Abstract
Background: Determining the optimal duration of extracorporeal cardiopulmonary resuscitation (ECPR) remains challenging, as patient outcomes may vary significantly based on individual characteristics. We aimed to establish critical time thresholds for achieving favorable neurological outcomes with ECPR across different risk groups, potentially providing more tailored guidance for clinical decision-making. Methods: This single-center retrospective study screened 279 adult patients who received ECPR between 2013 and 2020. Through multivariate analysis of various clinical parameters, we developed a pragmatic bedside risk stratification framework to identify groups with different prognostic profiles. The primary outcome was neurological status at discharge, assessed by the Cerebral Performance Categories scale. Results: In multivariate analysis, age greater than 50 years with asystole (adjusted odds ratio [OR]: 4.89, 95% confidence interval [CI]: 1.41-17.00) or pulseless electrical activity (adjusted OR: 9.70, 95% CI: 2.80-33.60), aspartate transaminase (adjusted OR: 1.52, 95% CI: 1.15-1.99), creatinine (adjusted OR: 2.08, 95% CI: 1.30-3.34), initial lactate (adjusted OR: 1.88, 95% CI: 1.27-3.45), and low-flow time (adjusted OR: 3.50, 95% CI: 2.02-6.06) were associated with poor neurological outcomes. Based on these findings, we identified three distinct risk groups showing different acceptable low-flow time thresholds: low-risk (38 min), moderate-risk (27 min), and high-risk (20 min). Notably, no favorable neurological outcomes were observed beyond 70 min in the low-risk group and 90 min in moderate/high-risk groups. Risk group stratification effectively predicted neurological outcomes across different low-flow time intervals. Conclusions: Risk-stratified evaluation of low-flow time (cardiac arrest to ECMO pump-on) provides clinically relevant thresholds for different patient groups, suggesting that continuation of ECPR may be warranted in low-risk patients even with extended low-flow times. This approach may enable more personalized decision-making in ECPR implementation.