Abstract Introduction Employment of VA-ECMO in the treatment of refractory cardiogenic shock is increasing. Weaning rates are highly variable between studies. Methods We designed a retrospective observational single-centre study in order to analyse procedural success rate after including scheduled and standardized levosimendan infusion as part of our weaning protocol. Procedural success rate includes weaning from VA-ECMO, bridging to other mechanical circulatory support (MCS) or transplantation. Results A total of 37 VA-ECMO were managed in our Cardiac Intensive Care Unit (CICU) between January 2019 and March 2021. Most frequent aetiologies are reflected in table 1. Ventricular unloading was accomplished in 62.2% (23). We included four phases in our weaning protocol as described in figure 1. Second phase was used to select patients who could benefit from levosimendan infusion. This group included patients who accomplished the following criteria: appropriate aetiological treatment of shock, tolerance for 50% VA-ECMO flow reduction, noradrenaline dose <0,5 ug/kg/min, inotropic score <10 and correction of blood lactate, kidney and liver function parameters. Levosimendan was administered in 62.2% (23). 5 patients (13.5%) were not considered candidates to benefit from levosimendan (2 of them had early systolic function recovery and 3 patients seemed to have no chance to recover). Average time from VA-ECMO implantation to levosimendan infusion was 3.6 days, with a medium dose of 0,1 ug/kg/min. Norepinephrine medium dose was 0,12 mg/kg/min, with a medium inotropic score of 3 and vasoactive-inotropic score of 16,1. Levosimendan infusion did not cause any major side effects and was well tolerated by entire cohort even though norepinephrine dose had to be slightly increased in some cases. Whole sample procedural success rate reached 64.9% (24) whereas weaning rate from VA-ECMO without further support was 51.4% (19). 5 patients were upgraded to another MCS and 4 of them survived after transplantation. The subgroup which meets inclusion criteria to levosimendan reached a procedural success rate of 82.6% (19) and a weaning rate of 73.9% (19). Differences between patients with and without levosimendan treatment were statistically significant (procedural success rate: 82.6% vs 35.7%, p 0.006, weaning rate: 73.9% vs 14.3%, p<0.001). Meanwhile there were no significant differences regarding in-hospital mortality rate: 47.8% (11) vs 71.4% (10), p 0.160, even though there was a downward trend in levosimendan treated patients. Most frequent causes of death are shown in table 1. Only 3 patients died due to cardiac failure (1 was rejected for cardiac surgery and 2 were not candidates to other advanced therapies). Conclusions Standardized levosimendan infusion as part of VA-ECMO weaning protocol could enhance myocardial recovery and significantly increase procedural success rate. Randomized or propensity score matching studies would be desirable to prove this hypothesis. Funding Acknowledgement Type of funding sources: None. Table 1Weaning protocol