Journal Article

Longitudinal characterization of electroencephalography features in consciousness recovery following severe traumatic brain injury: a case series study in male patients

De Koninck, Beatrice P., Apinis-Deshaies, Amélie, Maschke, Charlotte, GirardPepin, Rosalie, Blain-Moraes, Stefanie, De Beaumont, Louis, Bernard, Francis, Williams, Virginie, Arbour, Caroline, Williamson, David, Duclos, Catherine, Briand, Marie-Michèle, Brazeau, Daphnee, Martens, Géraldine

2025

Overview

Crucial clinical decisions in the acute-to-subacute stages of severe traumatic brain injury (sTBI) are mostly based on neurological exams and behavioral assessments. Although electroencephalography (EEG)-derived indices of consciousness show prognostic potential, their effectiveness in tracking individual recovery over time remains unclear. This study characterizes the longitudinal recovery of consciousness following sTBI, tracking EEG spectral power, and aperiodic exponent markers, considering sex, etiology, and age. Four medically stable, non-sedated sTBI patients were recruited 7–26 days post-injury, based on etiology (sTBI), age (M=30 years, SD=1.41 years), lesion severity (diffuse axonal injury). Behavioral responsiveness was assessed daily using the Coma Recovery Scale-Revised (CRS-R), alongside 5-min resting-state EEG for 6 days, with an additional recording a week later. Changes in power spectral distribution and the aperiodic component were observed over time, even within similar diagnostic categories. The aperiodic component exhibited a similar improvement trajectory to behavioral responsiveness, with a progressive flattening of the EEG slope in two individuals who recovered consciousness. In contrast, individuals whose CRS-R category remained static showed inconsistent fluctuations in the aperiodic component over time. Improvements in CRS-R scores were accompanied by changes in absolute power for theta, alpha, and beta frequency bands. However, the “ABCD” and Maximum Frequency Peak frameworks showed limitations and inconsistencies when compared to behavioral outcomes. This longitudinal within-subject design captured neurophysiological changes along patient-specific recovery trajectories, revealing substantial fluctuations within individual EEG markers, despite controlling for typical confounds like etiology, age, lesion severity, and sex. Combining spectral power and the aperiodic exponent may support the development of more dynamic and reliable markers to track changes in brain activity associated with consciousness recovery, potentially improving diagnostic accuracy, outcome prediction, and therapeutic interventions. Ethical approval for this study has been given by the Research Ethics Board of the CIUSSS du Nord-de-l’Île-de-Montréal (Project ID 2021-2279).

• Publisher: Oxford University Press
• Subject: acute–subacute care / Clinical Psychology / disorders of consciousness / EEG / Experimental and Cognitive Psychology / Human health sciences / Neurologie / Neurology / Neurology (clinical) / Psychiatry and Mental Health / Sciences de la santé humaine / severe traumatic brain injury