Original Articles

Cardio-Renal Effects of the A1 Adenosine Receptor

Antagonist SLV320 in Patients With Heart Failure
Veselin Mitrovic, MD; Petar Seferovic, MD; Slobodan Dodic, MD; Mirjana Krotin, MD;
Aleksander Neskovic, MD; Kenneth Dickstein, MD; Hanka de Voogd, MD; Christiane Böcker, PhD;
Dieter Ziegler, MD; Michael Godes, MD; Roumen Nakov, MD; Hans Essers, MSc;
Cees Verboom, MD; Berthold Hocher, MD

Background—Blocking the tubuloglomerular feedback mechanism with adenosine A1 receptor antagonists seems to
improve diuresis and sodium excretion without compromising the glomerular filtration rate in patients with heart failure.
However, the direct cardiac effects of this compound class have not been investigated to date.
Methods and Results—In total, 111 patients (109 men and 2 women) received a 1-hour infusion of 5, 10, and 15 mg SLV320,
an adenosine A1 receptor antagonist, placebo, or 40 mg furosemide. Mean age was 57.9 years, mean ejection fraction was
28.1%, 82 patients were of New York Heart Association class II, and 29 patients were of New York Heart Association class
III. Hemodynamic parameters (heart rate, blood pressure, pulmonary capillary wedge pressure, mean pulmonary arterial
pressure, systemic vascular resistance, right atrial pressure, and cardiac output) were determined. Kidney function was
assessed by cystatin C measurements and by analysis of urine output and urine electrolytes. In addition, pharmacokinetics of
SLV320 and ex vivo inhibition of adenosine A1 receptor activity were performed. SLV320 was well tolerated, and no serious
adverse events were observed. Heart rate, blood pressure, pulmonary capillary wedge pressure, mean pulmonary arterial
pressure, right atrial pressure, and cardiac output were not altered by any dose of SLV320. Pulmonary capillary wedge
pressure was significantly (P⫽0.04) decreased by furosemide (⫺6.2⫾5.9 mm Hg). Systemic vascular resistance was
significantly (P⫽0.04) increased in the furosemide group (⫹166.70⫾261.87 dynes 䡠 s⫺1 䡠 cm⫺5), whereas all SLV320 groups
showed no significant alterations of systemic vascular resistance. Changes from baseline cystatin C plasma concentrations
decreased after 10 mg SLV320 (⫺0.093⫾0.137 mg/L, P⫽0.046), whereas furosemide resulted in a significant (P⫽0.03)
increase of cystatin C (⫹0.052⫾0.065 mg/L) versus baseline. All values represent mean changes⫾SD from baseline at 3
hours postdosing: SLV320 (10 and 15 mg) increased significantly sodium excretion and diuresis compared with placebo
during the 0- to 6-hour collection period postdosing.
Conclusions—SLV320 infusion shows no immediate effects on cardiac hemodynamics. SLV320 might improve
glomerular filtration rate while simultaneously promoting natriuresis and diuresis.
Clinical Trial Registration— clinicaltrials.gov Indentifier: NCT00160134.
(Circ Heart Fail. 2009;2:523-531.)
Key Words: adenosine A1 antagonist 䡲 renal function 䡲 congestive heart failure 䡲 diuretics

I n the past decade, it became more and more evident that

patients suffering from both chronic renal failure and
chronic heart failure (HF) are characterized by a poor
Editorial see p 519
Clinical Perspective on p 531
outcome with respect to morbidity and mortality. The under- causal factors leading to the cardiorenal syndrome.1,2 Thus,
lying mechanisms are not yet completely understood. Inflam- new approaches are urgently needed. There is already evi-
mation, vascular and tissue calcification, anemia, and direct dence that risk factors related to kidney function correlate
cardiotoxic effect of yet unknown molecules that accumulate much better with outcome compared with classical cardiac
in patients with impaired kidney function are suspected to be risk factors in patients with HF.3 Moreover, a recent study

Received June 12, 2008; accepted September 3, 2009.

