Correlation of Gray-to-White matter ratio on Non-Contrast CT with neurological outcome in Post-Cardiac arrest syndrome
Zeinab El Mawla, Jida Mulki, Mahmoud Hassoun, Ahmad AoudeIntroduction
Post-cardiac arrest syndrome (PCAS) is a major cause of mortality and neurological disability. Early prognostication is essential for clinical decision-making, but variability in interpretation of Non-Contrast CT (NCCT)) Gray to White Matter Ratio (GWR) measurement techniques limits its clinical standardisation. This study evaluated the association between the Gray-to-White Matter Ratio (GWR) on early non-contrast CT (NCCT) and 6 -month neurological outcomes, and compared manual versus automated GWR measurements.
Methodology
A retrospective cohort of 120 patients who are comatose with PCAS patients with return of spontaneous circulation (ROSC) was analysed. NCCT scans (6–24 hours post-ROSC) were used to calculate GWR across basal ganglia, supratentorial cortex, and cerebellum using both manual and automated methods. Outcomes were classified using the Cerebral Performance Category (CPC). Prognostic performance was evaluated using receiver operating characteristic (ROC) analysis, and multivariable modelling was performed using Firth logistic regression.
Results
Higher GWR values were significantly associated with favourable neurological outcomes across all regions and measurement methods (p<0.001). Automated global GWR demonstrated the highest prognostic accuracy (AUCarea under the curve=0.869; sensitivity 84.2%, specificity 84.1%) and was an independent predictor of poor outcome. Manual measurements showed moderate-to-good inter-rater reliability.
Conclusion
GWR is a robust prognostic biomarker for neuroprognostication after cardiac arrest. Automated GWR measurement improves reproducibility and supports the standardisation of imaging-based prognostic assessment, with potential implications for clinical decision-making in post-cardiac arrest care.