Secondary Logo

Journal Logo

Levosimendan in Off-Pump Coronary Artery Bypass: A Four-Times Masked Controlled Study

Barisin, Stjepan MD, MSc; Husedzinovic, Ino MD, PhD; Sonicki, Zdenko MD, PhD; Bradic, Nikola MD; Barisin, Ana MD; Tonkovic, Dinko MD, PhD

Journal of Cardiovascular Pharmacology: December 2004 - Volume 44 - Issue 6 - p 703-708
Original Article

We tested the hypothesis that levosimendan produced beneficial hemodynamic effects during and after off-pump coronary artery bypass grafting in patients with good preoperative left ventricular function. Levosimendan at low dose (12 μg/kg), high dose (24 μg/kg), or placebo were administered in thirty-one patients in a randomized and four-times masked controlled study.

Heart rate was not significantly different between experimental groups. Significant increases in cardiac output and left ventricular ejection fraction occurred after high-dose (P < 0.001; P = 0.006) and low-dose levosimendan (P = 0.001; P = 0.002). Both doses of levosimendan produced significant increased stroke volume and decreased systemic vascular resistance. Mean arterial pressure, pulmonary capillary wedge pressure, and left ventricular end-systolic volume were not significantly different between groups.

The low-dose levosimendan produced better hemodynamic response than high-dose and may be preferable in patients undergoing off-pump coronary artery bypass grafting.

From the Clinical Department of Anesthesiology, Reanimatology, and Intensive Care Medicine, University Hospital Dubrava, Zagreb, Croatia.

Received for publication August 2, 2004; accepted September 22, 2004.

Reprints: Stjepan Barisin, University Hospital Dubrava, Clinical Department of Anesthesiology, Reanimatology, and Intensive Care Medicine, Gojko Susak Avenue 6, Zagreb, 10000 Croatia (e-mail:

Levosimendan is a calcium sensitizer that enhances the contractile force of the myocardium by binding to troponin C without increasing intracellular calcium concentration at therapeutic doses. It does not impair relaxation of intact paced guinea pig papillary muscles or of isolated failing human myocardium.1,2 In addition, levosimendan exerts cardioprotective effects including coronary artery vasodilatation via activation of triphosphate-regulated potassium channels at a dose that enhances myocardial contractility.3,4

Despite cardioplegic protection, cardiopulmonary bypass is associated with postoperative myocardial stunning, hypothermia, formation of microemboli, and systemic inflammatory response syndrome, all of which may prolong recovery from coronary artery bypass grafting (CABG).5 There is an association between improved patient outcome and off-pump coronary artery bypass grafting (OPCABG), although immediate postoperative left ventricular dysfunction occurs with both types of surgery.6 Manipulation of the beating heart during OPCABG, especially on the posterior and lateral left ventricular vessels produces significant fluctuation in the patients' hemodynamic status.7,8 Inotropic support and use of vasoconstrictors are often necessary for the long-lasting myocardial dysfunction after brief periods of ischemia.

The brief period of ischemia (myocardial stunning) during direct coronary occlusion is characterized by reversibly impaired postischemic systolic and diastolic function. The mechanism of myocardial stunning is directly connected with decreased myofibril calcium sensitivity.9,10

Two clinical studies have shown a positive effect of levosimendan on myocardial function after cardiopulmonary bypass in patients undergoing elective heart surgery.11,12 In the first study, levosimendan improved cardiac performance without increasing myocardial oxygen consumption or changing myocardial substrate utilization after cardiac surgery. The second study showed that levosimendan augments cardiac performance after cardiopulmonary bypass in patients with normal preoperative left ventricular function.

The goal of our investigation is testing the hypothesis that levosimendan, as a novel inodilator, could produce beneficial hemodynamic effects and enhance left ventricular performance during and after OPCABG in patients with good preoperative left ventricular function.