From the Kerckhoff-Klinik, Department of Cardiology and Cardiosurgery (V.M.), Bad Nauheim, Germany; Klinicki Center Serbia (P.S.), Institut za
kardiovaskularne bolest, Belgrad; Medicinski Facultet Uni Novi Sad (S.D.), Institut za kardiovaskularne bolest, Sremska Kamenica; Clinical Centre Bezanijska
Kosa, Department of Cardiology (M.K.), Zemum; Dedinje Cardiovascular Institute Milana Tepica 1 (A.N.), Belgrade, Serbia; Stavanger Universitetssykehus
Kardiologisk divisjon (K.D.), Stavanger, Norway; Solvay Pharmaceuticals Research Laboratories (H.V., C.B., D.Z., R.N., H.E., C.V., B.H.), Hannover, Germany and
Weesp, The Netherlands; and Center for Cardiovascular Research/Department of Pharmacology and Toxicology (M.G., B.H.), Charité, Campus Mitte, Berlin, Germany.
Hanka de Voogd, Christiane Böcker, Dieter Ziegler, Roumen Nakov, Hans Essers Cees Verboom, and Berthold Hocher are research employees of
Solvay Pharmaceuticals.
Correspondence to Berthold Hocher, MD, Solvay Pharmaceuticals, Hans Böckler Allee 20, D-30173 Hannover, Germany. E-mail
[email protected]
© 2009 American Heart Association, Inc.
Circ Heart Fail is available at http://circheartfailure.ahajournals.org DOI: 10.1161/CIRCHEARTFAILURE.108.798389

523 at BIBL DE L'UNIV LAVAL on July 3, 2015

Downloaded from http://circheartfailure.ahajournals.org/
524 Circ Heart Fail November 2009

