Background: Veno-arterial extracorporeal membrane oxygenation (VA-ECMO) can re-establish tissue perfusion in refractory cardiac arrest requiring cardiopulmonary resuscitation (CPR). Levosimendan showed potential benefit in VA-ECMO weaning and mortality but was not studied in extracorporeal CPR (ECPR). This study examined levosimendan effect on in-hospital survival with good neurological outcomes in ECPR.

Methods: A retrospective cohort study recruited patients receiving ECPR between January 2015 and March 2021. Neurological outcome was expressed by cerebral performance category.

Results: Eighty-seven patients were included with mean age of 45.4 years and 86.2% males. Eighteen (20.7%) patients received levosimendan. Baseline characteristics were similar between groups except dopamine use was significantly higher in comparator group (odds ratio = 0.134; 95% CI: 0.024-0.752, p=0.022). ~70% of patients in either group suffered in-hospital cardiac arrest and 70% had asystole. In > 50% of them, myocardial infarction was the cardiac arrest etiology. Overall, ~ 50% of patients received electrical shocks (Table 1). In levosimendan group, duration of CPR prior to VA-ECMO was numerically longer, lactate levels between second and fourth days after ECPR were higher (p=0.047) and VA-ECMO support was longer (p=0.019) (Table 2). Survival to decannulation (27.8% vs 26.1%) or hospital discharge (27.8% vs 24.6%) did not differ between groups. Almost all survived patients (n=22) had good neurological outcome. Four and three patients died in levosimendan and comparator groups (p=0.281), respectively, within 6 months post discharge. Length of ICU and hospital stays were similar. (Table 2).

Conclusions: Levosimendan did not improve survival in ECPR. Well-designed trials are needed in this population.