European Heart Journal (2002) 23, 1422–1432

doi:10.1053/euhj.2001.3158, available online at http://www.idealibrary.com on

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)
V. S. Moiseyev1, P. Põder2, N. Andrejevs3, M. Y. Ruda4, A. P. Golikov5,
L. B. Lazebnik6, Z. D. Kobalava1, L. A. Lehtonen7, T. Laine2, M. S. Nieminen8 and
K. I. Lie9 on behalf of RUSSLAN Study Investigators
1
Hospital No. 64, Russian University of People’s Friendship, Moscow, Russia; 2Orion Pharma, Research Center,
Espoo, Finland; 3P. Stradin’s Research and Teaching Hospital, Latvian Medical Academy, Riga, Latvia; 4A.L.
Myasnikov Institute of Cardiology, Cardiology Research Center, Moscow, Russia; 5Cardiology Clinic, Sklifosovsky
Research Institute for Emergency Medicine, Moscow, Russia; 6Department of Gerontology and Geriatrics of
Russian Medical Academy for Postgraduate Training, Moscow, Russia; 7Department of Clinical Pharmacology,
University of Helsinki, Finland; 8Helsinki University Central Hospital, Cardiology Division, Helsinki, Finland;
9
Academic Medical Center, University of Amsterdam, The Netherlands

Aims To evaluate the safety and efficacy of levosimendan P=0·033) and over 24 h (4·0% vs 8·8%; P=0·044). Mor-
in patients with left ventricular failure complicating acute tality was lower with levosimendan compared with placebo
myocardial infarction. at 14 days (11·7% vs 19·6%; hazard ratio 0·56 [95% CI
0·33–0·95]; P=0·031) and the reduction was maintained at
Methods and Results Levosimendan at different doses the 180-day retrospective follow-up (22·6% vs 31·4%; 0·67
(0·1–0·4
g . kg
1 . min
1) or placebo were administered [0·45-1·00], P=0·053).
intravenously for 6 h to 504 patients in a randomised,
placebo-controlled, double-blind study. The primary Conclusions Levosimendan at doses 0·1–
end-point was hypotension or myocardial ischaemia of 0·2
g . kg
1 . min
1 did not induce hypotension or is-
clinical significance adjudicated by an independent Safety chaemia and reduced the risk of worsening heart failure and
Committee. Secondary end-points included risk of death death in patients with left ventricular failure complicating
and worsening heart failure, symptoms of heart failure and acute myocardial infarction.
all-cause mortality. The incidence of ischaemia and/or (Eur Heart J, 2002; 23: 1422–1432, doi:10.1053/euhj.2001.
hypotension was similar in all treatment groups (P=0·319). 3158)
A higher frequency of ischaemia and/or hypotension was  2002 The European Society of Cardiology. Published by
only seen in the highest levosimendan dose group. Elsevier Science Ltd. All rights reserved.
Levosimendan-treated patients experienced lower risk of
death and worsening heart failure than patients receiving Key Words: Left ventricular failure, myocardial infarction,
placebo, during both the 6 h infusion (2·0% vs 5·9%; levosimendan, hypotension, ischaemia, mortality.

Manuscript submitted 4 December 2001, and accepted 12 December 2001.

This study was presented in part at the 72nd Scientific Sessions (7–10 November 1999) of American Heart Association in Atlanta, Georgia,
U.S.A.
*All investigators and study sites are listed in the Appendix.
Correspondence: Professor Markku S. Nieminen, MD, PhD, FACC, Helsinki University Central Hospital, Department of Internal
Medicine, Cardiology Division, 00029, Haartmaninkatu 4, Helsinki, Finland.

0195-668X/02/$35.00  2002 The European Society of Cardiology. Published by Elsevier Science Ltd. All rights reserved.
 The RUSSLAN study 1423

Introduction The protocol and any relevant amendments were

reviewed and approved by local Ethics Committees.
The prognosis of patients with heart failure complicating Written informed consent was obtained from all
acute myocardial infarction remains poor, despite cur- patients. The first patient was randomized to the study
rent standard therapy with diuretics, vasodilators and on 13 June 1996. The final 180-day mortality follow-up
angiotensin-converting-enzyme (ACE) inhibitors. Pre- of the last patient was completed on 27 April 2000.
vious studies indicate a 20–40% annual mortality rate
in this patient population, suggesting the need for
additional therapeutic options[1–3]. Positive inotropic
agents, including dobutamine and phosphodiesterase Study population
inhibitors, have also been studied in this patient popu-
lation, but the data regarding their efficacy and safety is Study patients were recruited at 21 centres in Russia and
still limited[4–6]. In addition, the results of several clinical Latvia. Inclusion criteria were: acute myocardial infarc-
trials with various positive inotropic drugs in patients tion (according to World Health Organization criteria)
with heart failure have shown increased mortality[7–10]. during the previous 5 days; evidence of left ventricular
Levosimendan is a novel drug developed for the failure on chest X-ray (pulmonary venous congestion or
treatment of decompensated heart failure. Levosi- pulmonary oedema); and a clinical need for inotropic
mendan is a calcium sensitizer that increases the contrac- therapy on the basis of symptomatic heart failure despite
tile force of the myocardium by enhancing the sensitivity conventional therapy. Exclusion criteria comprised:
of myofilaments to calcium without increasing intra- right ventricular infarction; systolic blood pressure
cellular calcium concentration at therapeutic doses[11–13]; <90 mmHg; sustained ventricular tachycardia or fre-
the risk of cardiac arrhythmias is similar to placebo[14]. quent ventricular non-sustained tachycardias not related
It improves cardiac contractility without increasing oxy- to thrombolysis; atrial fibrillation with a rapid ventricu-
gen consumption[15,16] and, in theory, should not induce lar response; immediate need for cardiac pacing, percu-
ischaemic episodes. Levosimendan has also vasodilatory taneous transluminal coronary angioplasty, or coronary
and antiischaemic properties attributable to its effects on artery bypass grafting; myocardial rupture, or severe
adenosine triphosphate (ATP)-dependent potassium mitral valve insufficiency; cardiac tamponade; use of
channels[17–20]. beta-adrenergic agonists within 30 min of the start of the
Previous clinical trials have established the favourable study; adult respiratory distress syndrome; septic shock;
haemodynamic effects of intravenously administered history of moderate or severe renal failure (serum cre-
levosimendan in patients with moderate or severe heart atinine >250
mol . l
1); clinically relevant hepatic fail-
failure[21–23]. Therapeutic options that improve cardiac ure; allergy requiring medication; and participation in
function without detrimental effects are limited in heart another clinical trial within 1 month before study entry.
failure complicating acute myocardial infarction. Lev- Women with childbearing potential and patients with
osimendan may confer clinical benefits in this setting due agonal status were ineligible for the study.
to its lack of detrimental effects on myocardial oxygen
consumption. A previous open-label dose-controlled
study with three different bolus doses of levosimendan
has demonstrated the haemodynamic efficacy of levosi- Study design
mendan in patients with acute myocardial infarction[24].
However, the safety and efficacy of longer infusions had This was a randomized, placebo-controlled and double-
to be addressed in a large-scale placebo-controlled clini- blind study. A computer-generated randomization
cal trial in acute myocardial infarction patients. This schedule, based on permuted blocks and balanced within
study was therefore conducted to assess the short- and each centre, was used to allocate patients to placebo or
the long-term safety and efficacy of different 6-h infu- one of four dose regimens of levosimendan (SIMDAX,
sions of levosimendan in patients with decompensated Orion Pharma, Finland): 6
g . kg
1 loading dose+
heart failure complicating acute myocardial infarction 0·1
g . kg
1 . min
1 continuous infusion; 12
g . kg
1
compared with placebo. This is the first clinical study to loading dose+0·2
g . kg
1 . min
1 continuous infu-
address the safety and efficacy of a calcium-sensitizing sion; 24
g . kg
1 loading dose+ 0·2
g . kg
1 . min
1
drug in this setting. continuous infusion; 24
g . kg
1 loading dose+
0·4
g . kg
1 . min
1 continuous infusion (Fig. 1). The
loading dose was infused over a period of 10 min, and
the continuous infusion was maintained for 5 h and
Methods 50 min. The placebo was identical in appearance to the
active drug. All the formulations and vials were made to
The RUSSLAN study (Randomised stUdy on Safety look identical and had either no active ingredient or
and effectivenesS of Levosimendan in patients with left different amounts of levosimendan. The volumes infused
ventricular failure due to an Acute myocardial iNfarct) into the patients of the five different treatment arms were
was conducted according to the Declaration of Helsinki also identical. Study medications were introduced via a
of the World Medical Assembly and its amendments. peripheral vein using a calibrated infusion pump.