demonstrated that improvement of renal function by an A1 screening); (3) subjects with 2nd or 3rd degree atrioventricular block
adenosine receptor antagonist in patients with acutely decom- or sick sinus syndrome; (4) subjects with a heart rate of ⬍50 or ⬎110
bpm as measured on ECG (at screening); and (5) subjects with a
pensated HF may translate in reduced hospitalization and
transplanted heart. The study protocol was approved by all partici-
60-day mortality.4 pating institutional review boards, and all patients gave written
The adenosine system is involved in several key functions informed consent for participation in this study.
of both the kidney and the heart. Adenosine acts through 4 The clinical study was divided into 4 periods:
different receptors: A1, A2A, A2B, and A3.5 In the kidney, ● Screening (visit 1 [day 7 to day 2])
adenosine plays a key role in the tubuloglomerular feedback ● Pretreatment period (visit 2a [day 1]) (catheterization session)
mechanism and thus exerts an inhibitory action on glomerular ● Treatment period (visit 2b [day 1]) (60-minute infusion, 12-hour
hemodynamics and glomerular filtration rate (GFR) through postdosing hemodynamic assessments, pharmacokinetics mea-
A1 receptors.6,7 In addition, adenosine has antinatriuretic (and surements, and observation for 24 hours)
● Posttreatment period (visit 3 [day 3 to day 8]) (follow-up)
antidiuretic) effects, through activation of tubular A1 recep-
tors, that promote sodium reabsorption.8 –10 In the heart, A1 The plan was to enroll 110 subjects in the study with 22 subjects
receptor activation may be deleterious with regard to ische- randomly assigned to each treatment group. Each subject received 1
mia/reperfusion injury, through promotion of neutrophil che- infusion of SLV320, placebo, or furosemide. Subjects were assessed
moattraction and adhesion,11–13 whereas A2 and/or A3 acti- at screening as eligible and willing to participate in the clinical study.
Subjects who met all of the inclusion and none of the exclusion
vation is protective in this setting. Given these complex criteria were catheterized at the pretreatment period (day 1) and
preclinical data, human studies addressing the cardiac effects received 1 of the following treatment regimens:
of A1 adenosine antagonists are urgently needed.
● SLV320 5 mg IV
In patients with congestive HF, A1 receptor antagonists
● SLV320 10 mg IV
might increase diuresis without compromising GFR, and is in ● SLV320 15 mg IV
contrast to the loop diuretic furosemide, which increases ● Placebo IV (saline)
diuresis at the expense of a decreased GFR.14,15 Moreover, ● Furosemide 40 mg IV
loop diuretics might be harmful for patients with acutely
The randomization was performed in blocks of 5 and stratified by
decompensated HF.16 center.
SLV320, a pyrrolopyrimidine derivative, is a selective The doses of SLV320 were chosen based on the phase 1 data in
adenosine A1 receptor antagonist.17 Use of this compound in healthy volunteers; the selected doses showed no safety signals and
rats with 5/6 nephrectomy showed that SLV320 prevented the had clear diuretic properties. Study medication was infused over a
development of uremia-related cardiac fibrosis.17 In line with 1-hour period. Baseline medications, such as diuretics, angiotensin-
converting enzyme inhibitors, ␤-blockers, and nitrates and other
this finding are data indicating that cardiac overexpression of vasodilators, if applicable, were withheld until after the 12-hour
the A1 adenosine receptor in mice causes myocardial fibrosis postdosing hemodynamic assessments were completed, unless re-
associated with an increased mortality.18 quired in an emergency. Baseline medications varied substantially in
