Extracorporeal Membrane Oxygenation Bridges Inoperable Patients to Definitive Cardiac Operation

Extracorporeal membrane oxygenation (ECMO) offers an opportunity for patient recovery through complete cardiopulmonary support but is associated with complications that limit duration and overall utility. We examine the role of ECMO as a potential bridge to high-risk cardiac surgery in otherwise inoperable cases. This study reports a retrospective, multi-institution experience examining all patients for whom ECMO was used preoperatively as a bridge to definitive cardiac surgery without exception. A consecutive patient database (December 2011 through August 2017) was utilized. European System for Cardiac Risk Evaluation (EuroSCORE) 2 was calculated as a metric of patient acuity and risk assessment. Observed and expected mortality were compared. Twelve adult patients fit inclusion criteria and were supported with ECMO during the study period. There were five males and seven females. Average age was 56 (39-77) years. All 12 patients were supported with venoarterial ECMO for cardiogenic shock. This was done in preparation for corrective conventional cardiac surgery. Definitive cardiac surgical procedures included complex valve (n = 5), left ventricular assist device (n = 3), coronary artery bypass grafting (CABG; n = 2), CABG/ventricular septal defect repair (n = 1), and mitral valve replacement/CABG (n = 1). Average time of ECMO support was 200 (range 113-379) hours. Three patients were decannulated from ECMO at the conclusion of definitive cardiac surgery. Risk assessed by Logistic EuroSCORE 2 ranged from 64% to 89%. Average EuroSCORE 2-predicted mortality representing all 12 patients was 77%. Thirty day mortality was 25% (3/12), and hospital mortality was 33% (4/12). Seven patients are still alive today, with a mean survival of 37 (range 2-64) months. Two deaths were associated with gastrointestinal bleeding and two with evolving liver failure. Mean difference between the EuroSCORE 2 prediction model and actual observed 30 day mortality rate was 42.33 (95% CI 36.86-47.98) with a two-tailed, one-sample t test value of p < 0.001. ECMO can successfully be utilized as a bridge to conventional cardiac surgical procedures in critically ill patients, with a historically high mortality.