Abstract
BACKGROUND: Selecting appropriate patients for extracorporeal cardiopulmonary resuscitation (ECPR) in cases of out-of-hospital cardiac arrest remains challenging, particularly for those with initial nonshockable rhythms. Our aim is to develop a prediction score to identify suitable ECPR candidates in patients with initial nonshockable rhythm. METHODS: The data were sourced from the SAVE-J II (Study of Advanced Life Support for Ventricular Fibrillation With Extracorporeal Circulation in Japan II) study, a retrospective multicenter observational study in Japan. Included were adult patients with out-of-hospital cardiac arrest who underwent ECPR with initial pulseless electrical activity or asystole. The primary outcome was survival to hospital discharge. We developed a prediction score, employing logistic regression analysis and internally validating it with 1000 bootstrap samples. The performance of the score in predicting a favorable neurological outcome at discharge was also evaluated. RESULTS: Among 648 eligible patients, 86 (13.3%) survived to hospital discharge. The median age was 60.5 years, 75.9% (492) were male, and 74.4% (482) had pulseless electrical activity as the initial rhythm. Three clinical predictors for the START-ECPR Score (Signs of Life, Transient ROSC, Not Asystole Rhythm to ECPR Score) were identified: shockable rhythm or pulseless electrical activity at hospital arrival, transient return of spontaneous circulation before hospital arrival, and signs of life at hospital arrival. Survival rates were 4.4% (7/159) for a score of 0, 10.7% (38/356) for a score of 1, and 30.8% (39/130) for scores of 2 to -3. The bias-corrected C-index for the score was 0.696. For predicting favorable neurologic outcomes at discharge, the C-index was 0.761. CONCLUSIONS: We developed a straightforward 3-factor prediction score for predicting survival to hospital discharge and favorable neurologic outcomes in patients with out-of-hospital cardiac arrest with initial nonshockable rhythms receiving ECPR.