Back to Top | Article Outline


Study Patients

Of 33 patients undergoing OPCABG who were enrolled in this randomized, placebo-controlled and 4-times masked study, 31 patients met inclusion criteria and were entered in the data analysis. The Hospital Ethical Committee approved the study protocol. All the participants were informed about the investigation and signed the Informed Consent form before the cardiac surgery. All patients were younger than 75 years, all of them were classed as I or II degree of Cardiac Anesthesia Risk Evaluation score,13 and also had angiographically verified coronary artery disease (Table 1). Patients with myocardial infarction and cerebral stroke within 3 months were excluded from the study. Also, patients with significant valvular stenosis, second- or third-degree atrioventricular block, preexisting left ventricular dysfunction (ejection fraction <40%, preoperative inotropic and/or mechanical support), significant pulmonary disease, severe hepatic (bilirubin 1.5 and ASTand ALT 2 times higher than maximal approved limit) and renal disease (serum creatine more than 1.3 mg/dL), or sepsis with multiple organ failure were excluded from enrolment. Exclusion criteria during OPCABG were hemodynamic instability (systolic blood pressure <90 mm Hg, heart rate >100 beats/min, elevation or depression of the ST-segment by more than 1 mm on 12-lead ECG), inotropic support before administration of placebo or levosimendan, sustained ventricular arrhythmias requiring antiarrhythmic support or immediate need for cardiac pacing. Two patients were subsequently excluded from the study and from data analysis. One patient was randomized to the placebo group because he fulfilled 2 exclusion criteria during operation (110 beats/min and elevation of the ST-segment in inferior leads by 2.3 mm). The second patient did not receive high-dose levosimendan because the surgeon changed his decision and performed CABG.



Back to Top | Article Outline

Anesthesia Protocol

Long-acting medications (angiotensin converting enzyme inhibitors, calcium antagonists, and nitrates) were discontinued 24 hours before starting the study. All patients received their scheduled cardiac medications on the morning of surgery and premedication half an hour before the surgery. As premedication, midazolam (Dormicum®, F. Hoffman-La Roche Ltd., Basel, Switzerland) in dose of 0.07-01 mg/kg IM was prescribed. Anesthesia was induced with midazolam 0.1 mg/kg IV, fentanyl (Fentanyl®, Janssen Pharmaceutica, Beerse, Belgium) 5 to 7 μg/kg IV, and pancuronium-bromide (Pavulon®, N. V. Organon, Oss, the Netherlands) 0.08 to 0.12 mg/kg IV. After endotracheal intubation, the lungs were mechanically ventilated using positive pressure and 100% O2 (2 L/min, tidal volume 8 mL/kg, ventilatory frequency 12/min) (Cato Dräger, Lübeck, Germany). Anesthesia was maintained with continuous infusion of fentanyl (0.1-0.15 μg/kg/min) and midazolam (0.5-0.75 μg/kg/min). Additional doses of pancuronium-bromide (0.1 mg/kg) were administered as required to maintain neuromuscular blockade during surgery.

Back to Top | Article Outline

Hemodynamic Monitoring

In the operating room an arterial catheter (Arrow International, Reading, PA) was inserted into the left radial artery and a 5-lumen Swan-Ganz catheter 7.5 Fr (Arrow International) into the right internal jugular vein. Heart rate, arterial and central venous pressure values, and the progression of the catheter through the jugular vein into a pulmonary artery branch could be seen on a monitor (Hewlett Packard Viridia CMS; Böblingen, Germany) as well as the changes in pulmonary arterial pressure curve and values. Radial and pulmonary arterial pressure transducers (Peter von Berg, Kirchseeon, Germany) were zeroed at the level of the left atrium. Thermodilution cardiac output measurements were obtained using an inert indicator 10 mL of 5% glucose at room temperature. The thermodilution curve was monitored on the thermodilution monitor (Cardiac Output Computer; Arrow International). Other hemodynamic parameters, such as mean arterial pressure, central venous pressure, mean pulmonary arterial pressure, pulmonary capillary wedge pressure, and heart rate were measured by Swan-Ganz catheter immediately after the assessment of cardiac output. Just after the induction of anesthesia, the transesophageal echo 7.0 MHz Multi-Plane probe (ATL Ultramark HDI 3000®; Philips Medical Systems, Bothell, WA) was inserted through the mouth and positioned in the stomach. With the probe tip in the stomach, superior angulation (flexing the scope) in the 0° image plane yielded the transgastric mid short-axis view. At this level, we determined left ventricular systolic function by recording the cross-sectional view of the left ventricle.

