Levosimendan in patients undergoing cardiac surgery: does timing matter?

Type of funding sources: None.

Levosimendan (LEV) is an inotropic drug that improves myocardial contractility without increasing oxygen requirement and its use in cardiac surgery has been questioned. Early meta-analysis showed that it was associated with reduced mortality, however recent randomized trials failed to show a benefit of this drug in some specific subsets of patients.

This study sought to evaluate if the timing of the initiation of LEV infusion could have impact in short-term prognosis.

This was an unicenter retrospective study of patients submitted to a cardiac surgery between December 2009 and December 2018 and who did LEV during hospitalization. Three groups were established upon the beginning of LEV administration: preoperative, intraoperative and postoperative. Statistical comparisons between groups are presented below in this order.

Out of 141 patients, 40,4% belonged to the preoperative group, 31,2% to the intraoperative group and 28,4% to the postoperative group. Demographic data was balanced between the three groups with the exception of past history of peripheral artery disease (3,6% vs 15,9% vs 2,5%, p=0,024) and previous cardiac surgery (19,3% vs 4,5% vs 2,5%, p=0,009). The type of cardiac surgery, whether with or without cardiopulmonary bypass, was similar in all groups. Left ventricular ejection fraction was not significantly different (36,9±12,3% vs 41,2±12,5% vs 42,2±12,2%, p=0,080) and the same happened with the proportion of patients with right ventricle dysfunction (12,3% vs 9,1% vs 10%, p=0,865). Moreover, total dose of LEV and its infusion duration were not different (p=0,117 and p=0,242, respectively).

Compared to the postoperative group, preoperative and intraoperative initiation was not independently associated with a significantly lower risk of in-hospital cardiogenic shock requiring inotropic drugs (adjusted OR 0,84 (95% CI 0,23-3,06) and 1,04 (95% CI 0,30-3,60), respectively) or mechanical circulatory support (adjusted OR 1,29 (95% CI 0,18-9,34) and 0,57 (95% CI 0,06-5,64)), myocardial infarction (adjusted OR 1,25 (95% CI 0,16-9,87) and 1,46 (95% CI 0,20-10,51)) and death (adjusted OR 1,85 (95% CI 0,28-12,18) and 4,41 (95% CI 0,81-24,00)). However, preoperative LEV was associated with less acute kidney injury (adjusted OR 0,36 (95% CI 0,14-0,93)) compared to postoperative initiation. Still, the need for renal replacement therapy didn’t differ significantly between all groups. Overall, subgroup analysis showed identical results independently of biventricular systolic function, type of surgery and need of cardiopulmonary bypass.

In this heterogeneous population submitted to a cardiac surgery, timing of LEV initiation did not have a significant impact on short-term in-hospital major adverse outcomes, independently of biventricular systolic function and type of surgery.