Eur Heart J, Vol. 23, issue 18, September 2002

1424 V. S. Moiseyev et al.

504 patients
randomized

Placebo Levosimendan Levosimendan Levosimendan Levosimendan

102 patients 6 µg.kg–1 + 12 µg.kg–1 + 24 µg.kg–1 + 24 µg.kg–1 +
0·1 µg.kg–1 min–1 0·2 µg.kg–1 min–1 0·2 µg.kg–1 min–1 0·4 µg.kg–1 min–1
103 patients 100 patients 99 patients 100 patients

8 withdrawals (7·8%) 10 withdrawals (9·7%) 6 withdrawals (6·0%) 5 withdrawals (5·1%) 12 withdrawals (12·0%)

94 completed (92·2%) 93 completed (90·3%) 94 completed (94·0%) 94 completed (94·9%) 88 completed (88·0%)

Figure 1 Trial profile.

Investigators were mandated to stop the infusion of 1 mm in a 12-lead ECG. Clinically significant hypoten-
the study medication if patients experienced: sympto- sion and/or ischaemia, reported by the investigator, were
matic hypotension, heart rate >130 beats . min
1 sus- evaluated by the Safety Committee. For the evaluation,
tained for 10 min, or any serious adverse event. Baseline investigators provided the committee with case record
assessments included blood pressure (measured with a forms, ECGs and copies of patients’ hospital files.
sphygmomanometer or with an automatic blood press- Committee meetings were held after every 100 patients
ure measuring device), heart rate (determined from the recruited. Members of the Committee were unaware of
ECG) and respiratory rate. Baseline assessments of patient treatment allocation at the time of assessment.
dyspnoea, fatigue, anginal pain and physical signs of Secondary end-points included the combined risk of
heart failure were also made. Patients were permitted to death and worsening heart failure during the first 6 and
receive all appropriate therapy for the management of 24 h after the start of the infusion, a change in dyspnoea
both acute myocardial infarction and heart failure. and fatigue at the end of the infusion and death for any
Patients developing persistent hypotension or reason over 14 days after the start of the infusion.
refractory heart failure during infusion, were permitted Patients were considered to have worsening heart failure
intravenous dopamine (3–9
g . kg
1 . min
1). if they experienced onset or worsening of any following
Throughout the 6-h infusion period, patients were conditions: dyspnoea, fatigue, pulmonary congestion or
assessed for hypotension or myocardial ischaemia of oedema, heart failure or cardiogenic shock. The severity
clinical significance, symptoms of heart failure, haemo- of dyspnoea and fatigue was assessed both by the patient
dynamics, urinary output and adverse events. In ad- and the investigator before and after the infusion as
dition to spontaneous reporting, an adverse event a score from 1 to 4, where 1 represented none and
inquiry was undertaken by an investigator at the end of 4 disabling. An increase in score was categorized
the infusion and at 24 h after the start of the infusion. as ‘worse’, a decrease as ‘better’, and no ‘change’ as
Patient survival was evaluated at 14 days following the ‘unchanged’. In the worst-rank symptom analysis,
start of the infusion. An additional 180-day mortality patients were considered to be worse if, in addition to
follow-up was conducted after the end of the study. The the changes in actual dyspnoea and fatigue scores during
information for the 180-day follow-up was obtained the 6 h study infusion, they (1) died; (2) developed
from Official Inhabitant Registries and patients’ hospital worsening heart failure; or (3) received a new drug for
files. Confirmation of survival was obtained through the treatment of heart failure.
telephone contact with the patients.
Blood samples, drawn before the start of infusion and
immediately after infusion, were used to determine Statistical analysis
serum creatine kinase-MB levels. A chest X-ray was
repeated within 12–30 h after the start of the infusion. Statistical analyses were based on the intention-to-treat
principle, performed at a two-sided 0·05 level of signifi-
cance. Analyses were carried out using SAS 6.12 statisti-
End-points cal software (SAS Institute Inc., Cary, NC, US) for
Windows.
The primary end-point was the proportion of patients In a separate pilot study, evaluating the effects of
developing hypotension or ischaemia of clinical signifi- three bolus infusions of levosimendan in patients with
cance adjudicated by an independent safety committee. acute myocardial infarction, levosimendan improved
Clinically significant hypotension was defined as: (1) haemodynamics, but did not cause myocardial
symptomatic hypotension (obligatory) or (2) an asymp- ischaemia in Holter-recordings and did not cause hypo-
tomatic drop in systolic blood pressure of more than tension[24]. However, the regulatory authorities (FDA)
10 mmHg (at the discretion of the investigator). Clini- required that a population of about 500 patients would
cally significant ischaemia was defined as: (1) aggrava- be required to assess the risk/benefit ratio of levosi-
tion or a new onset of anginal pain; or (2) further mendan in patients with acute myocardial infarction.
depression or elevation of the ST-segment by more than A placebo-controlled study with 6 h infusions was