Back to Top | Article Outline

Study Design

In this 4-times masked clinical study, the patients did not know which medication they received, the anesthesiologist did not know which medication was used during the surgery, and finally, the statistician and the authors during the data accumulation and the writing of text also did not know which patients were in each group. After opening the patients' group code, comparative results were determined. A computer randomization schedule based on permuted blocks was used to allocate patients to placebo or 1 of 2 doses of levosimendan (Simdax®; Orion Corporation, Espoo, Finland): 12 μg/kg (low-dose levosimendan) and 24 μg/kg loading dose (high-dose levosimendan). The placebo was identical in appearance to the active drug (riboflavinophosphate 0.4 mg, ethanol 100 mg, and 5% glucose yielding a yellow color). Drug infusions were administered 20 minutes before starting of surgery over a period of 10 minutes. Study medications were introduced via a central vein using an infusion pump Omnifuse Graseby (SIMS Graseby Ltd.; Watford, UK). After induction and before starting medication, 500 mL hydroxyethylstarch 6% solution (HAES-sterile 6% in saline 0.9%, Fresenius Kabi, Bad Homburg, Germany) was administered in all patients to optimize preload. Before starting and immediately after ending drug infusion, heart rate, mean arterial pressure, pulmonary capillary wedge pressure, cardiac output, stroke volume, systemic vascular resistance, left ventricular ejection fraction and left ventricular end-systolic volume were measured. The measurements were repeated thereafter 5, 20, and 50 minutes after drug administration.

Back to Top | Article Outline


Numerical data were described by mean, standard deviation, median and interquartile range. Qualitative data were described by frequencies. To compare the three independent groups (placebo, low-dose and high-dose levosimendan) the Kruskal-Walis test was applied. To find differences between the 3 groups, we tested the groups in pairs by the Mann-Whitney U test corrected for multiple testing. For comparing time points in one group, we tested the differences in time within each group with the Friedman test. To find differences between each group of patients, we tested time-dependent variable pairs within each group by the Wilcoxon Signed Rank test corrected for multiple testing. For data analysis the SAS System for Windows Release 6.12 software (SAS Institute Inc., Cary, NC) was used.

Back to Top | Article Outline


Demographics, preoperative, and operative data were similar between experimental groups (Table 1). Except for a decrease in heart rate that occurred in high-dose levosimendan compared with those treated with placebo, no differences in hemodynamic parameters were observed between groups in the baseline measurements (Table 2).



Compared with placebo, the high-dose (P < 0.001) and low-dose levosimendan (P = 0.003) caused an increase in heart rate. Compared with the baseline measurement, all values of heart rate were significantly higher (P = 0.005 for all) in patients treated with high-dose levosimendan. In patients receiving low-dose levosimendan, compared with the baseline measurement, all values of heart rate were also significantly higher (P = 0.016 for 0, 20, and 50 minutes; P =0.029for 5 minutes) (Table 2).

No differences in mean arterial pressure and pulmonary capillary wedge pressure were observed between and within groups at baseline and at other measurements after levosimendan or placebo administration.

