 to October 2024 in the Department of Functional Neurosensory Explorations at Gonesse Hospital, involving patients hospitalized in the ICU. Sociodemographic, clinical, electroencephalographic, and prognostic variables were collected. During data analysis, we investigated EEG patterns associated with poor prognosis and epileptic abnormalities. RESULTS: A total of 76 participants met our selection criteria, with a mean age of 61 ± 18.8 years and a predominance of men (sex ratio: 1.92). The main indications for EEG were neuroprognostic evaluation in 53 patients (68.5%), seizure assessment in 32 patients (40.8%), and unexplained neurological deterioration in four patients (5.3%). The mean duration of EEG recordings was 50 minutes; 17 patients (22.4%) experienced clinical seizures. Epileptic abnormalities were observed in 43 patients (56.6%), including electrographic seizures in 18 patients (23.7%), non-specific interictal activity in 10 patients (13.2%), non-convulsive status epilepticus (NCSE) in eight patients (10.5%), and lateralized periodic discharges (LPDs) in seven patients (9.2%). These findings accounted for 41.9%, 23.3%, 18.6%, and 16.3% of the epileptic abnormalities, respectively. Regarding non-epileptic abnormalities, 63 patients (82.9%) exhibited generalized slow waves, followed by focal slow waves in 27 patients (35.5%), generalized periodic discharges in 15 patients (19.7%), and burst-suppression patterns in 11 patients (14.5%). There was a statistically significant correlation between epileptic abnormalities and status epilepticus (p = 0.0003), ischemic stroke (p = 0.0317), and abnormal brain imaging (p = 0.0421). Poor prognostic factors included the absence of EEG reactivity (p = 0.0001), burst suppression (p = 0.0002), and the absence of cerebral activity (p = 0.0163). CONCLUSION: EEG in the ICU setting not only provides valuable information regarding the epileptic origin of coma but also offers important prognostic insights.