Sustained Hemodynamic Effects of Intravenous Levosimendan
Background— The short-term infusion of levosimendan (Levo) improves hemodynamic function in patients with decompensated heart failure. The metabolites of Levo have a prolonged half-life, and one is hemodynamically active. The goal of this study was to determine whether the hemodynamic effects of Levo are sustained during a long-term infusion and beyond the discontinuation of drug infusion.
Methods and Results— Patients with decompensated heart failure received escalating infusion rates of intravenous Levo (n=98) or placebo (n=48) for 6 hours. At the end of 6 hours, 85 of the Levo-treated patients were continued on open-label drug for a total of 24 hours, at which time they were randomized 1:1 to an additional 24 hours of Levo (n=43) or placebo (n=42).The hemodynamic effects observed at 24 hours were maintained at 48 hours in both the continuation and withdrawal groups and did not differ between groups. Although the plasma concentration of Levo decreased rapidly in the withdrawal group, concentrations of the active metabolite OR-1896 were similar in the continuation and withdrawal groups at 24 hours and increased further (3.5-fold to 4-fold) and to a similar extent in both groups at 48 hours.
Conclusions— The hemodynamic effects of Levo were maintained during a 48-hour continuous infusion and for at least 24 hours after discontinuation of a 24-hour infusion. The active metabolite OR-1896 increased for at least 24 hours after cessation of drug infusion and may account for the prolonged hemodynamic effects of Levo.
Received September 4, 2002; accepted September 30, 2002.
Levosimendan (Levo) is a myocardial calcium sensitizer1,2 and vasodilator3,4 that is being evaluated for the short-term treatment of patients with decompensated heart failure.5–8 Although Levo has an elimination half-life of ≈1 hour, its active metabolite, OR-1896, has a much longer elimination half-life of ≈70 to 80 hours.9 We previously found that the intravenous administration of Levo for 6 hours increased stroke volume and decreased systemic vascular resistance (SVR) and pulmonary artery wedge pressure (PAWP) in a dose-dependent manner in patients with decompensated heart failure.5
The goal of the present study was to determine whether the effects of Levo are sustained during continuous infusion for up to 48 hours. In addition, in view of the long half-life of the active metabolite, we examined whether the hemodynamic effects of Levo are sustained after withdrawal of a 24-hour infusion. Patients who in the prior report5 received Levo during the initial 6 hours of placebo-controlled infusion were continued for an additional 24 hours on open-label Levo at half the prior infusion rate. At 24 hours, half of these patients (n=43) had Levo continued and half (n=42) had Levo withdrawn in a double-blind, randomized manner. Both groups were followed up for an additional 24 hours, during which hemodynamics and plasma levels of Levo and its metabolites were measured.
A total of 146 patients with New York Heart Association functional class III or IV heart failure of either ischemic or nonischemic origin who were admitted to a hospital for the management of decompensated heart failure were recruited between June 1995 and May 1997 in 20 centers in the United States. All patients had a documented left ventricular ejection fraction ≤30% by echocardiogram or radionuclide ventriculogram in the preceding 6 months. To be eligible for the study, patients needed to have a right heart catheter placed for clinical purposes that demonstrated a PAWP ≥15 mm Hg and a cardiac index ≤2.5 L · min−1 · m−2.
The main exclusion criteria were angina-limited exercise, unstable angina or acute myocardial infarction within the previous 8 weeks, obstructive cardiomyopathy, uncorrected primary stenotic valve, history of ventricular flutter or fibrillation or symptomatic ventricular tachycardia, symptomatic primary pulmonary disease, supine systolic blood pressure <85 mm Hg or >200 mm Hg, resting heart rate >115 bpm, serum creatinine >2.5 mg/dL, liver transaminases >2 times the upper limit of normal, uncorrected hypokalemia or hyperkalemia (serum potassium <3.5 mmol/L or >5.5 mmol/L), or treatment with another investigational therapy in the 30 days before study inclusion.
The present study reflects the continuation phase of a double-blind, placebo-controlled study that was reported previously.5 In that study, 98 patients received Levo for 6 hours, and 48 received placebo. After 6 hours, 85 of the patients in the Levo group were continued for a total of 24 hours on open-label Levo. At 24 hours, the patients were randomized 1:1 in a double-blind manner to continue on Levo (continuation group) or placebo (withdrawal group) for an additional 24 hours.