Cardiac output was significantly higher in patients receiving high-dose levosimendan compared with those receiving placebo (P = 0.006 for 0 minutes; P = 0.030 for 15 minutes; P = 0.027 for 50 minutes). Although all values of cardiac output were higher in patients with high-dose compared with patients treated with low-dose levosimendan, only the value measured immediately after administration of drug was significantly higher (P = 0.036). Compared with baseline measurement, all values of cardiac output were significantly higher in patients treated with high-dose levosimendan (P = 0.005 for all). In patients receiving low-dose levosimendan, compared with baseline measurement, cardiac output was significantly higher immediately (P = 0.080), 5 minutes (P = 0.033), and 20 minutes (P = 0.080) after administration of drug (Table 2).

Significant increase in stroke volume and decrease in systemic vascular resistance occurred after high-dose (P = 0.016; P < 0.001) and low-dose levosimendan (P = 0.002; P = 0.001) (Table 3). In contrast to the finding during placebo administration, this increase in stroke volume and decline in systemic vascular resistance was dose related.



Significant increase in left ventricular ejection fraction occurred after both doses of levosimendan. The peak effect on left ventricular ejection fraction was 5 minutes after administration of both doses (Table 3).

No differences in left ventricular end-systolic volume were observed between groups in all measurements. Compared with the baseline measurement, left ventricular end-systolic volume was significantly lower immediately and 5 minutes after administration of levosimendan: for high-dose (P = 0.005; P = 0.013), and for low-dose levosimendan (P = 0.013; P = 0.018) (Table 3).

Back to Top | Article Outline


The past few years have been marked by striking advances in OPCABG because of technical progress in the revascularization of the posterior arteries of the heart. In this type of surgery immediate postoperative left ventricular dysfunction often occured.6 Transient ischemia from the direct coronary occlusion often resulted in myocardial stunning followed by hemodynamic instability.14 Animal studies, which investigated the hemodynamic changes during heart displacement, have indicated that such displacement has its primary deleterious effects on the right heart.15 The main cause of hemodynamic instability (decreases of cardiac output and arterial pressure; increases of left and right atrial pressures; and left and right ventricular end-diastolic pressures) is the disturbance of ventricular diastolic filling by direct ventricular compression.16 Although exposure of the posterior and lateral vessels caused a decrease in left ventricular contractile state these cardiovascular disturbances could be easily corrected by anesthetic intervention, such as fluid loading, low doses of inotropic agents, or boluses of vasoconstrictors.7,8,16,17

There is still not uniformestablished opinion about vasoactive drug support in OPCABG. Catecholamines are not recommended in this setting because they could cause significant hemodynamic disturbances (primarily tachycardia) and increase in myocardial oxygen consumption. Vasoconstrictors are the medications mostly used in these situations, but they could have an effect on cardiac output and on the patency of arterial grafts due to increasing systemic vascular resistance. Because of the unfavorable effects of the previously mentioned medications, levosimendan could be an alternative agent for the maintenance of hemodynamic stability in OPCABG.

Myocardial dysfunction during this procedure is characterized by impaired postischemic systolic function, which is related to decreased myofibril calcium sensitivity.10 Calcium sensitization has been proposed as a novel therapeutic approach by which cardiac performance may be enhanced without predisposition to calcium-induced arrhythmias or an increase in myocardial oxygen demand.18 Clinical data indicate that levosimendan produces hemodynamic improvement in patients with normal preoperative left ventricular function undergoing heart surgery, but the role of myofilament calcium sensitizers has not been described until now in OPCABG.11,12

For preventing systemic hypotension during heart elevation, we administrated 500 mL of 6% hydroxyethylstarch solution and Ringer lactate to maintain a central venous pressure above 10 mm Hg. Most patients had been on beta-blocking agents, long-acting nitrates, or calcium channel blockers prior to surgery, but for all of them, only short-acting medications were given on the day of surgery. Regarding preoperative and operative data, there were no significant differences between groups of patients (Table 1).