In this clinical trial, we focused on short-term effects. This the patients with HF and were withheld during hemodynamic
was a randomized, placebo-controlled, double-blind, multi- assessment to get a more uniform situation in this very first phase 2
trial with SLV320.
center, parallel-group, single-dose study analyzing the car- The investigation started in the early morning meaning that all
diac and renal efficacy as well as safety of SLV320 in patients patients took their last usual oral drug administered the evening
with stable HF in comparison to furosemide and placebo. before. At pretreatment, after a resting period of 30 minutes, cardiac
output (CO) and heart rate measurements were determined at
Methods 10-minute intervals until baseline stability had been established.
This was a randomized, placebo-controlled, double-blind, multi- Baseline stability was defined as CO and heart rate measurements
center, parallel-group, single-dose study to evaluate hemodynamic showing ⬍10% variability at 2 consecutive time points.
and renal effects of single IV doses of the A1 adenosine receptor A balloon-tipped, thermodilution pulmonary artery catheter was
antagonist, SLV320 (5, 10, and 15 mg as 1-hour infusion) compared inserted using standard percutaneous techniques. The antecubital, sub-
with placebo (1-hour saline infusion) or furosemide (40 mg as clavian, femoral, or internal jugular venous approach was allowed. Each
5-minute bolus and 55-minute saline) during 12-hour right heart participating study center was to follow their standard procedures.
catheterization in subjects with stable HF requiring diuretic treat- The hemodynamic variables were measured at: 0, 0.5, 1, 1.5, 2, 3,
ment. The primary end point was to evaluate the maximum reduction 4, 5, 6, 8, and 12 hours. Postdosing is defined relative to the start
in the pulmonary capillary wedge pressure (PCWP) from baseline time of the infusion.
during the first 12 hours (regardless when the maximum reduction The following hemodynamic measurements were made: PCWP
may occur during this 12 hours) after dosing with any IV dose of in mm Hg; CO in L/min (determined by thermodilation); heart rate in
SLV320 in subjects with HF (New York Heart Association [NYHA] bpm (from ECG); pulmonary arterial systolic pressure and pulmonary
classes II–III) requiring diuretics compared with dosing with placebo. arterial diastolic pressure in mm Hg; systemic arterial systolic pressure
The study enrolled patients at 6 clinical sites. Inclusion criteria and systemic arterial diastolic pressure in mm Hg; mean arterial pressure
included NYHA classes II–III HF with an ejection fraction of ⬍35% in mm Hg; and right atrial pressure (RAP) in mm Hg.
measured by echocardiography at screening and the presence of At each measurement time point for hemodynamic variables,
edema, despite a daily furosemide dose of at least 80 mg. Baseline measurements were recorded 3 times (5 times for subjects with atrial
GFR, as measured by estimated creatinine clearance at screening fibrillation) and the mean captured on the case report form.
(MDRD [the Modification of Diet in Renal Disease study] formula),
was at least 30 mL/min per 1.73 m2 or serum creatinine was ⬍1.9 Efficacy Data for Urinary Output and Excretions
mg/dL. Main exclusion criteria were (1) the subjects’ conditions Serum creatinine and cystatin C concentrations were measured
were so unstable that they required hospitalization (for cardiovascu- immediately before dosing (0 hour) and at 0.5, 1, 1.5, 2, 3, 4, and 8
lar disease) or adjustment of background medications for HF; (2) hours. Urine was collected during the 0 to 6- and 6 to 12-hour
subjects with a sitting systolic blood pressure of ⬍90 mm Hg (at intervals. Patients were asked to empty their bladder before collec-

