To assess the effect of preoperative levosimendan on mortality at day 90 in patients with left ventricular ejection fraction (LVEF) ≤ 40%, and to investigate a possible differential effect between patients undergoing isolated coronary artery bypass grafting (CABG) versus CABG combined with valve replacement surgery.

Pooled analysis of two multicentre randomised controlled trials (RCT) investigating prophylactic levosimendan versus placebo prior to CABG surgery on mortality at day 90 in patients with LVEF ≤ 40%. A meta-analysis of all RCT investigating the same issue was also conducted.

A cohort of 1084 patients (809 isolated CABG, and 275 combined surgery) resulted from the merging of LEVO-CTS and LICORN databases. Seventy-two patients were dead at day 90. The mortality at day 90 was not different between levosimendan and placebo (Hazard Ratio (HR): 0.73, 95% CI: 0.41–1.28, p = 0.27). However, there was a significant interaction between the type of surgery and the study drug (p =  0.004). We observed a decrease in mortality at day 90 in the isolated CABG subgroup (HR: 0.39, 95% CI: 0.19–0.82, p = 0.013), but not in the combined surgery subgroup (HR: 1.73, 95% CI: 0.77–3.92, p = 0.19). The meta-analysis of 6 RCT involving 1441 patients confirmed the differential effect on mortality at day 30 between the 2 subgroups.

Preoperative levosimendan did not reduce mortality in a mixed surgical population with LV dysfunction. However, the subgroup of patients undergoing isolated CABG had a reduction in mortality at day 90, whereas there was no significant effect in combined surgery patients. This finding requires confirmation with a specific prospective trial.