Eur Heart J, Vol. 23, issue 18, September 2002

 The RUSSLAN study 1425

therefore conducted and the incidence of clinically sig- dyspnoea and about 90% had pulmonary rales at rest
nificant hypotension and ischaemia in the placebo group despite previous treatment).
represents the spontaneous variation of the primary
end-point in this patient population.
Baseline characteristics were summarized using ap-
propriate descriptive statistics; values for each character- Primary end-point
istic were compared among the five treatment groups
The safety committee considered that 65 patients had
using analysis of variance (ANOVA) with effects for
clinically significant ischaemia or hypotension (Table 3).
treatment, centre and treatment by centre interaction or
No significant differences among the five treatment
the non-parametric Cochran–Mantel–Haenszel (CMH)
groups were observed in the proportion of patients who
test, controlling for centre. In a primary analysis for a
experienced the primary end-point (P=0·319). When all
primary end-point, the differences between treatment
four levosimendan groups were combined and compared
groups in the proportions of patients experiencing clini-
with placebo, the proportions of patients who experi-
cally significant ischaemic and/or hypotensive events
enced clinically significant hypotension and/or ischaemia
were tested using the CMH row means score test,
in the placebo and levosimendan groups were similar
controlling for centre. The same analysis was tested
(10·8% vs 13·4%, respectively, P=0·456). There was,
comparing the placebo group and the pooled levosi-
however, a weak relationship between the dose of lev-
mendan group. The relationship between dose and
osimendan and the risk of hypotension and/or ischaemia
frequency of event(s) was evaluated using the CMH
(P=0·054), which was attributable to a higher frequency
non-zero correlation test; the effects in each treatment
(19·0%) of ischaemia and hypotension among patients
arm were weighed according to the total quantity of
who received the highest levosimendan infusion rate
drug (in mg . kg
1) due over the 6 h infusion period.
(24
g . kg
1 +0·4
g . kg
1 . min
1).
The combined risk of death and worsening heart
failure were expressed using a time-to-event model. The
log-rank test was used for detecting differences between
placebo and pooled levosimendan groups. Cumulative Secondary end-points
survival curves for placebo and pooled levosimendan
groups were constructed by the Kaplan–Meier method Death and worsening heart failure
and the differences between the curves were tested for The combined risk of death and worsening heart failure
significance using the Cox proportional hazards model. was lower among patients treated with levosimendan
Survival time in the model was calculated as the differ- than among patients receiving placebo during both the
ence in days from the start of infusion to the event or to 6 h infusion period (2·0% vs 5·9%, respectively;
the last follow-up date. The relationship between dose P=0·033), and 24 h after the start of infusion (4·0% vs
and frequency of event(s) was evaluated using the CMH 8·8%, respectively; P=0·044) (Fig. 2). There was no
non-zero correlation test, controlling for centre. relationship between the dose of levosimendan and the
Changes in overall clinical status, symptoms of heart combined risk of death and worsening heart failure
failure, anginal pain, jugular venous distension, periph- during both the 6 h infusion and 24 h after start of
eral oedema, urinary output, pulmonary congestion and infusion (Table 4). All-cause mortality among
creatine kinase-MB values were evaluated using levosimendan-treated patients was significantly lower
ANOVA methods or by use of the CMH row mean than with placebo for the 14-day period after the start of
scores test, controlling for centre. The frequency of the treatment (11·7% vs 19·6%, respectively; 0·56 [95%
adverse events in the five treatment groups was com- CI 0·33–0·95]; P=0·031); this difference was also seen
pared using Fisher’s exact test. Dose-relations of adverse when the follow-up was extended to 180 days (22·6% vs
events were tested using the CMH non-zero correlation 31·4%, respectively; 0·67 [0·45-1·00]; P=0·053) (Fig. 3).
test, controlling for centre. There was no relationship between the dose of levosi-
mendan and all-cause mortality during both the 14-day
and 180-day follow-up (Table 4).

Results Symptoms of heart failure

There were no differences among the treatment groups
Patient characteristics with respect to changes in dyspnoea or fatigue scores. In
the worst rank analysis, however, patients treated with
The five treatment groups were well matched regarding levosimendan were judged by the investigators to have
baseline characteristics (Table 1) and concomitant medi- experienced worsening dyspnoea less frequently than
cations (Table 2). Diabetes mellitus was more prevalent those receiving placebo (10·8% vs 17·0%, respectively,
among levosimendan patients and cerebrovascular dis- P=0·042). This treatment difference was also reflected in
ease in the placebo group. All patients had pulmonary patient self-assessments (11·0% vs 16·7%, respectively,
congestion or oedema on the chest X-ray despite con- P=0·056). In the worst rank analysis patients treated
ventional therapy (nitrates and diuretics) and were with levosimendan experienced also worsening fatigue
highly symptomatic (more than 90% of patients had less frequently than patients receiving placebo in both

Eur Heart J, Vol. 23, issue 18, September 2002

 1426
V. S. Moiseyev et al.

Table 1 Baseline characteristics

Levosimendan Levosimendan Levosimendan Levosimendan

Placebo 6
g . kg
1 + 12
g . kg
1 + 24
g . kg
1 + 24
g . kg
1 +
Variable P-value*
(n=102) 0·1
g . kg
1 . min
1 0·2
g . kg
1 . min
1 0·2
g . kg
1 . min
1 0·4
g . kg
1 . min
1
(n=103) (n=100) (n=99) (n=100)