Downloaded from http://circheartfailure.ahajournals.org/ at BIBL DE L'UNIV LAVAL on July 3, 2015

 Mitrovic et al A1 Adenosine Receptor Antagonist and Heart Failure 525

tion start and at the end of each collection period. The volume of measurement at baseline and at postbaseline. Continuous demo-
urine and excretion of sodium, potassium, chloride, and uric acid graphic and baseline variables presented are summarized using
were evaluated. mean, median, standard deviation, minimum, maximum, and number
of available observations. Categorical demographic and baseline
Pharmacokinetic Measurements variables are summarized by counts and percents.
Blood samples of 5 mL in heparinized tubes for determining the
plasma concentrations of SLV320 were collected during the treat- Results
ment period at 0 (predose), 0.5, 1, 1.5, 2, 3, 4, 8, 12, and 24 hours One hundred fifty-one patients were screened for the study: 40
after start of the infusion. The samples were taken from the right patients did not fulfill the inclusion and exclusion criteria and
heart catheter line (right atrium port). SLV320 plasma concentrations
were determined by validated liquid chromatography tandem mass were thus rejected from participation. The remaining 111 pa-
spectrometry methods. The concentrations of SLV320 were calcu- tients were included in the study. This was 1 patient more than
lated by suitable linear regression curve fitting. Quality control planned. The first subject’s first visit was January 5, 2005; the last
samples were analyzed throughout the study. The measured concen- subject’s last visit was November 29, 2005. Patients’ characteristics
trations of these quality control samples were used to determine the (109 men and 2 women) are given in Tables 1 and 2. Mean age
intraday and interday precision and accuracy of the method.
was 57.9 years, mean ejection fraction was 28.10, 82 patients
In Vitro Inhibition of A1 Receptor Activity were of NYHA class II, and 29 patients were of NYHA class III.
The ex vivo functional A1 antagonism in plasma samples was All patients were taking angiotensin-converting enzyme inhibi-
quantified as follows: yeast cells (Saccharomyces cerevisiae) ex- tors: 86 took ␤-blockers, 47 took cardiac glycosides, 5 were on
pressing the human adenosine A1 receptor were inoculated in thiazides, and 2 were receiving spironolactone. There were no
LT-medium (SD medium lacking leucine and tryptophan, pH 6.8) to significant differences between the placebo group, the 3 SLV320
a density of 3⫻105 cells/mL and incubated at 30°C with agitation for
16 hours (overnight). The plasma samples were stored at ⫺20°C and treatment groups, and the furosemide group (Tables 1 and 2),
were thawed on the day of the experiment. Assays were conducted in although mean baseline cystatine C was lowest in the placebo
a final volume of 100 ␮L in 96-well microtiter plates. Incubations group.
were performed as follows: 10 ␮L plasma was added followed by 10
␮L of the A1 receptor agonist 5⬘-N-ethylcarboxamidoadenosine Pharmacokinetics
(final concentration, 1 ␮mol/L in water) and 80 ␮L of the cell
Plasma concentrations of SLV320 after IV infusions are
suspension (final cell concentration, 1.6⫻105 cells/mL). Additional
incubations without plasma (replaced by water) acted as 0% inhibi- presented in Figure 1 in semilogarithmic scale. The plasma
tion (100% stimulation) controls. Furthermore, incubations without half-life was 1.42, 1.50, and 2.13 hours for the 5-, 10-, and
5⬘-N-ethylcarboxamidoadenosine and with water replacing plasma 15-mg SLV320 infusions, respectively.
were included to determine basal fluorescence by ␤-galactosidase
activity. Control plasma (without compound) had no inhibitory effect Adenosine Antagonism
on the adenosine A1 receptor. All samples were tested in duplicate,
and the means are reported. All steps were performed under The ability of a patient’s plasma to inhibit A1 adenosine
sterile conditions. The plates were agitated briefly and incubated receptor activity was similar for the SLV320 10- and 15-mg
for 4 hour at 30°C. The ␤-galactosidase activity was determined treatment groups and only slightly lower for the SLV320
using the fluorescent ␤-galactosidase substrate fluorescein-di-␤D- 5-mg treatment group.
galactopyranoside (FDG, Molecular Probes). The FDG solution In contrast, there were no changes from baseline adenosine
contains 2.5% Triton X-100 to harvest the cells. FDG/Triton X-100
was added to all wells at 20 ␮L/well (final concentration, 80 ␮mol/L). A1% inhibition time profiles (Figure 2) for placebo and
After 45 minutes of incubation, the reaction was stopped by use of 20 40-mg furosemide treatment groups, indicating no relevant
␮L/well of a 1 mol/L Na2CO3. Relative fluorescence intensity was A1 adenosine receptor antagonistic activity of placebo and
determined using a fluorometer (excitation 485 nm and emission 535 nm). furosemide treatment.
Statistical Methods Kidney Function
The sample size estimates were based on the primary end point—
All 3 SLV320 treatment groups showed a small mean
maximum change from baseline in PCWP. The sample size was
calculated using ANOVA with Dunnett’s test to correct for the decrease from baseline in cystatin C during the 12-hour
multiple comparisons. postdosing period, but compared with placebo, these differ-
A sample size of 22 subjects per treatment group was considered ences did not reach statistical significance after Dunnett‘s
to be sufficient to detect a 5-mm Hg difference in change from correction. In contrast, cystatin C showed an increase from
baseline in PCWP (a between-subject standard deviation of 5 mm Hg
baseline in the furosemide treatment group during the 12-
was assumed) between SLV320 and furosemide at an overall 5%
significance level with a power of 80%. hour postdosing period. The differences between furosemide
All efficacy parameters were analyzed using an analysis of and each of the 3 SLV320 doses were statistically significant
covariance (ANCOVA) with the baseline value as covariate and at all measured time points after Dunnett‘s correction. This
treatment, country, and NYHA classification as factors. The signif- was especially notable with the first 4 hours postdosing.
icance of the treatment effects of each of the 3 dose levels of SLV320
was examined in comparison with placebo and furosemide. Furo-
Because our study is a more an exploratory phase 2 study,
semide and placebo were also compared. P values ⬍0.05 were we also present data without Dunnett’s test for adjustment for
considered significant. We followed the intent-to-treat principles. multiple comparisons: urinary sodium and chloride excretion
Data are presented with (Dunnett’s test) and without correction for increased in a dose-dependent manner in patients receiving 5,
multiple comparison. Only if the overall F test for treatment is 10, or 15 mg SLV320 IV, whereas potassium (data not
significant, subsequent significant unadjusted pairwise comparisons
are considered. The results are presented for the entire study shown) excretion was not affected. Urine volume was also
population, consisting of all randomized subjects who receive the increased in a dose-dependent manner during the first 6 hours
single dose of study medication and have at least 1 evaluable efficacy after IV administration of SLV320. The diuretic effect of all
Downloaded from http://circheartfailure.ahajournals.org/ at BIBL DE L'UNIV LAVAL on July 3, 2015
526 Circ Heart Fail November 2009