Our results confirm the findings of Lilleberg et al11 that a single intravenous loading dose of levosimendan produces hemodynamic changes that are characteristic of the agents with inodilation effect (Table 2, Table 3). After administering levosimendan, heart rate increased in both groups of patients (the highest median value was 80 beats/min), which was a reflection of baroreceptor activation caused by lessening of left ventricular afterload. Absolute heart rate values in patients receiving levosimendan were within clinically acceptable values, but also were lower than in the patients who received the placebo. Furthermore, there was no lowering of mean arterial pressure after administering levosimendan, which could mean that levosimendan has a favorable effect on stroke volume. In contrast to a study of patients with ischemic heart disease associated with left ventricular dysfunction that was accompanied with profound decrease in filling pressures (central venous and pulmonary capillary wedge pressure) in our investigation both doses of levosimendan did not have an effect on pulmonary capillary wedge pressure.19 Because of direct inotropic effect, loading doses of levosimendan produce rapid increases in cardiac output and stroke volume, and these increases were dose related. In patients undergoing CABG, Lilleberg et al11 have shown similar stroke volume responses after administration of 2 different doses of levosimendan. In that study, 10 minutes after administration of 8 μg/kg levosimendan, stroke volume increased maximally for 6 mL/beat, and 30 minutes after administration of 24 μg/kg levosimendan for 10 mL/beat. In our study the stroke volume response on administration of low- and high-dose levosimendan was faster (immediately after the end of drug infusion) and stronger (10 and 18 mL/beat). The moderate decreases in left ventricular afterload after administration of levosimendan are desirable and were not accompanied by a fall in systemic arterial pressures because of successfully optimizing preload before starting levosimendan infusion. Systolic response (left ventricular ejection fraction and end-systolic volume) as measured by transesophageal ultrasound was stronger in patients treated with low-dose compared with patients with high-dose levosimendan. Increasing left ventricular ejection fraction (11% in patients receiving lowdose and 8% in patients receiving high-dose levosimendan) has special importance during exposure of lateral and posterior coronary vessels. The ultrasound measurement of left ventricular end-systolic volume, as a powerful predictor of mortality, shows decreasing end-systolic volume in patients treated with both doses of levosimendan.20

These results must be interpreted within the constraints of several limitations. First of all, patients could receive crystalloids and colloid solution before the beginning OPCABG, and this may prevent filling pressures from decreasing. Decreasing filling pressures after administration of levosimendan is not so prominent in patients with good preoperative left ventricular function. Secondly, as it is known that the effect of levosimendan is prolonged after the completion of infusion, the aim of this study was calculating only the acute hemodynamic effects of levosimendan during the OPCABG. There was no previous study to compare these results. Although the study has been well controlled, because of a low number of patients, these results must be cautiously extrapolated to additional patients who will undergo the same surgical procedure. The present results may not be directly applicable to patients with severely compromised preoperative left ventricular performance, who would possibly have higher benefit from usage of levosimendan.

In conclusion, levosimendan was without the unfavorable effects of previously used inotropic and vasoconstrictor agents, and in this study produced good cardiac performance during and after the OPCABG procedure. The low-dose levosimendan used in this study demonstrated equivalent or a slightly better efficacy than the high-dose, because it produced stronger systolic response and lower peripheral vasodilatation. The low-dose levosimendan could be preferable in this patient population undergoing OPCABG. Levosimendan offers a promising therapeutic option for management with optimal hemodynamic stability and enhances left ventricular performance during and after OPCABG in patients with good preoperative left ventricular function.

Back to Top | Article Outline


The authors wish to thank to all staff in the Clinical Department of Anesthesiology, Reanimatology, and Intensive Care Medicine, Division of Cardiac Anesthesia and Intensive Care, University Hospital Dubrava who helped in our work during the study.