Eur Heart J, Vol. 23, issue 18, September 2002

Age (years) 6811 6712 6810 6611 6711 0·779
Sex (M/F) 57/45 52/51 44/56 53/46 54/46 0·968
Weight (kg) 7612 7914 7815 7612 7813 0·385
Time since AMI (days) 1·6 1·2 1·91·2 1·91·3 1·81·2 2·01·3 0·148
Anginal pain (%) 3·9 5·8 4·0 7·1 5·1 0·803
Dyspnoea (%) 97·1 89·3 94·0 93·9 94·9 0·228
Fatigue (%) 54·9 49·5 52·0 53·5 49·5 0·816
Pulmonary rales (%) 89·2 89·3 86·0 89·9 92·0 0·744
S3 with tachycardia (%) 15·7 11·7 12·0 12·1 9·1 0·633
Signs of poor tissue perfusion (%) 61·8 58·3 55·0 55·6 57·0 0·545
Previous MI (%) 21·6 32·0 31·0 30·3 20·0 0·145
Hypertension (%) 60·8 73·8 72·0 64·6 66·0 0·256
Atrial fibrillation (%) 2·9 6·8 2·0 7·2 10·0 0·097
Peripheral vascular disease (%) 6·9 4·9 9·0 4·0 6·0 0·588
Gastrointestinal disease (%) 30·4 23·3 19·0 25·3 27·0 0·333
Pulmonary disease (%) 26·5 28·2 26·0 19·2 20·0 0·348
Diabetes mellitus (%) 9·8 16·5 26·0 22·0 22·0 0·038
Cerebrovascular disease (%) 17·6 7·8 20·0 10·1 16·0 0·033

*Comparison of levosimendan groups versus placebo based on CMH test (except ANOVA for age and weight).
 Table 2 Baseline and concomitant medication during 24 h after start of infusion

Levosimendan Levosimendan Levosimendan Levosimendan

Placebo 6
g . kg
1 + 12
g . kg
1 + 24
g . kg
1 + 24
g . kg
1 +
P-value*
(n=102) 0·1
g . kg
1 . min
1 0·2
g . kg
1 . min
1 0·2
g . kg
1 . min
1 0·4
g . kg
1 . min
1
(n=103) (n=100) (n=99) (n=99)

Thrombolytics (%) 15·7 17·5 22·0 14·1 16·2 0·487

Cardiac glycosides (%) 12·6 10·8 13·0 20·2 9·1 0·146
Dopamine (%) 13·6 14·7 8·0 9·1 6·1 0·226
Intravenous inotropes 6·8 9·8 7·0 7·1 4·0 0·624
(others than cardiac glycosides and dopamine) (%)
Diuretics (%) 69·9 75·5 76·0 75·8 74·7 0·883
ACE-inhibitors (%) 44·7 46·1 48·0 47·5 48·5 0·991
Beta-blockers (%) 40·8 42·2 38·0 32·3 42·4 0·520
Calcium channel blockers (%) 14·6 10·8 12·0 14·1 14·1 0·881
Nitrates (%) 94·2 97·1 98·0 97·0 96·0 0·822
Antiarrhythmics (%) 30·1 22·5 29·0 20·2 26·3 0·385
Analgesics (%) 79·6 84·3 79·0 80·8 81·8 0·893
Acetylsalicylic acid (%) 88·3 90·2 88·0 85·9 86·9 0·920
Heparin/heparin analogues (%) 77·7 88·2 84·0 84·8 87·9 0·183

*CMH statistics (row mean score difference).

The RUSSLAN study

Eur Heart J, Vol. 23, issue 18, September 2002

1427
1428 V. S. Moiseyev et al.

Table 3 Incidence of clinically significant ischaemia or hypotension, adjudicated by the Safety Committee

Levosimendan Levosimendan Levosimendan Levosimendan

Placebo 6
g . kg
1 + 12
g . kg
1 + 24
g . kg
1 + 24
g . kg
1 +

1
1
1
1
1
1
(n=102) 0·1
g . kg . min 0·2
g . kg . min 0·2
g . kg . min 0·4
g . kg
1 . min
1
(n=103) (n=100) (n=99) (n=100)

Hypotension only 5 (4·9%) 7 (6·8%) 4 (4·0%) 5 (5·1%) 9 (9·0%)

Ischaemia only 4 (3·9%) 0 (0·0%) 7 (7·0%) 5 (5·1%) 8 (8·0%)
Hypotension and ischaemia 2 (2·0%) 4 (3·9%) 1 (1·0%) 2 (2·0%) 2 (2·0%)
Hypotension and/or ischaemia*† 11 (10·8%) 11 (10·7%) 12 (12·0%) 12 (12·1%) 19 (19·0%)§

*P=0·319 for comparison between all treatment groups (CMH row means score test).
†P=0·456 for comparison of combined levosimendan groups versus placebo (CMH row means score test).
§P=0·054 for dose–response relation (CMH non-zero correlation test).

investigator (10·6% vs 17·0%, respectively, P=0·047) Fewer patients treated with levosimendan required a
and patient assessments (10·8% vs 16·7%, respectively, new vasodilator, diuretic or positive inotropic drug for
P=0·045). the treatment of heart failure than with placebo (7·2% vs
13·7%, respectively, P=0·003) during the 6 h infusion.
Other indices of clinical status
There were no significant differences among the five
treatment groups as regards change of overall clinical
status, anginal pain, jugular venous distension, periph-
Haemodynamic responses
eral oedema, urinary output and pulmonary congestion. Levosimendan produced dose-dependent decreases in
systolic and diastolic blood pressure and increases in
100
heart rate at the end of the 6 h treatment period (Table
Levosimendan
Patients without event (%)

5). The effect on blood pressure was most marked at the

95 highest dose (24
g . kg
1 +0·4
g . kg
1 . min
1)
90 Placebo studied, where placebo-corrected decreases of up to
85 6 mmHg were noted. The effect on heart rate was small
at the lower doses, but the highest dose produced a
80
placebo-corrected increase of 11 beats . min
1. The
75 respiratory rate was unaffected.
70
65
0 2 4 6 8 10 12 14 16 18 20 22 24 Adverse events
Time (h)
During the 6 h infusion period, adverse events were
Figure 2 Combined risk of death and worsening heart
recorded in 23·4% of patients receiving levosimendan
failure during the first 24 h after start of infusion. The
combined risk of death and worsening heart failure was compared with 17·6% in the placebo group (P=0·233).
2·0% in levosimendan group and 5·9% in placebo group The only statistically significant differences between
during the 6–h infusion period (P=0·033, log-rank) and levosimendan and placebo were observed in the
4·0% in levosimendan group and 8·8% in placebo group frequencies of sinus tachycardia and myocardial
(P=0·044) during the first 24 h after start of infusion. rupture (Table 6). Sinus tachycardia was most common

Table 4 Incidence of death and worsening heart failure and all-cause mortality in all dose groups

Levosimendan Levosimendan Levosimendan Levosimendan

Placebo 6
g . kg
1 + 12
g . kg
1 + 24
g . kg
1 + 24
g . kg
1 +
End-point 0·1
g . kg
1 . min
1 0·2
g . kg
1 . min
1 0·2
g . kg
1 . min
1 0·4
g . kg
1 . min
1 P-value*
(n=102)
(n=103) (n=100) (n=99) (n=100)

Death or worsening 5·9 2·9 2·0 1·0 2·0 0·094

heart failure at 6 h (%)
Death or worsening 8·8 5·8 3·0 3·0 4·0 0·089
heart failure at 24 h (%)
Mortality at 6 h (%) 3·9 1·9 1·0 0·0 0·0 0·015
Mortality at 24 h (%) 4·9 3·9 1·0 1·0 2·0 0·127
Mortality at 14 days (%) 19·6 12·6 10·0 13·1 11·0 0·112
Mortality at 180 days (%) 31·4 26·2 16·0 27·3 21·0 0·088

*For dose-relation (CMH non-zero correlation test).

