Low-Flow Duration and Outcomes of Extracorporeal Cardiopulmonary Resuscitation in Adults With In-Hospital Cardiac Arrest: A Nationwide Inpatient Database Study

OBJECTIVES: Although existing guidelines recommend commencing cannulation for extracorporeal cardiopulmonary resuscitation (ECPR) within 10-20 minutes of failed conventional resuscitation efforts for cardiac arrest, there is little supportive evidence. The present study aimed to determine the association of low-flow duration with survival-to-discharge rate in in-hospital cardiac arrest patients who received ECPR. DESIGN: A nationwide retrospective cohort study analyzed a nationwide inpatient database in Japan. Low-flow duration was defined as the time interval from initiation of chest compression to termination of chest compression. We assessed the association between low-flow duration and survival-to-discharge rate by predicting estimates with covariate adjustment stratified by categories of low-flow duration. SETTING: More than 1,600 acute-care hospitals in Japan. PATIENTS: All in-hospital cardiac arrest patients greater than or equal to 18 years old who received ECPR during hospitalization from July 2010 to March 2018. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Among 303,319 in-hospital cardiac arrest patients, 9,844 (3.2%) received ECPR in 697 hospitals during the study period and 9,433 were eligible in the study. The overall survival-to-discharge rate was 20.5% (1,932/9,433). The median low-flow duration was 26.0 minutes (interquartile range, 12.0-46.0 min) in the overall cohort. The highest and lowest estimated survival-to-discharge rates were 35.1% in the group with low-flow duration 0-5 minutes and 7.9% in the group with low-flow duration greater than 90 minutes. The estimated survival-to-discharge rate dropped sharply by about 20% during the first 35 minutes of low-flow duration (decreasing by about 3% every 5 min), followed by small decreases after the first 35 minutes. CONCLUSIONS: The estimated survival-to-discharge rate was markedly decreased by approximately 20% during the first 35 minutes of low-flow duration. Whether we should wait for the first 10-20 minutes of cardiac arrest without preparing for ECPR is questionable.