Back to Top | Article Outline


1. Haikala H, Nissinen E, Etemadzadeh E, et al. Troponin C-mediated calcium sensitization induced by levosimendan does not impair relaxation. J Cardiovasc Pharmacol. 1995;25:794-801.
2. Haikala H, Linden I-B. Mechanisms of action of calcium-sensitizing drugs. J Cardiovasc Pharmacol. 1995;26(suppl 1):S10-S19.
3. Rump AF, Acar D, Rösen R, et al. Functional and antiischaemic effects of the phosphodiesterase inhibitor levosimendan in isolated rabbit hearts. Pharmacol Toxicol. 1994;74:244-248.
4. Kersten JR, Montgomery MW, Pagel PS, et al. Levosimendan, a new positive inotropic drug, decreases myocardial infarct size via activation of K(ATP) channels. Anesth Analg. 2000;90:5-11.
5. Sohrabi F, Mispireta LA, Fiocco M, et al. Effects of off-pump coronary artery bypass grafting on patient outcome. J Investig Med. 2003;51:27-31.
6. Chang PP, Sussman MS, Conte JV, et al. Postoperative ventricular function and cardiac enzymes after on-pump versus off-pump CABG surgery. Am J Cardiol. 2002;89:1107-1110.
7. Do QB, Goyer C, Chavanon O, et al. Hemodynamic changes during off-pump CABG surgery. Eur J Cardiothorac Surg. 2002;21:385-390.
8. Torracca L, Schreuder JJ, Quarti A, et al. Acute effects of beating heart coronary surgery on left ventricular performance. Ann Thorac Surg. 2002;74:S1348-S1352.
9. Sunderdiek U, Korbmacher B, Gams E, et al. Myocardial efficiency in stunned myocardium. Comparison of Ca+-sensitization and PDE III-inhibition on energy consumption. Eur J Cardiothorac Surg. 2000;18:83-89.
10. Gao DW, Atar D, Barkx PH, et al. Relationship between intracellular calcium and contractile force in stunned myocardium. Circ Res. 1995;76:1036-1048.
11. Lilleberg J, Nieminen MS, Akkila J, et al. Effects of a new calcium sensitizer, levosimendan, on hemodynamics, coronary blood flow and myocardial substrate utilization early after coronary artery bypass grafting. Eur Heart J. 1998;19:660-668.
12. Nijhawan N, Nicolosi AC, Montgomery MW, et al. Levosimendan enhances cardiac performance after cardiopulmonary bypass: a prospective, randomized placebo-controlled trial. J Cardiovasc Pharmacol. 1999;34:219-228.
13. Dupuis JY, Wang F, Nathan H, et al. The cardiac anesthesia risk evaluation score. Anesthesiology. 2001;94:194-204.
14. Grubitzsch H, Ansorge K, Wollert HG, et al. Stunned myocardium after off-pump coronary artery bypass grafting. Ann Thorac Surg. 2001;71:352-355.
15. Gründeman P, Borst C, van Herwaarden JA, et al. Hemodynamic changes during displacement of the beating heart by the Utrecht Octopus method. Ann Thorac Surg. 1997;63:S88-S92.
16. Mathison M, Edgerton JR, Horswell JL, et al. Analysis of hemodynamic changes during beating heart surgical procedures. Ann Thorac Surg. 2000;70:1355-1360.
17. Resano FG, Stamou SC, Lowery RC, et al. Complete myocardial revascularization on the beating heart with epicardial stabilization: anesthetic consideration. J Cardiothorac Vasc Anesth. 2000;14:534-539.
18. Greenberg B, Borghi C, Perrone S. Pharmacotherapeutic approaches for decompensated heart failure: a role for the calcium sensitiser, levosimendan? Eur J Heart Fail. 2003;5:13-21.
19. Lilleberg J, Sundberg S, Nieminen MS. Dose-range study of a new calcium sensitizer, levosimendan, in patients with left ventricular dysfunction. J Cardiovasc Pharmacol. 1995;26(suppl 1):S63-S69.
20. Schiller NB, Foster E. Analysis of left ventricular systolic function. Heart. 1996;75(suppl 2):17-26.

Levosimendan; calcium sensitizers; coronary artery bypass; thermodilution; echocardiography; transesophageal

© 2004 Lippincott Williams & Wilkins, Inc.