RESULTS: Twelve trials were eligible (n = 1867) for inclusion in the data synthesis. In comparison to the placebo cohort, the levosimendan cohort showed a significant reduction in mortality (TSA = inconclusive; ρ = 0.002; I2 = 0%; FEM: OR 0.56; 95% CI 0.39, 0.80), especially in the subgroups of preoperative severe low LVEF ≤ 30% (ρ = 0.003; OR 0.33; 95% CI 0.16, 0.69), preoperative administering of levosimendan (ρ = 0.001; OR 0.46; 95% CI 0.29, 0.74) and patients who had bolus followed by infusion of levosimendan (ρ = 0.005; OR 0.50; 95% CI 0.30, 0.81). However, the effect on mortality was not significant in the subgroup analysis of high quality trials (ρ = 0.14; OR 0.73; 95% CI 0.47, 1.12). The levosimendan cohort showed a significantly lower incidence of low-cardiac-output-syndrome (ρ