During the initial 6-hour infusion period, Levo was initiated with a bolus of 6 μg/kg followed by a continuous infusion of 0.1 μg · kg−1 · min−1 for 1 hour. Thereafter, at hourly intervals, a 6 μg/kg bolus was administered, and the continuous infusion rate was increased by increments of 0.1 μg · kg−1 · min−1 until a maximum infusion rate of 0.4 μg · kg−1 · min−1 was achieved or a dose-limiting event occurred. At 6 hours, the achieved Levo infusion rate was halved and continued for a total of 24 hours. At 24 hours, patients randomized to the continuation group received Levo at the rate selected for the open-label infusion, and patients randomized to the withdrawal group received a placebo infusion.
The study drug was discontinued temporarily or permanently if any of the following dose-limiting events occurred: (1) a heart rate >130 bpm or an increase in heart rate >15 bpm above baseline for 10 minutes, (2) symptomatic hypotension or a drop in systolic blood pressure to <75 mm Hg, (3) a decrease in PAWP to ≤10 mm Hg, or (4) any adverse event that, in the opinion of the site investigator, required dose modification. If a dose-limiting event occurred, the study drug was discontinued until the event resolved and was then restarted at the next lower dose.
At least 2 hours after insertion of a pulmonary artery catheter, 2 sets of hemodynamic measurements, separated by 10 minutes, were averaged and taken as baseline hemodynamics. Measurements included PAWP, pulmonary artery pressure, right atrial pressure, and cardiac output (thermodilution). Heart rate was determined from the ECG, and blood pressure was determined by arm cuff or intra-arterial monitor. Stroke volume, SVR, and pulmonary vascular resistance were calculated by standard equations. Hemodynamic measurements were obtained at baseline, repeatedly during the first 6 hours as described previously,5 at 23.5 and 24 hours, and at 25, 26, 28, 30, 47.5, and 48 hours. The measurements at 5.5 and 6 hours, 23.5 and 24 hours, and 47.5 and 48 hours were averaged.
Dyspnea and fatigue were evaluated by both the patient and the physician at baseline and at 6, 24, and 48 hours. Patients were asked to grade their symptoms according to a 5-point scale, with 1 denoting the most severe symptoms (extreme dyspnea or fatigue) and 5 denoting no symptoms.
Levo plasma concentrations were determined from blood samples (5 mL) drawn within 10 minutes of the start of the infusion, repeatedly during the first 6 hours, repeatedly between 24 and 30 hours, and at 48 and 54 hours. An additional 5-mL blood sample was collected for the measurement of plasma levels of the metabolites OR-1855 and OR-1896 at the following time points: within 10 minutes of the start of the infusion and at 24, 30, 48, and 54 hours. Plasma concentrations of Levo, OR-1855, and OR-1896 were determined by a liquid chromatography tandem mass spectrometry method. Levo concentrations were analyzed by a previously described method.10 Metabolite concentrations were analyzed with 2 different HPLC-mass spectrometry/mass spectrometry systems (PE Sciex API III and PE Sciex API 300), and the results of the 2 systems were cross-validated. Within-batch precision of the methods varied from 2.2% to 7.2% for OR-1855 and from 1.7% to 6.8% for OR-1896.
An adverse event inquiry was undertaken at baseline and at 6, 24, and 48 hours. Serious adverse events, including mortality, were followed for 14 days after infusion.
Changes in hemodynamics between 6 and 24 hours were evaluated by paired t test. Changes in hemodynamics between 24 and 48 hours were evaluated with an ANOVA model with effects for treatment, center, and treatment-by-center interaction. Symptoms of dyspnea or fatigue were assessed by the Cochran-Mantel-Haenszel test. Differences were considered significant if the null hypothesis could be rejected at the 5% confidence level. Statistical calculations were performed with SAS software. All data are shown as mean±SEM.
The protocol was approved by the Institutional Review Board of each participating center and was performed in accordance with institutional guidelines and the Declaration of Helsinki. All patients gave written informed consent before entering the study.
Patient Demographics and Disposition
The baseline characteristics of the patients in the Levo continuation and withdrawal groups were not different (Table 1). Of the 85 patients randomized to Levo continuation or withdrawal, 16 in the continuation group and 11 in the withdrawal group were discontinued from study medication. The majority of discontinuations in both groups were due to an exaggerated hemodynamic response (ie, decrease in PAWP to ≤10 mm Hg either alone or in combination with an increase in heart rate).
PAWP (Figure 1), mean pulmonary artery pressure, SVR, and systolic and diastolic blood pressures all decreased significantly between 6 and 24 hours (Table 2). Heart rate, stroke volume (Figure 2), and cardiac index were unchanged between 6 and 24 hours (Table 2).
