Predictors of mortality after extracorporeal cardiopulmonary resuscitation

Objective: Extracorporeal membrane oxygenation (ECMO) is a promising adjunct to cardiopulmonary resuscitation (CPR) in refractory cardiac arrest (CA). Factors associated with outcome are incompletely characterised. The aim of our study was to identify pre-ECMO factors associated with in-hospital mortality after extracorporeal CPR (ECPR). Design: Retrospective analysis of a prospective cohort of patients. Setting: Academic quaternary referral hospital. Participants: All patients who underwent ECPR from January 2012 through April 2017. Interventions: A retrospective chart review was performed for CPR and ECMO. A multivariable logistic regression was performed to identify factors associated with mortality after ECPR. Main outcome measures: Primary outcome was in-hospital mortality. Secondary outcomes included survival with favourable neurologic outcome, days on ECMO, and intensive care unit (ICU) length of stay. Results: During the study period, 75 patients received ECPR. Median age was 59 years, 81% were male, 51% had out-of-hospital CA, and 57% had an initial shockable rhythm. Median time from arrest to ECMO was 91 minutes (IQR, 56–129) for non-survivors and 51 minutes (IQR, 37–84) for survivors (P = 0.02). Twenty-six patients (39%) were successfully separated from ECMO, with 31% surviving to hospital discharge and 29% with a cerebral performance category score of 1 or 2. In multivariable analysis, significant predictors of in-hospital mortality were ongoing CPR at the time of ECMO initiation (P < 0.01) and arrest to ECMO cannulation time (P = 0.02). Conclusion: Following ECPR, the factors most strongly associated with mortality were ongoing CPR at the time of ECMO initiation and arrest to ECMO cannulation time. Interventions aimed at reducing time to ECMO initiation may lead to improved outcomes.