Validation of the rCAST score and comparison to the PCAC and FOUR scores for prognostication after out-of-hospital cardiac arrest

Abstract

AimEarly, accurate outcome prediction after out-of-hospital cardiac arrest (OHCA) is critical for clinical decision-making and resource allocation. We sought to validate the revised post-Cardiac Arrest Syndrome for Therapeutic hypothermia (rCAST) score in a United States cohort and compare its prognostic performance to the Pittsburgh Cardiac Arrest Category (PCAC) and Full Outline of UnResponsiveness (FOUR) scores. MethodsThis is a single-center, retrospective study of OHCA patients admitted between January 2014-August 2022. Area under the receiver operating curve (AUC) was computed for each score for predicting poor neurologic outcome at discharge and in-hospital mortality. We compared the scores’ predictive abilities via Delong’s test. ResultsOf 505 OHCA patients with all scores available, the medians [IQR] for rCAST, PCAC, and FOUR scores were 9.5 [6.0, 11.5], 4 [3, 4], and 2 [0, 5], respectively. The AUC [95% confidence interval] of the rCAST, PCAC, and FOUR scores for predicting poor neurologic outcome were 0.815 [0.763–0.867], 0.753 [0.697–0.809], and 0.841 [0.796–0.886], respectively. The AUC [95% confidence interval] of the rCAST, PCAC, and FOUR scores for predicting mortality were 0.799 [0.751–0.847], 0.723 [0.673–0.773], and 0.813 [0.770–0.855], respectively. The rCAST score was superior to the PCAC score for predicting mortality (p = 0.017). The FOUR score was superior to the PCAC score for predicting poor neurological outcome (p < 0.001) and mortality (p < 0.001). ConclusionThe rCAST score can reliably predict poor outcome in a United States cohort of OHCA patients regardless of TTM status and outperforms the PCAC score.