Effects of levosimendan on mortality in patients undergoing cardiac surgery: A systematic review and meta‐analysis

We sought to determine the impact of levosimendan on mortality following cardiac surgery based on large‐scale randomized controlled trials (RCTs).

opener, has been shown to decrease mortality in small randomized controlled trials (RCTs) and meta-analyses, involving patients undergoing cardiac surgery. [7][8][9][10] A Bayesian network analysis suggested that levosimendan might be the best inotrope for survival advantage among patients undergoing cardiac surgery. 11 However, these meta-analyses were based on small clinical trials with varying drug dosages, administered at different times, to patients undergoing a variety of cardiac surgical procedures. Recently, two large-scale RCTs (LEVO-CTS and CHEETAH) did not find a survival benefit for levosimendan in cardiac surgery. 12,13 We therefore performed a systematic review of all available randomized studies and a meta-analysis of placebo-controlled randomized trials to evaluate the effectiveness and safety of levosimendan in the treatment of patients undergoing cardiac surgery. and ClinicalTrials.gov using the search terms levosimendan, simdax, heart surgery, and cardiac surgery. The reference lists of retrieved articles and relevant reviews were carefully screened for further studies.

| Inclusion and exclusion criteria
Studies were included if they met the following requirements: (1) trial included patients undergoing cardiac surgery; (2) patients were randomly assigned to receive levosimendan or controls. No restrictions on dose or time of administration were applied; (3) one or more of the outcomes mentioned below were reported. Duplicate publications, secondary reporting format, abstracts, nonadult studies, and nonrandomized trials were excluded.

| Data extraction and outcomes of interest
Two authors (PL. Chen and XQ. Wu) independently extracted and summarized data from the identified studies into a standardized data extraction form, including study baseline characteristics (design, sample size,comparators, and follow-up), levosimendan infusion dose and duration, as well as primary outcomes and secondary outcomes. Any discrepancies were resolved by consensus.
The primary outcomes evaluated were all-cause mortality (in-hospital or within 30 days), serious adverse events, and myocardial infarction. The secondary endpoints included hypotension and low cardiac output syndrome. We contacted corresponding authors for detailed outcomes if they were missing in the original papers. The quality evaluation of included trials was performed using the Cochrane risk of bias assessment tool. 14

| Data synthesis and analysis
Computations were conducted with Review Manager 5.2 (Cochrane Collaboration, Oxford, UK). Odds ratio (OR) with 95% corresponding confidence interval (CI) was applied for dichotomous outcomes.
Statistical heterogeneity in studies was measured with I 2 statistic and the chi-square test. A P-value< 0.10 or an I 2 > 50% indicated the presence of high heterogeneity. The fixed-effects models were applied if there was no significant heterogeneity. Otherwise, the randomeffects model was considered. P < 0.05 is considered statistically significant for hypothesis testing.

| Risk of bias assessment of included studies
The methodological quality was satisfactory across most studies, as summarized in Figure 7. All included studies adopted a randomized, prospective, and comparative design. There was low risk of bias for selective reporting of outcomes, or incomplete outcome data. Two Levosimendan is a novel inotrope with an action mechanism differing from that of other catecholamines. Through stabilizing the binding of calcium to troponin C, levosimendan strengthens actin-myosin crossbridging and improves contractile force. It also reduces peripheral vascular resistance by acting on ATP-sensitive potassium channels of vascular smooth muscle cells. 33 Since it does not elevate intracellular concentrations of free calcium, levosimendan does not increase myocardial oxygen consumption. As levels of calcium fall in diastole, levosimendan does not impair myocardial relaxation which can occur with other catecholamines. 34 Furthermore, OR-1896, known as the active metabolite of levosimendan, has a half-life for elimination of 80 h. Consequently, a 24-h infusion may offer hemodynamic support for approximately 1 week. 35 Previous small clinical trials and meta-analyses of RCTs showed that levosimendan was associated with higher rates of survival, a lower incidence of periprocedural myocardial infarction, and lower rates of inotrope use than placebo, dobutamine, or milrinone among patients undergoing cardiac surgery. 8-10 These positive results were not confirmed in our present metaanalysis of placebo-controlled randomized trials. In addition, a survival benefit of levosimendan was also not seen in trials involving patients with severe sepsis or heart failure. [36][37][38] It was postulated that patients undergoing cardiac surgery who develop postoperative low cardiac output syndrome would benefit from levosimendan, due to the transient nature of postoperative depressed myocardial function. 39,40 Myocardial stunning is believed to be the main risk factor for the development of perioperative heart failure, and the heart function usually recovers within approximately 24-48 h. [39][40][41] Considering its unique mechanism of action (increase in cardiac function with a neutral effect on oxygen consumption), levosimendan seemed to be the ideal inotrope in cardiac surgery patients. However, in our present study, treatment with levosimendan did not show a significant reduction in mortality as compared with placebo, nor did it decrease the incidence of myocardial infarctions, hypotension, low cardiac output syndrome or other adverse events. Our results do not support the treatment with levosimendan in addition to standard care for perioperative cardiac dysfunction among patients undergoing cardiac surgery.
There may be several factors responsible for the different results of clinical trials with levosimendan in cardiac surgery. 10,17,25,42,43 Several previous trials of levosimendan in cardiac surgery, such as the LEVO-CTS trial, involved patients with reduced ejection fraction at baseline. 12 Other trials, however, including the CHEETAH trial focused on patients with ongoing myocardial dysfunction requiring inotropic support. 13 The dose of levosimendan differed among most trials. It is proposed that higher bolus doses of levosimendan may be more effective. 25,44,45 However, such regimens have been showed to be associated with a higher incidence of adverse events and less survival in other studies. 9 Moreover, the timing of administration of levosimendan may be of great importance, and treatment with levosimendan that is started just before surgery may not be effective to avoid perioperative myocardial The following limitations of this study must be taken into account.
First, all four trials assessed levosimendan as added to routine clinical care rather than comparing levosimendan with other inotropic agents.
Second, this study was conducted based on summary statistics instead of individual patient level data. There may be some confounders (eg, cardiac surgical operation, ejection fraction at baseline, etc.) at the patient-level influencing the clinical effect of levosimendan, but that were not available. Third, the dose of levosimendan varied across trials.
In conclusion, this study indicates that despite its unique inotropic and cardioprotective properties, levosimendan did not result in a higher rate of survival nor did it positively affect other secondary outcomes as compared with placebo for patients undergoing cardiac surgery. Well-designed, multicenter, large-scale RCTs are required to determine the role of levosimendan in cardiac surgery.

ACKNOWLEDGMENT
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.