Eur Heart J, Vol. 23, issue 18, September 2002

 The RUSSLAN study 1429

leading to withdrawal from study infusion, eight

were receiving placebo (7·8%) and 33 were receiving
levosimendan (8·2%).
During the first 24 h after the start of the infusion,
adverse events were recorded in 29·4% of patients receiv-
ing levosimendan compared with 26·5% in the placebo
group (P=0·625). The incidence of adverse events
was highest in the highest levosimendan dose
(24
g . kg
1 +0·4
g . kg
1 . min
1) group (36·0%).

Discussion
Figure 3 Overall survival in 180 days after start of
infusion. The mortality rates at 14 days were 11·7% in the This study demonstrates that levosimendan is both well
levosimendan group and 19·6% in the placebo group
tolerated and effective in patients with left ventricular
(P=0·031, Cox Proportional Hazards); at 180 days the
rates were 22·6% and 31·4%, respectively (P=0·053). failure complicating acute myocardial infarction. The
patients enrolled into the study were highly sympto-
matic. Of the 504 randomized patients, nearly all had
in the highest levosimendan dose (24
g . kg
1 + either dyspnoea at rest, pulmonary rales or signs of
0·4
g . kg
1 . min
1) group (5·0%). Myocardial rup- peripheral hypoperfusion (Table 1). Earlier studies in
ture occurred more frequently with placebo than with post-acute myocardial infarction patients with similar
levosimendan (3·9% vs 0·25%, respectively, P=0·027). clinical characteristics have identified them to be at very
Of the 41 patients who experienced adverse events high risk of death[1–3]. In common with these earlier

Table 5 Mean changes in blood pressure and heart rate after 30 min and 6 h

Levosimendan Levosimendan Levosimendan Levosimendan

Placebo 6
g . kg
1 + 12
g . kg
1 + 24
g . kg
1 + 24
g . kg
1 +
(n=102) 0·1
g . kg
1 . min
1 0·2
g . kg
1 . min
1 0·2
g . kg
1 . min
1 0·4
g . kg
1 . min
1
(n=103) (n=100) (n=99) (n=100)

Systolic blood pressure (mmHg) 123·5 (21·1) 123·3 (19·5) 128·0 (19·7) 125·0 (22·2) 125·5 (18·7)
baseline mean (SD)
 30 min
1·6
3·0
1·7
2·1
3·0
 6 h*
1·3
2·1
4·2
5·4
7·9
Diastolic blood pressure (mmHg) 76·8 (13·2) 74·7 (11·4) 76·0 (12·8) 75·0 (12·4) 74·9 (12·9)
baseline mean (SD)
 30 min
2·4
3·4
3·0
4·5
3·8
 6 h†
2·5
3·1
4·3
4·6
8·0
Heart rate (beats/min) 83·8 (16·0) 81·8 (13·9) 81·5 (17·8) 84·7 (16·9) 80·0 (17·2)
baseline mean (SD)
 30 min
1·5 0·5 1·1 4·7 5·2
 6 h† 0·0 2·0 3·7 3·9 11·4

*P=0·012 for dose-relation (CMH non-zero correlation test).

†P=0·001 for dose-relation (CMH non-zero correlation test).

Table 6 Adverse events during 6 h infusion

Levosimendan Levosimendan Levosimendan Levosimendan

Adverse event Placebo 6
g . kg
1 + 12
g . kg
1 + 24
g . kg
1 + 24
g . kg
1 + P-value*
(n=102) 0·1
g . kg
1 . min
1 0·2
g . kg
1 . min
1 0·2
g . kg
1 . min
1 0·4
g . kg
1 . min
1
(n=103) (n=100) (n=99) (n=100)

Ventricular extrasystoles 1 (1·0%) 3 (2·9%) 1 (1·0%) 4 (4·0%) 9 (6·0%) 0·198

Atrial fibrillation 2 (2·0%) 1 (1·0%) 4 (4·0%) 3 (3·0%) 3 (3·0%) 0·653
Other atrial arrhythmia 1 (1·0%) 1 (1·0%) 1 (1·0%) 1 (1·0%) 2 (2·0%) 0·952
Sinus tachycardia 2 (2·0%) 0 (0·0%) 0 (0·0%) 3 (3·0%) 5 (5·0%) 0·028
Hypertension 1 (1·0%) 1 (1·0%) 3 (3·0%) 0 (0·0%) 0 (0·0%) 0·206
Nausea 0 (0·0%) 1 (1·0%) 1 (1·0%) 2 (2·0%) 1 (1·0%) 0·611
Headache 1 (1·0%) 2 (1·9%) 3 (3·0%) 1 (1·0%) 1 (1·0%) 0·796
Myocardial rupture 4 (3·9%) 1 (1·0%) 0 (0·0%) 0 (0·0%) 0 (0·0%) 0·027

*Fisher’s exact test.