Before randomization at 24 hours, the hemodynamics were similar in the Levo continuation and withdrawal groups (Table 3). At 48 hours, the hemodynamics were similar in the Levo withdrawal and continuation groups and generally unchanged from hemodynamics at 24 hours (Table 3), with the exception of heart rate, which was higher (P<0.001) in both groups at 48 hours (versus 24 hours), and SVR, which was lower (P=0.071) in both groups at 48 hours (versus 24 hours).
At 48 hours, there were no significant differences in symptoms of dyspnea or fatigue between the Levo continuation and withdrawal groups.
Plasma Concentrations of Levo and Its Metabolites
The mean dose of Levo was 0.26±0.08 μg · kg−1 · min−1 at 6 hours, which yielded plasma concentrations of Levo at 6 hours of ≈120 ng/mL in both the continuation and withdrawal groups (Figure 3). At 24 hours, the mean dose of Levo was 0.14±0.08 and 0.18±0.09 μg · kg−1 · min−1 in the Levo continuation and withdrawal groups, respectively, which yielded Levo plasma concentrations that averaged 63±35 and 70±46 ng/mL in these groups, respectively. Between 24 and 48 hours, plasma levels remained constant in the continuation group, whereas they decreased rapidly in the discontinuation group. Plasma levels of the metabolites OR-1896 and OR-1855 increased to a similar extent (3- to 4-fold) in both the continuation and withdrawal groups from 24 to 54 hours (Figure 4).
The frequency and nature of adverse events were similar in the Levo continuation and withdrawal groups, except that the Levo continuation group had more frequent reports of headache (21% versus 14%), nausea (7% versus 2%), and vomiting (5% versus 0%) than the withdrawal group. Hypotension occurred more frequently in the withdrawal group (7% versus 5%). One case of ventricular tachycardia occurred in the withdrawal group.
No deaths occurred during the 48-hour infusion period. During the 14-day follow-up, 5 deaths occurred, 3 in patients randomized to placebo for the initial 6-hour infusion period (6.3%) and 2 in patients randomized to Levo (2.0%).
We previously demonstrated that a short-term 6-hour infusion of Levo to patients with decompensated heart failure due to left ventricular systolic dysfunction results in beneficial hemodynamic effects, with decreases in left and right heart filling pressures and SVR and increases in stroke volume and cardiac index.5 We now report 2 new findings with regard to the long-term infusion of Levo for 24 to 48 hours. First, the hemodynamic effects of Levo appear to be sustained for at least 48 hours with continuous infusion. Second, after a 24-hour infusion, the hemodynamic effects are sustained for at least another 24 hours after withdrawal of the drug. This latter observation appears to reflect accumulation of the long-lived, active metabolite OR-1896.
The present report reflects the 24- and 48-hour data in patients who received a placebo-controlled infusion of escalating doses of Levo over a 6-hour period.5 At the end of the 6-hour infusion, there were reductions in right atrial pressure, PAWP, pulmonary artery pressure, and SVR and reciprocal increases in stroke volume and cardiac index. In the patients who initially received Levo, the infusion rate was reduced by 50%, and the infusion was continued for an additional 24 hours. At 24 hours, the initial hemodynamic effects seen at 6 hours were sustained or augmented, with further reductions in systolic and diastolic arterial pressure, PAWP, and SVR. In the patients who were then randomized to Levo for an additional 24 hours, the hemodynamic effects observed at 24 hours persisted.
These observations suggest that the acute hemodynamic effects of Levo are maintained for at least 48 hours and thus, that clinically important tolerance does not occur. This conclusion is limited by the lack of a placebo group beyond 6 hours; therefore, it is possible that other changes in therapy contributed to the hemodynamic improvement observed at 24 and 48 hours.
The hemodynamic effects of Levo were sustained after double-blind withdrawal for 24 hours. These persistent effects cannot be attributed to Levo, because there was a rapid fall in the plasma level of Levo in the first few hours after drug withdrawal, consistent with its known elimination half-life of ≈1 hour. However, at 24 hours, the plasma levels of 2 metabolites of Levo, OR-1896 and OR-1855, were measurable in the plasma and continued to increase between 24 and 48 hours in both the continuation and withdrawal groups. Both metabolites are known to have long elimination half-lives on the order of 70 to 80 hours,9 which would account for the continued increase in these metabolite levels even up to 24 hours after cessation of the Levo infusion. Likewise, in a previous study,9 we found that metabolite levels continued to increase for at least 24 hours after withdrawal of levosimendan.
