Back

Application of severity classification after return of spontaneous circulation for predicting long-term outcomes in cardiac arrest survivors

Lin, J.-J.; Huang, C.-H.; Chen, W.-T.; Lee, H.-Y.; Chen, W.-J.; Chang, W.-T.; Ong, H.-N.; Tsai, M.-S.

2025-11-06 intensive care and critical care medicine
10.1101/2025.11.03.25339449 medRxiv
Show abstract

BackgroundAccurately predicting outcomes for cardiac arrest (CA) survivors is critical, and several severity scores have been developed to predict outcomes at hospital discharge. However, the ability of these scores to predict long-term outcomes at 6 months and 1 year in CA survivors remains unexplored. MethodsThis retrospective observational study enrolled 1773 adult patients with nontraumatic CA in the emergency department of National Taiwan University Hospital between January 2011 and June 2023 without interhospital transfer. Data on clinical variables, resuscitation events, post-arrest care, and outcomes were collected, and the disease severity was classified using the sCAHP, rCAST, and TIMECARD scores. The outcomes included 6-month and 1-year survival. The predictive ability of these scores was evaluated using receiver operating characteristic curves. ResultsAmong the 1773 enrolled patients, 501 (28.3%) survived to discharge, 361 (20.4%) survived for 6 months, and 346 (19.5%) survived for 1 year. The area under the receiver operating characteristic curves (AUCs) for 6-month survival were 0.787, 0.733, and 0.810 for the sCAHP, rCAST, and TIMECARD scores, respectively. For 1-year survival, the AUC values were 0.788, 0.734, and 0.816, respectively. The sCAHP and TIMECARD scores demonstrated better discriminatory performance than did the rCAST score for both 6-month and 1-year survival. Compared with the sCAHP and rCAST scores, the TIMECARD score exhibited superior discriminatory ability for 1-year survival, particularly in younger patients (age [≤]65 years), those with in-hospital CA (IHCA), and those with good pre-arrest neurological status. ConclusionThe sCAHP, rCAST, and TIMECARD scores demonstrated good discrimination and calibration in predicting long-term outcomes in CA survivors. The TIMECARD score exhibited better discriminatory performance for 1-year survival, particularly in the subgroup of younger age, IHCA, and good pre-arrest neurological function.

Published in International Journal of Cardiology (predicted rank #3) · training set

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.