Table 1. Patient Population

SLV320 (n⫽22)
Placebo Furosemide
5 mg 10 mg 15 mg (n⫽22) (n⫽22)
Age, y 58.4⫾11.5 59.2⫾12.5 56.9⫾9.6 57.2⫾10.4 57.9⫾10.9
Male/female 23/0 21/1 21/1 22/0 22/0
Height, m 1.79⫾0.1 1.74⫾0.1 1.75⫾0.1 1.80⫾0.1 1.79⫾0.1
Weight, kg 89.7⫾16.6 78.9⫾14.2 85.3⫾14.3 90.8⫾14.5 86.5⫾15.6
BMI, kg/m2 28.0⫾4.5 26.0⫾3.4 27.8⫾3.4 27.9⫾3.5 27.1⫾4.2
Sitting SBP, mm Hg 121.9⫾20.1 127.1⫾15.7 130.2⫾13.9 129.0⫾15.1 129.6⫾19.8
Sitting DBP, mm Hg 75.9⫾9.7 79.5⫾9.1 80.5⫾9.9 82.8⫾8.7 77.7⫾10.2
Mean arterial pressure, mm Hg 90.2⫾11.7 88.9⫾15.2 98.0⫾14.3 95.8⫾16.8 92.0⫾11.4
Heart rate, bpm 74.0⫾15.0 70.0⫾8.1 72.8⫾11.6 72.5⫾12.2 69.5⫾10.8
LVEF, % 28.0⫾4.0 27.2⫾5.5 27.7⫾5.9 28.5⫾5.0 29.2⫾4.6
Sodium, mmol/L 141.1⫾2.8 140.3⫾1.5 140.7⫾3.1 141.7⫾3.4 141.2⫾2.9
Potassium, mmol/L 4.7⫾0.5 4.6⫾0.4 4.8⫾0.4 4.6⫾0.4 4.7⫾0.5
Creatinine, ␮mol/L 97⫾13 91⫾19 91⫾16 96⫾14 91⫾17
Cystatin C, mg/L 0.96⫾0.2 1.05⫾0.3 0.95⫾0.2 0.81⫾0.1 1.03⫾0.2
Cardiac index, L/min per m2 2.17⫾0.3 2.14⫾0.4 2.32⫾0.2 2.17⫾0.3 2.24⫾0.3
Cardiac output, L/min 4.53⫾0.19 4.1⫾0.19 4.55⫾0.18 4.63⫾0.17 4.71⫾0.15
Pulmonary artery systolic pressure, mm Hg 40.9⫾11.6 42.7⫾13.1 42.1⫾13.1 41.2⫾13.2 44.5⫾11.0
Pulmonary artery diastolic pressure, mm Hg 20.3⫾7.2 19.2⫾6.3 20.3⫾7.7 18.1⫾5.6 20.4⫾6.7
Pulmonary artery pressure, mm Hg 27.8⫾7.7 29.0⫾7.2 28.9⫾8.0 27.3⫾6.5 29.5⫾6.9
Right atrial pressure, mm Hg 6.7⫾4.7 6.5⫾3.4 7.2⫾3.1 7.3⫾3.4 8.1⫾4.7
Pulmonary vascular resistance, dynes/s per cm5 197⫾113 200⫾125 202⫾133 155⫾72 183⫾88
Systemic vascular resistance, dynes/s per cm5 1126⫾358 1194⫾357 1234⫾305 1249⫾364 1055⫾201
PCWP, mm Hg 17.09⫾4.2 19.36⫾4.2 18.32⫾4.6 18.73⫾6.2 19.18⫾5.3
Baseline characteristics of the 5 groups analyzed in this study. Data are given as mean⫾SD.

SLV320 doses was lower compared with 40 mg of furo- Cardiac Hemodynamics

semide (Figure 3). The drug effects were no longer present or No statistically significant differences between any SLV320
detectable in the second urine collection period: urine excretion dose and placebo were observed at various time points and
and sodium excretion during the 7- to 12-hour collection period overall for the 12-hour postdosing assessment period for all
were not significantly different between groups (Table 3). hemodynamic outcomes after Dunnett‘s correction for mul-
SLV320 treatment (10-mg group) reduced plasma concen- tiple comparison.
trations of cystatin C significantly, whereas furosemide treat- Treatment with placebo led to a minor nonsignificant
ment had the opposite significant effect (Figure 4). The 5- and reduction of PCWP ⬇2 mm Hg. Furosemide caused a signif-
15-mg groups (with exception of the 15 mg, 0.5- and 2-hour icant decrease of PCWP (after Dunnett‘s correction) versus
time points) showed a nonsignificant reduction of plasma placebo. The effect was, however, not significant compared
cystatin C. The effect of SLV320 on cystatin C was present with the higher doses of SLV320. The maximal effect of
longer than the plasma level of SLV320. furosemide, ⬇⫺6 mm Hg, was achieved after 3 to 5 hours

Table 2. Heart Failure Classes and Reasons for Heart Failure in the Study Population
SLV320

5 mg 10 mg 15 mg Placebo Furosemide
Diabetes, n (%) 6 (26.1) 3 (13.6) 9 (40.9) 7 (31.8) 7 (31.8)
History of hypertension, n (%) 6 (26.1) 9 (40.9) 14 (63.6) 11 (50.0) 12 (54.5)
NYHA classification, n (%)
Class II 15 (65.2) 16 (72.7) 18 (81.8) 17 (77.3) 16 (72.7)
Class III 8 (34.8) 6 (27.3) 4 (18.2) 5 (22.7) 6 (27.3)
Etiology of HF, n (%)
Ischemic heart disease 9 (39.1) 13 (59.1) 15 (68.2) 9 (40.9) 8 (36.4)
Nonischemic heart disease 14 (60.9) 9 (40.9) 7 (31.8) 13 (59.1) 14 (63.6)
Baseline categorical characteristics of the 5 groups analyzed in the study. Data are given as n (%).