Eur Heart J, Vol. 23, issue 18, September 2002

1430 V. S. Moiseyev et al.

findings[1–3], the 14-day follow-up showed a 20% was associated with a significant reduction in the com-
mortality in the placebo group; this rose to 31% at the bined risk of death and worsening heart failure and also
180-day follow-up. Compared with the recent large-scale in the need for new medications for heart failure during
observational studies in patients with acute myocardial the infusion period. The treatment benefit on combined
infarction, there were no major differences between these risk of death and worsening heart failure was still
and RUSSLAN-study regarding the use of concomitant evident 24 h after the start of the treatment.
medication, except for the lower usage of thrombolytics The continued mortality benefit up to 180 days after a
compared with Western Europe. This, however, was 6-h infusion is noteworthy. It is evident, however, that
similar to that reported in the U.S.A.[25–27]. It is note- the risk reduction attributable to levosimendan was
worthy that the study was performed without invasive achieved during the first 14 days of follow-up. After 14
haemodynamic monitoring, thus reflecting common days the Kaplan–Meier curves are parallel indicating no
clinical practice[28], which also helped to ensure that further additional survival benefit after that time (Fig.
investigators’ knowledge of the changes in haemody- 3). Similar long-term results have also been seen in trials
namic parameters did not bias the assessments of with short-term therapy with thrombolytic agents and
symptoms of heart failure. beta-blockers[31–33]. However, this is the first time, that
In post-acute myocardial infarction patients it is the decrease in mortality in this patient population was
especially important not to increase the ischaemic achieved by the use of an intravenous positive inotropic
burden. An improved cardiac contractility must not be drug. This interesting finding is in accordance with
obtained at the expense of an increase in oxygen demand previous pharmacological results. In a dog study levosi-
and further ischaemic events. The study was therefore mendan was found to reduce myocardial infarct size,
designed primarily as a randomized double-blind dose- suggesting cardioprotective effects[19]. In another,
safety trial with a placebo group. The end-points chosen recently published study racemic simendan improved
for this study — hypotension and ischaemia, are survival in rats with healed myocardial infarction[34]. It
relevant to both the study population and to the has also been shown that the haemodynamic benefits of
mechanisms of action of levosimendan — improved a 6-h levosimendan infusion in patients with heart
cardiac contractility and vasodilation[11–13,17–20,29]. The failure were not at the expense of increased sympatho-
proportion of patients experiencing hypotension mimetic stimulation or autonomic imbalance, which are
and/or ischaemia during the 6 h infusion was similar in known to be associated with an increased proarrhythmic
the combined levosimendan groups and the placebo risk[35]. Thus levosimendan possesses a unique combina-
group. A higher risk of hypotension and/or ischaemia tion of antiischaemic and inodilatory properties and
compared with placebo was observed only with therefore favourable clinical results in patients with
the highest levosimendan dose (24
g . kg
1 + ischaemic pump failure are not surprising[36].
0·4
g . kg
1 . min
1). These findings are consistent Given the similarity in patient populations, differences
with the results of a previous dose-finding study in in mortality rates cannot be attributed to the differences
patients with congestive heart failure, excluding patients in the baseline characteristics. Especially noteworthy is
with acute myocardial infarction, that identified 0·05– the finding that the prevalence of diabetes, a disease
0·2
g . kg
1 . min
1 as the optimal infusion rate for known to have an adverse effect on survival in patients
levosimendan[21]. with acute myocardial infarction[37–38], was higher
The effects of levosimendan on the improvement of among levosimendan-treated patients. However, when
symptoms of heart failure during the infusion period evaluating the mortality results, one should take into
were small. Given the short duration of the study and account that the study was not prospectively designed
the relatively insensitive methods used to appraise and powered to show a difference in mortality as
changes in symptom severity, this finding is not unex- an end-point. Nevertheless, the significant difference
pected. There have been few studies in patients with left observed suggests that levosimendan may have favour-
ventricular failure due to acute myocardial infarction; able effects on long-term mortality outcomes in addition
moreover, no published placebo-controlled double- to its beneficial effects on haemodynamics (such as
blind study has reported significant improvement of increased stroke volume and cardiac output, reduced
symptoms in this patient population. pulmonary capillary wedge pressure)[21–23]. This
As clinical measures like symptom inquiry provide possibility needs to be confirmed in a prospective
only subjective evidence of changes in clinical status, mortality trial.
there was a need to provide more objective evidence, The combined risk of death and worsening heart
which would better characterize the change in clinical failure and all-cause mortality at 24 h, 14 days and 180
status[30]. Especially in the acute setting, the clinical days showed no dose-relation, and the frequency
stabilization of the patient (i.e. prevention of heart of events was lower in all the levosimendan groups
failure worsening) is also an important clinical goal. For than in the placebo group (Table 4). However, a
this purpose, ‘the combined risk of death and worsening dose-relation regarding all-cause mortality was seen
heart failure’ was used as a pre-defined end-point in our during the 6 h infusion period. It is important that the
study. ‘Worsening heart failure’ included all possible highest levosimendan dose (24
g . kg
1 . min
1 +
adverse clinical events, indicating deterioration of clini- 0·4
g . kg
1 . min
1), showing a higher incidence of
cal condition in this patient population. Levosimendan ischaemia and/or hypotension during the 6 h infusion,

Eur Heart J, Vol. 23, issue 18, September 2002

 The RUSSLAN study 1431

100 severe heart failure after acute myocardial infarction: a

randomized double-blind trial. Eur Heart J 1993; 14:
80 696–700.
60 [6] Karlsberg RP, DeWood MA, DeMaria AN, Berk MR,
Risk

Lasher KP for The Milrinone-Dobutamine Study Group.

40
Comparative efficacy of short-term intravenous infusions of
20 milrinone and dobutamine in acute congestive heart failure
0 following acute myocardial infarction. Clin Cardiol 1996; 19:
21–30.
–20 [7] Krell MJ, Kline EM, Bates ER et al. Intermittent, ambulatory
Benefit

