Modeling of Intensive Care Risk Factors for Spreading Depolarizations in Severe Traumatic Brain Injury

Area of Science:

• Neuroscience
• Critical Care Medicine
• Biomedical Engineering

Background:

• Spreading depolarizations (SDs) are a critical secondary injury mechanism in severe brain trauma, impacting patient outcomes.
• SDs are associated with metabolic failure and are influenced by routinely managed neurocritical care variables like blood pressure and oxygenation.

Purpose of the Study:

• To analyze risk factors for SD occurrence in patients with severe brain trauma.
• To develop and validate predictive models for SD occurrence using routinely monitored physiological data.

Main Methods:

• Retrospective analysis of electrocorticographic (ECoG) monitoring data from 137 patients with severe brain trauma.
• Alignment of SD timestamps with hourly nursing chart data, including mean arterial pressure (MAP), intracranial pressure (ICP), temperature, brain tissue oxygenation (PbrO2), and blood gases.
• Development of multivariate regression and Poisson models to predict SD occurrence, incorporating both hourly physiological variables and fixed patient factors.

Main Results:

• Univariate and multivariate analyses identified associations between SDs and lower MAP, heart rate, P_aCO₂, and higher temperature, S_aO₂, and glucose.
• Initial models achieved moderate performance (AUC ~0.72), improving with random-effects variables (AUC = 0.84).
• A final Poisson model incorporating hourly predictors and fixed factors achieved high performance (AUC = 0.916) in predicting hours with SDs.

Conclusions:

• Key systemic physiological variables are significant risk factors for SD occurrence in severe brain trauma.
• Predictive models integrating physiological data show potential for identifying patients at risk of SDs and secondary injury.
• The developed model offers a potential tool for neurocritical care to guide interventions aimed at preventing SDs and improving outcomes.