Long-term outcomes of extracorporeal membrane oxygenation support for postcardiotomy shock

Objective: Extracorporeal membrane oxygenation (ECMO) is a widely used technique for treating postcardiotomy cardiogenic shock (PCS); however, no study has compared the long-term outcomes of patients who receive ECMO support for PCS with those of the general population post cardiac surgery. Methods: A total of 1141 patients who received ECMO after cardiac surgery between 2000 and 2011 were identified by using the Taiwan National Health Insurance Research Database. For each patient, we matched 5 non-ECMO patients who had undergone cardiac surgery by using propensity scores calculated for age, sex, 12 comorbid diseases, Charlson score, hospital level, type of cardiac surgery, and year of index hospitalization. The outcomes included all-cause mortality, readmission for any cause, and medical expenditure. Results: The incidence of ECMO use after cardiac surgery in Taiwan was 1.91%. The ECMO group had a significantly higher risk of in-hospital mortality than did the non-ECMO group (61.7% vs 6.8%, odds ratio 22.34, 95% confidence interval 19.06-26.18). The risks of all-cause mortality and first readmission for any cause were greater in the ECMO group than that in the control group (P<.001, P<.001) in the first year, whereas no difference was observed after the first year of follow-up (P = .209, P = .474). Similar results were observed regarding the medical expenditure of admission after index admission discharge. Conclusions: Patients receiving ECMO for PCS had similar outcomes to those of the non-ECMO group after the first year of follow-up despite significantly poor outcomes during the in-hospital course.