–40 dobutamine infusions in patients with severe congestive heart

failure. Am Heart J 1986; 112: 787–91.
–60 [8] Packer M, Carver JR, Rodeheffer RJ, Ivanhoe RJ, DiBianco
–80 R, Zeldis SM for the PROMISE Study Research Group.
–100 Effect of oral milrinone on mortality in severe chronic heart
6 + 0·1 12 + 0·2 24 + 0·2 24 + 0·4 failure. N Engl J Med 1991; 325: 1468–75.
Levosimendan infusion rate [9] Cohn JN, Goldstein SO, Greenberg BH et al. A dose-
dependent increase in mortality with vesnarinone among
Figure 4 Placebo–adjusted risk-benefit ratio of levosi- patients with severe heart failure. N Engl J Med 1998; 339:
mendan. Placebo-adjusted means that the risk for each 1810–6.
levosimendan group is expressed as a percentage of the [10] O’Connor CM, Gattis WA, Uretsky BF, Kirkwood FA,
corresponding risk in the placebo group. Black columns McNulty SE, Grossman SH for the FIRST Investigators.
represent the risk of clinically significant hypotension or Continuous intravenous dobutamine is associated with an
increased risk of death in patients with advanced heart failure:
ischaemia and white columns represent the combined risk Insights from the Flolan International Randomized Survival
of death and worsening heart failure during 6 h infusion. Trial (FIRST). Am Heart J 1999; 138: 78–86.
[11] Haikala H, Nissinen E, Etemadzadeh E, Levijoki J, Lindén
I-B. Troponin C-mediated calcium sensitization induced by
was safe and effective in this respect, i.e. ischaemia and levosimendan does not impair relaxation. J Cardiovasc Phar-
hypotension during infusion were not adversely affecting macol 1995; 25: 794–801.
the short-term efficacy and long-term safety. However, [12] Lancaster MK, Cook SJ. The effects of levosimendan on
[Ca 2+]i in guinea-pig isolated ventricular myocytes. Eur J
combining the results of the primary end-point and the
Pharmacol 1997; 339: 97–100.
combined risk of death and worsening heart failure as [13] Kristof E, Szigeti G, Papp Z et al. Cardiac responses to
the secondary end-point indicate that the risk–benefit calcium sensitizers and isoproterenol in intact guinea pig
ratio of a 6 h infusion of levosimendan was favourable hearts. Effects on cyclic AMP levels, protein phosphorylation,
up to 0·2
g . kg
1 . min
1 due to a higher incidence of myoplasmic calcium concentration and left ventricular func-
tion. Ann N Y Acad Sci 1998; 853: 316–9.
ischaemia and/or hypotension in the highest dose group
[14] Singh BN, Lilleberg J, Sandell E-P, Ylönen V, Lehtonen L,
(Fig. 4). Toivonen L. Effects of levosimendan on cardiac arrhythmias:
In conclusion, the RUSSLAN study shows that a 6 h electrophysiologic and ambulatory electrocardiographic find-
infusion of levosimendan (0·1–0·2
g . kg
1 . min
1) ings in Phase II and Phase III clinical studies in cardiac failure.
did not increase clinically significant hypotension or Am J Cardiol 1999; 83 (Suppl I): (I)16–(I)20.
[15] Ukkonen H, Saraste M, Akkila J et al. Myocardial efficiency
ischaemia. Levosimendan also decreased the incidence during calcium sensitization with levosimendan: a noninvasive
of worsening heart failure and reduced both short- study with positron emission tomography and echocardiogra-
and longer-term mortality. Levosimendan offers a prom- phy in healthy volunteers. J Clin Pharmacol Ther 1997; 61:
ising therapeutic option for the management of left 596–607.
ventricular failure complicating an acute myocardial [16] Ukkonen H, Saraste M, Akkila J et al. Myocardial efficiency
during levosimendan infusion in congestive heart failure. J
infarction. Clin Pharmacol Ther 2000; 68: 522–31.
[17] Yokoshiki H, Katsube Y, Sunagawa M, Sperelakis N. The
novel calcium sensitizer levosimendan activates the ATP-
References sensitive K + channel in rat ventricular cells. J Pharmacol Expt
Ther 1997; 283: 375–83.
[1] Dwyer EM, Greenberg HM, Steinberg G and The Multicenter [18] Yokoshiki H, Katsube Y, Sunagawa M, Sperelakis N.
Postinfarction Research Group. Clinical characteristics and Levosimendan, a novel Ca2+ -sensitizer activates the
natural history of survivors of pulmonary congestion glibenclamide-sensitive K + channel in rat arterial myocytes.
during acute myocardial infarction. Am J Cardiol 1989; 63: Eur J Pharmacol 1997; 333: 249–59.
1423–8. [19] Pataricza J, Hohn J, Petri A, Balogh A, Papp JG. Comparison
[2] The Acute Infarction Ramipril Efficacy (AIRE) Study inves- of the vasorelaxing effect of cromakallim and the new inodi-
tigators. Effect of ramipril on mortality and morbidity in lator, levosimendan, in human isolated portal vein. J Pharm
survivors of acute myocardial infarction with clinical evidence Pharmacol 2000; 52: 213–17.
of heart failure. Lancet 1993; 342: 821–8. [20] Kersten JR, Montgomery MW, Pagel PS, Warltier DC.
[3] Mahon NG, O’Rorke CO, Codd MB, McCann HA, McGarry Levosimendan, a new positive inotropic drug, decreases myo-
K, Sugrue DD. Hospital mortality of acute myocardial infarc- cardial infarct size via activation of K (ATP) channels. Anesth
tion in the thrombolytic era. Heart 1999; 81: 478–82. Analg 2000; 90: 5–11.
[4] Gillespie TA, Ambos HD, Sobel BE, Roberts R. Effects of [21] Nieminen MS, Akkila J, Hasenfuss G, Kleber FX, Lehtonen
dobutamine in patients with acute myocardial infarction. Am LA, Mitrovic V on behalf of the Study Group. Haemody-
J Cardiol 1977; 39: 588–94. namic and neurohumoral effects of continuous infusion of
[5] Caldicott LD, Hawley K, Heppell R, Woodmansey PA, levosimendan in patients with congestive heart failure. J Am
Channer KS. Intravenous enoximone or dobutamine for Coll Cardiol 2000; 36: 1903–12.