Downloaded from http://circheartfailure.ahajournals.org/ at BIBL DE L'UNIV LAVAL on July 3, 2015

 Mitrovic et al A1 Adenosine Receptor Antagonist and Heart Failure 527

Figure 1. Geometric mean plasma concentration-time profiles

after single 1-hour infusions of 3 doses of SLV320 (semilogarithmic Figure 2. Arithmetic mean change from baseline adenosine A1
scale). The sample sizes for pharmacokinetics are different from receptor% inhibition-time profiles in the different treatment groups.
the numbers randomized because 1 small study center was not Samples from all 111 study patients were used for this analysis.
able to perform pharmacokinetics. The 12-hour points are missing
for the 10- and 5-mg doses of SLV320 because these values were
below the limit of quantification (LOQ) of the SLV320 assay.
SLV320 group). For these events, a dose-response relation-
ship could not be established.
and persisted over several hours. Five milligrams of SLV320
had a placebo-like effect, whereas 10 and 15 mg of SLV320
Discussion
This study simultaneously analyzed cardiac and renal effects
showed a nonsignificant trend of PCWP values consistently
of an A1 adenosine receptor antagonist in patients with stable
lower than placebo (Figure 5).
HF. SLV320 increased sodium and chloride excretion as well
RAP was not reduced by any of the SLV320 doses
as diuresis in a dose-dependent manner. In contrast to
compared with placebo, whereas furosemide decreased RAP
furosemide treatment, SLV320 treatment reduced cystatin C
(Figure 5). Systemic and pulmonary vascular resistance were
plasma concentration. The hemodynamic measurements re-
not altered by any of the SLV320 doses. By contrast,
vealed no safety concerns; total peripheral resistance was not
systemic vascular resistance was significantly (P⬍0.05, with-
altered after SLV320 treatment, whereas furosemide treat-
out Dunnett‘s correction) increased in the furosemide group
ment increased total peripheral resistance in these patients
compared with placebo at time points 0.5, 1, 3, 4, and 5 hours
with HF.
postdosing. At 1.5 and 2 hours postdosing, there was a
nonsignificant trend (P⬍0.1) for a higher systemic vascular Cardiac Effects
resistance in the furosemide-treated group. PCWP was not significantly lowered by any dose of SLV320
Mean arterial blood pressure was not significantly affected at any time point. This finding is not unexpected, given the
by either SLV320 or furosemide (Figure 5). Mean pulmonary mainly renal mode of action of an A1 adenosine receptor
arterial pressure was only significantly affected by furo- antagonist (Figure 5). Overall, SLV320 had also no signifi-
semide (after Dunnett‘s correction) treatment, and the higher cant effect on pulmonary artery pressure and RAP. As
doses of SLV320 showed only a trend toward reduction of reported by others (for review see Ref. 16), furosemide
mean pulmonary arterial pressure (Figure 5). treatment led to a significant and immediate increase in total
CO remained stable after treatment with SLV320, whereas peripheral resistance. This was not observed by any dose of
furosemide showed a trend toward reduction of CO (data not SLV320. All SLV320 doses were neutral with respect to
shown). systemic vascular resistance. This is clearly an advantage for
patients with HF.
Safety Results
No deaths or other serious adverse events were reported Renal Effects
during the study. Of the total 111 subjects, 11 (9.9%) subjects This study demonstrates that SLV320 increased sodium and
had at least 1 treatment-emergent adverse event. Only 1 chloride excretion as well as diuresis in a dose-dependent
subject terminated the study prematurely because of a manner without compromising plasma cystatin C. Plasma
treatment-emergent adverse event (“sacral pain”), which was cystatin C even decreased in the 10-mg SLV320 IV group. In
severe and considered unrelated to SLV320 treatment. Five contrast, furosemide treatment increased plasma levels of
(4.5%) subjects had 6 drug-related treatment-emergent ad- cystatin C, indicative of worsening of kidney function. These
verse events. The treatment-emergent adverse events consid- data with respect to renal function fit well with recent studies
ered related to SLV320 were “dizziness,” 1subject (5-mg performed in patients with HF with other A1 receptor
SLV320 group); “nausea,” 2 subjects (5- and 10-mg SLV320 antagonists.14,15,19 –21
group); “transient hypertension” 1 subject (5-mg SLV320 A1 adenosine receptors have a dual mode of renal action7–10:
group), and “transient hypotension,” 2 subjects (5- and 15-mg (1) activation of tubular A1 adenosine receptor increases tubular
Downloaded from http://circheartfailure.ahajournals.org/ at BIBL DE L'UNIV LAVAL on July 3, 2015
528 Circ Heart Fail November 2009