In preclinical studies, OR-1896 has been shown to have hemodynamic effects similar to those of Levo,11–13 whereas OR-1855 appears to be much less active. OR-1896 and Levo possessed similar positive inotropic effects at the same concentrations in guinea pig hearts.11 Because Levo is 95% to 98% bound to plasma proteins in humans,14 whereas OR-1896 is only ≈40% bound to plasma proteins (data not published), the free fraction of OR-1896 is 12 to 30 times higher than that of a comparable concentration of Levo. Therefore, although the concentration of OR-1896 at 48 hours (≈7.5 ng/mL) was less than the concentration of Levo (≈60 ng/mL), the concentration of free OR-1896 may be comparable to a Levo concentration in excess of 90 to 200 ng/mL, a concentration range that is consistent with the demonstrated hemodynamic effects of Levo in the first 6 hours of infusion (Figure 3). It is also possible that accumulation of the metabolite contributed to the greater hemodynamic effects at 24 versus 6 hours. However, because this phase of the trial was not controlled, we cannot exclude indirect contributions from other therapies such as diuretics and/or an accentuated diuretic response secondary to the hemodynamic effects of Levo.
On the basis of these data, we cannot predict the duration of the hemodynamic effects after discontinuation of Levo infusion. However, it appears that after a 24-hour infusion, the hemodynamic effects are maintained for at least an additional 24 hours. Given the pharmacokinetics of OR-1896, it is likely that hemodynamic effects would be sustained for up to several days. The duration of effects would also depend on the infusion rate and duration of the drug administration. In a noninvasive study in patients with severe heart failure, heart rate was increased for at least 14 days after a 7-day Levo infusion was stopped.9 Additional studies will be needed to determine the duration and nature of hemodynamic effects after drug withdrawal and their relationship to metabolite levels.
A limitation of the present study is that 27 of 85 patients who entered the 24-hour double-blind withdrawal phase discontinued the study drug. However, in the majority of these cases (16 of 27), the reason involved a decrease in PAWP to ≤10 mm Hg or an excessive decrease in blood pressure, which makes it unlikely that withdrawals biased the data in favor of a sustained effect.
These results have several potentially important implications for the clinical use of Levo. First, they suggest that the hemodynamic effects of the drug are not hampered by tolerance. Second, they indicate that in most cases, the infusion of Levo need not be extended beyond 24 hours. Third, they suggest that prolonged infusions beyond 24 to 48 hours could lead to excessive metabolite accumulation, which might be associated with adverse events. The present data do not directly address the optimal duration of infusion. However, it is likely that an infusion of much less than 24 hours would not result in sustained effects because of insufficient accumulation of the metabolite. Conversely, it appears that a 24-hour infusion is safe, at least relative to therapy with dobutamine, and may be associated with a lower mortality rate.7 The unique pharmacokinetic profile of Levo should be considered in the design of future studies with the drug. Potentially, the sustained effects of Levo may be used to clinical advantage in the management of patients with decompensated heart failure by providing extended hemodynamic support.
Dr Kivikko is a current employee of Orion Pharma, which manufactures levosimendan. Dr Lehtonen is a former employee of Orion Pharma; he is also an inventor in several patents regarding the formulations and use of levosimendan and its metabolites. Dr Colucci is a consultant for Orion Pharma; he has been an investigator in several trials on levosimendan and has received study grants for this purpose. Dr Colucci has also made presentations in several meetings sponsored by Orion Pharma.
↵*The complete list of Study Investigators has been published previously (Circulation. 2002;102:2222–2227).
Pollesello P, Ovaska M, Kaivola J, et al. Binding of a new Ca2+ sensitizer, levosimendan, to recombinant human cardiac troponin C: a molecular modelling, fluorescence probe, and proton nuclear magnetic resonance study. J Biol Chem. 1994; 269: 28584–28590.
Slawsky MT, Colucci WS, Gottlieb SS, et al, for the Study Investigators. Acute hemodynamic and clinical effects of levosimendan in patients with severe heart failure. Circulation. 2000; 102: 2222–2227.
Moiseyev VS, Poder P, Andrejevs N, et al. Safety and efficacy of a novel calcium sensitizer, levosimendan, in patients with left ventricular failure due to an acute myocardial infarction: a randomized, placebo-controlled, double-blind study (RUSSLAN). Eur Heart J. 2002; 23: 1422–1432.
Kristof E, Szigeti G, Papp Z, et al. Cardiac responses to calcium sensitizers and isoproterenol in intact guinea pig hearts: effects on cyclic AMP levels, protein phosphorylation, myoplasmic calcium concentration, and left ventricular function. Ann N Y Acad Sci. 1998; 853: 316–319.