Early Prognostication After Out-of-Hospital Cardiac Arrest: Modified rCAST Score Incorporating Age and Brainstem Reflexes

TL;DR

This study improves a scoring system to predict poor neurological outcomes after cardiac arrest by adding age and brainstem reflexes, enhancing early prognostication.

Contribution

The modified rCAST (mCAST) score integrates age and brainstem reflexes to enhance prognostic accuracy for OHCA survivors.

Findings

Age >65 years and absence of brainstem reflexes were independent predictors of poor neurological outcomes.

The mCAST score showed higher prognostic accuracy (AUC 0.849) compared to the original rCAST (AUC 0.823).

The mCAST improved risk stratification in high-severity groups without reducing clinical feasibility.

Abstract

Background: Out-of-hospital cardiac arrest (OHCA) survivors demonstrate wide variation in neurological outcomes due to hypoxic–ischemic brain injury. Early prognostic stratification in the emergency department is essential to inform clinical decisions. This study aimed to improve the revised Cardiac Arrest Syndrome for Therapeutic hypothermia (rCAST) score by incorporating additional clinical variables and to evaluate its ability to predict poor neurological outcomes. Methods: This multicenter observational study analyzed OHCA survivors treated with targeted temperature management (TTM) between October 2015 and December 2018 at 22 university-affiliated hospitals participating in the Korean Hypothermia Network prospective registry. The primary outcome was poor neurological status at one month, defined as a Cerebral Performance Category (CPC) score of 3–5. Independent predictors were identified using multivariable logistic regression and incorporated into a modified rCAST (mCAST) score. Results: Among 881 included patients, age > 65 years (odds ratio [OR], 13.87; 95% confidence interval [CI], 7.38–26.08) and absence of brainstem reflexes (OR, 2.31; 95% CI, 1.29–4.12) were identified as independent predictors and added to the mCAST score. The mCAST demonstrated higher prognostic accuracy than the original rCAST (area under the curve [AUC], 0.849 vs. 0.823; p < 0.001). In the high-severity group, the mCAST identified a higher poor outcome rate (95.1% vs. 90.9%) while reducing the proportion of patients in this group (20.7% vs. 31.3%). Conclusions: The mCAST score improves early prognostic accuracy during the immediate post-cardiac arrest period by incorporating age and brainstem reflexes and may offer refined risk stratification without compromising clinical feasibility.

Figures