BackgroundVenoarterial extracorporeal membrane oxygenation (VA-ECMO) has been increasingly used in patients with refractory cardiogenic shock (CS) or out-of-hospital cardiac arrest. It is difficult to perform VA-ECMO weaning, which may cause circulatory failure and death. Levosimendan is an effective inotropic agent used to maintain cardiac output, has a long-lasting effect, and may have the potential benefit for VA-ECMO weaning. The study aimed to explore the relationship between the early use of levosimendan and the rate of VA-ECMO weaning failure in patients on VA-ECMO support for circulatory failure.MethodsAll patients who underwent VA-ECMO in our hospital for CS between January 2017 and December 2020 were recruited in this cohort study and divided into two groups: without and with levosimendan use. Levosimendan was used as an add-on to other inotropic agents as early as possible after VA-ECMO setting. The primary endpoint was VA-ECMO weaning success, which was defined as survival without events for 24 h after VA-ECMO withdrawl. The secondary outcomes were cardiovascular and all-cause mortality at the 30-day and 180-day follow-up periods post-VA-ECMO initialization.ResultsA total of 159 patients were recruited for our study; 113 patients were enrolled in the without levosimendan-use group and 46 patients were enrolled in the levosimendan-use group. In levosimendan-use group, the patients received levosimendan infusion within 24 h after VA-ECMO initialization. Similar hemodynamic parameters were noted between the two groups. Poorer left ventricular ejection fraction and a higher prevalence of intra-aortic balloon pumping were observed in the levosimendan group. An improved weaning rate (without vs. with: 48.7 vs. 82.6%; p < 0.001), lower in-hospital mortality rate (without vs. with: 68.1 vs. 43.5%; p = 0.007), and 180-day cardiovascular mortality (without vs. with: 75.3 vs. 43.2%; p < 0.001) were also noted. Patients administered with levosimendan also presented a lower rate of 30-day (without vs. with: 75.3 vs. 41.3%; p = 0.034) and 180-day (without vs. with: 77.0 vs. 43.2%; p < 0.001) all-cause mortality.ConclusionEarly levosimendan administration may contribute to increasing the success rate of VA-ECMO weaning and may help to decrease CV and all-cause mortality.