RESUMEN
OBJECTIVE: Investigate how a multitude of patient demographics and extracorporeal membranous oxygenation (ECMO)-related complications affect 30-day survival or survival to discharge. DESIGN: Retrospective observational study. SETTING: Urban university hospital, quaternary care center. PARTICIPANTS: Patients who underwent ECMO circulatory support from January 2012 to May 2016. INTERVENTIONS: Date-based data extraction, univariate and multivariate regression analysis. MEASUREMENTS AND MAIN RESULTS: The hospital database contained complete data for 235 adult patients who received venoarterial ECMO (74.04 %) and venovenous ECMO (25.96 %); 106 patients (45.11%) survived. The independent predictors significant in the odds of in-hospital mortality in a multiregression model were age (odds ratio [OR]â¯=â¯1.028, pâ¯=â¯0.008), extracorporeal cardiopulmonary resuscitation (ECPR) after unsuccessful high-quality CPR (ORâ¯=â¯7.93, p =0.002), cardiogenic shock as the primary indication for circulatory support (ORâ¯=â¯2.58, pâ¯=â¯0.02), acute kidney injury (AKI) before ECMO initiation (ORâ¯=â¯7.53, p < 0.001), time spent on ECMO in days (ORâ¯=â¯1.08, pâ¯=â¯0.03), and limb ischemia (ORâ¯=â¯3.18, pâ¯=â¯0.047). CONCLUSION: The most significant findings of advancing age, time spent on ECMO, AKI, ECMO use in the setting of cardiogenic shock, ECPR, and limb ischemia as a complication of ECMO all independently increase the odds of in-hospital and 30-day mortality. To the best of the authors' knowledge, this study is the first to demonstrate a significant relationship between limb ischemia and mortality.