Eur Heart J, Vol. 23, issue 18, September 2002

1432 V. S. Moiseyev et al.

[22] Slawsky MT, Colucci WS, Gottlieb SS, Greenberg BH, Appendix
Haeusslein E, Hare J on behalf of the Study Investigators.
Acute hemodynamic and clinical effects of levosimendan in
patients with severe heart failure. Circulation 2000; 102: Steering Committee: V. S. Moiseyev (Chairman), N. Andrejevs,
2222–7. A. P. Golikov, L. B. Lazebnik, M. Y. Ruda, Z. D. Kobalava
[23] Follath F, Hinkka S, Jäger D et al. Dose-ranging and safety (non-voting secretary)
with intravenous levosimendan in low-output heart failure: Safety Committee: K. I. Lie (University of Amsterdam, The
experience in 3 pilot studies and outline of the LIDO trial. Am Netherlands, Chairman), M. S. Nieminen (University of Helsinki,
J Cardiol 1999; 83 (Suppl I): (I)21–(I)25. Finland), R. G. Oganov (Russian Centre for Preventive Medicine,
[24] Luotolahti M, Lammintausta O, Ukkonen H et al. Levosi- Moscow, Russia), S. J. Pocock (London School of Hygiene
mendan, a calcium sensitizer and potassium channel opener, is and Tropical Medicine, UK), J. H. Svendsen (Rigshospitalet,
safe and improves left ventricular function in acute myocar- Copenhagen, Denmark)
dial infarction. Circulation 1998; 98 (Suppl I): (I)105–(I)106. Study Manager: P. Põder (Orion Pharma, Espoo, Finland).
Statisticians: T. Laine (Orion Pharma, Espoo, Finland), K.
[25] European Secondary Prevention Group. Translation of clini- Vaahtera (CRST, Turku, Finland)
cal trials into practice: a European population-based study of Principal Investigators: Valentin S. Moiseyev (Hospital No. 64,
the use of thrombolysis for acute myocardial infarction. Russian People’s Friendship University, Moscow, Russia),
Lancet 1996; 347: 1203–7. Nikolajs Andrejevs (P. Stradin’s Research And Teaching Hospital,
[26] Chen J, Radford MJ, Wang Y, Marcinak TA, Krumholz HM. Latvian Medical Academy, Riga, Latvia)
Do ‘America’s best hospitals’ perform better for acute myo- Investigators and study sites: Z. Kobalava, S. Terestsenko, I.
cardial infarction? N Engl J Med 1999; 340: 286–92. Karavaeva, E. Pavlicova, N. Shatkovskiy (Hospital No. 64,
[27] Michaels AD, Maynard C, Every NR, Barron HV for the Russian People’s Friendship University, Moscow, Russia); J.
National Registry of Myocardial Infarction 2 participants. Lacis, T. Andreeva (P. Stradin’s Research And Teaching Hospital,
Early use of ACE inhibitors in the treatment of acute myo- Latvian Medical Academy, Riga, Latvia); V. Almazov, B.
cardial infarction in the United States: Experience from Bondarenko, N. Perepech, A. Nedoshivin (Sankt-Petersburg Re-
National Registry of Myocardial Infarction 2. Am J Cardiol search Institute of Cardiology, Sankt-Petersburg, Russia); I.
1999; 84: 1176–81. Bokarew, M. Akseneva, S. Dovgolis, M. Ulibysheva (Hospital No.
[28] Zion MM, Balkin J, Rosenmann D et al. Use of pulmonary 20, Moscow Medical Academy, Moscow, Russia); A. Golikov, T.
artery catheters in patients with acute myocardial infarction. Zvereva, A. Butylin (Sklifosovsky Research Institute for Emer-
Analysis of experience in 5841 patients in the SPRINT gency Medicine, Moscow, Russia); N. Gratsiansky, O. Averkov, I.
Registry. Chest 1990; 98: 1331–5. Yavelov (Hospital No. 29, Centre for Atherosclerosis, Moscow,
Russia); V. Zadionchenko, G. Shekhyan, V. Sinyukaeva (Hospital
[29] Gill JB, Cairns JA, Roberts RS et al. Prognostic importance
No. 11, Moscow Stomatological Institute, Moscow, Russia); R.
of myocardial ischemia early after acute myocardial infarc-
Karpov, V. Markov, A. Repin (Tomsk Cardiology Research
tion. N Engl J Med 1996; 334: 65–70.
Centre, Tomsk, Russia); L. Lazebnik, I. Konev, O. Kuznetsov
[30] Packer M. Proposal for a new clinical end point to evaluate (Hospital No. 60, Russian Medical Academy for Postgraduate
the efficacy of drugs and devices in the treatment of chronic Training, Moscow, Russia); V. Lusov, N. Volov, I. Gordeev
heart failure. J Card Fail 2001; 7: 176–82. (Hospital No. 15, Russian Medical University, Moscow, Russia);
[31] ISIS-1 (First International Study of Infarct Survival) Collabo- V. Makolkin, V. Sulimov, N. Novikova, D. Andreev (Department
rative Group. Randomised trial of intravenous atenolol of Therapy No. 1, Moscow Medical Academy, Moscow, Russia);
among 16027 cases of suspected acute myocardial infarction: A. Michailov, V. Gelnov, D. Platuschichin (Hospital No. 7,
ISIS 1. Lancet 1986; ii: 57–65. Moscow Medical Academy, Moscow, Russia); L. Olbinskaya, V.
[32] Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Zacharova, G. Travin, D. Tschepkii, P. Sharaev (Hospital No. 36,
Miocardico (GISSI). Long-term effects of intravenous throm- Moscow Medical Academy, Moscow, Russia); L. Orlov, A. Ipatov
bolysis in acute myocardial infarction: Final Report of the (Hospital No 59, Moscow Stomatological Institute, Moscow,
GISSI study. Lancet 1987; ii: 871–4. Russia); M. Ruda, I. Staroverov, V. Gramovitch (Russian Cardi-
[33] ISIS-2. Randomised trial of intravenous streptokinase, oral ology Research Centre, Moscow, Russia); D. Zateichikov, S.
aspirin, both or neither among 17187 cases of suspected acute Gneushev, D. Privalov (Hospital No. 51, Moscow, Russia); B.
myocardial infarction. Lancet 1988; ii: 349–60. Sidorenko, A. Gruzdev, S. Vasetchkin, A. Zyryanov (Central
Clinical Hospital of Russian Government Medical Centre,
[34] Levijoki J, Pollesello P, Kaheinen P, Haikala H. Improved Moscow, Russia); G. Storozhakov, A. Selivanov, D. Tipisev (ZIL
survival with simendan after experimental myocardial infarc- Hospital, Russian Medical University, Moscow, Russia); Y.
tion in rats. Eur J Pharmacol 2001; 419: 243–8. Tokmachev, A. Arefjev, R. Striouk, A. Akimov, N. Guschanskaya
[35] Binkley PF, Nunziata E, Hatton PS, Leier CV. The positive (Hospital No. 67, Moscow Stomatological Institute, Moscow,
inotropic agent levosimendan mediates increased cardiac out- Russia); A. Vyortkin, O. Talibov, A. Laptev, L. Solodar (Hospital
put without progression of sympathovagal imbalance in No. 50, Moscow Stomatological Institute, Moscow, Russia); V.
patients with heart failure. Circulation 2000; 102 (Suppl II): Dvornikov, A. Martynov, S. Kakorin (Hospital No. 53, Russian
(II)720–(II)721. University of People’s Friendship, Moscow, Russia); G.
[36] Lehtonen LA. Levosimendan: a parenteral calcium-sensitising Arutyunov, A. Rozanov, A. Vershinine (Hospital No. 4, Russian
drug with additional vasodilatory properties. Exp Opin Invest University of People’s Friendship, Moscow, Russia), G.
Drugs 2001; 10: 945–60. Zhukovsky (Russian Centre for Preventive Medicine, Moscow,
[37] Karlson BW, Herlitz J, Hjalmarson A r . Prognosis of acute Russia)
myocardial infarction in diabetic and non-diabetic patients. Sponsor: The RUSSLAN-study was sponsored by Orion Pharma,
Diabet Med 1993; 10: 449–54. Espoo, Finland.
[38] McGuire DK, Emanuelsson H, Granger CB et al. for
GUSTO-IIb Investigators. Influence of diabetes mellitus on
clinical outcomes across the spectrum of acute coronary
syndromes. Findings from the GUSTO-IIb Study. Eur Heart
J 2000; 21: 1750–8.

Eur Heart J, Vol. 23, issue 18, September 2002