Table 2.

Population included in studies selected for meta-analysis

Author, study [reference]	Patients (n) and operations	LVEF cutoffs	Administration timing and dosages
Low LVEF only
Erb et al., 2014 [19]	33 On-pump CABG (with or without valve)	<30%	Before incision 12.5-mg total dose at 0.1 μg∙kg−1∙minute−1
Levin et al., 2012 [20]	252 On-pump CABG only	<25%	Preoperative 10-μg/kg bolus; 0.1 μg∙kg−1∙minute−1 for 23 h
Mehta et al., 2017 [13]	849 On-pump cardiac surgery	<35%	Before incision 0.2 μg/kg/minute for 1 h; 0.1 μg∙kg−1∙minute−1 for 23 h
Shah et al., 2014 [21]	50 Off-pump CABG only	<30%	Preoperative 0.133 μg∙kg−1∙minute−1 for 24 h
Sharma et al., 2014 [22]	40 CABG and mitral valve repair	<30%	Preoperative 200 μg∙kg−1 for 24 h
Lomivorotov et al., 2012a [17]	60 On-pump CABG only	<35%	Before incision 12-μg∙kg−1 bolus; 0.1 μg∙kg−1∙minute−1 for 24 h
LCOS only
–	–	–	–
Low LVEF and LCOS
Landoni et al., 2017 [14]	504 All cardiac surgery	<25%/or LCOSb	Mainly postoperativeb 0.05 μg∙kg−1∙minute−1 for 48 h or until ICU discharge

Abbreviations: CABG Coronary artery bypass grafting, LCOS Low cardiac output syndrome, LVEF Left ventricular ejection fraction

The studies are classified according to the subgroup of low LVEF and/or LCOS. We also report the number of patients in each study, the timing of levosimendan (placebo) administration, and the outcomes of interest of our meta-analysis reported by each study

^aThe study of Lomivorotov et al. [17] was a three-arm study with patients with low LVEF receiving preoperative levosimendan and intra-aortic balloon pump (IABP) vs levosimendan alone vs IABP alone. The data from the first two groups were included in a sensitivity analysis

^bIn this trial, only 4% of patients were randomized according to a preoperative low LVEF, 19% according to the need for IABP, 12% for difficult weaning from cardiopulmonary bypass, and 65% for postoperative LCOS