Epinephrine administration in venoarterial extracorporeal membrane oxygenation patients is associated with mortality: a retrospective cohort study.

AIMS Knowledge about the impact of epinephrine on the outcome in venoarterial (VA) extracorporeal membrane oxygenation (ECMO) patients is limited, and existing data are conflicting. METHODS AND RESULTS We conducted a retrospective cohort study in a 1500 bed tertiary university hospital. Five hundred and eighty-nine VA-ECMO patients were analysed. The median age was 57 years [47-65], 68% of male. The major indications for ECMO were post-cardiotomy cardiogenic shock (CS) (38%) and medical CS (36%). Two hundred and sixty-two (44.5%) patients received epinephrine alone or associated with another catecholamine while on ECMO. Baseline factors significantly associated with epinephrine administration were younger age, higher sequential organ failure assessment score, cardiac arrest at implantation, and intra-aortic balloon pump support at implantation, whereas medical CS and dobutamine administration were significantly associated with a lower risk of epinephrine administration. Epinephrine administration was independently associated with death [hazard ratio = 1.68 (1.44-2.23); P < 0.01]. A sensitivity analysis with propensity score inverse probability weighting in complete cases confirmed a significant association of epinephrine administration with death [hazard ratio = 1.69 (1.43-2.00); P < 0.001]. CONCLUSIONS Among patients who required VA-ECMO, epinephrine administration was associated with an increased risk for death.