Figure 3. Sodium excretion, chloride excretion, and diuresis during the first 6 hours after treatment with 3 different doses of SLV320, furo-
semide, or placebo. Samples from all 111 study patients were used for this analysis. Bars represent means⫾SEM *P⬍0.05 versus placebo.

sodium reabsorption and finally decreases sodium concentra- effects seem to last much longer than the half-life of the
tion at the macula densa; (2) activation of A1 adenosine compound. Similar observations were made with another A1
receptor at the vasa afferentia of the glomeruli plays a key adenosine receptor antagonist in patients with HF, thus this is
role in the tubuloglomerular feedback mechanism by exerting not a compound-related effect but rather a class effect.21
an inhibitory action on glomerular hemodynamics and GFR. The effects on sodium excretion and diuresis are most
Thus, both A1 adenosine receptor-mediated effects will have likely mediated by means of direct tubular effects of A1
opposite effects on GFR. This might explain the bell-shaped adenosine receptor antagonists.22
dose response curve of all investigated A1 adenosine receptor This study is the first to analyze the short-term GFR effects
antagonists in patients with HF with respect to control of by measuring cystatin C in a clinical HF trial; however, others
diuresis and GFR.14,15,19 –21 SLV320 had the clearest effect on are also currently addressing this issue (see http://clinicaltrials.
reduction of GFR measured by changes of cystatin C from gov/ct2/show/NCT00561483?term⫽cystatin⫹c&rank⫽1).
baseline with 10 mg SLV320, whereas the lower and higher Cystatin C is most likely a better marker for GFR compared
doses were less effective. with creatinine especially in patients with HF because cysta-
The long-lasting effect (at least 8 hours) on cystatin C tin C is independent of muscle mass and less dependent on
cannot be explained by the pharmacokinetics of SLV320. age and sex.23–25 Cystatin C was recently used as a marker for
Control of kidney function is complex, and the biological glomerular filtration in studies analyzing pathways of acute

Table 3. Urine Excretion and Sodium Excretion During the 7- to 12-h Postdosing Period
SLV320

5 mg 10 mg 15 mg Placebo Furosemide
Urine excretion (in mL/h) during the 105.3⫾74.9 79.9⫾43.3 110.1⫾87.1 85.61⫾72.7 90.1⫾72.6
7- to 12-h collection period
Sodium excretion (in mmol) during the 57.3⫾25.1 58.2⫾31.0 78.1⫾58.7 67.8⫾48.6 52.8⫾37.1
7- to12-h collection period
There were no statistical significant differences between groups. Data are given as mean⫾SD.

Downloaded from http://circheartfailure.ahajournals.org/ at BIBL DE L'UNIV LAVAL on July 3, 2015

 Mitrovic et al A1 Adenosine Receptor Antagonist and Heart Failure 529

ing changes in kidney function after acute IV administration

of a compound that might affect GFR. Although the experi-
ence in acute renal failure and the aforementioned advantages
of cystatin C measurements are compelling, systematic head-
to-head comparisons of the different methods to measure
short-term GFR alterations in patients with HF are needed to
confirm the applicability of the new method.

Safety
There were no major safety concerns in patients with HF. A
recent study performed with the A1 adenosine receptor
antagonist KW-3902 in patients with HF20,21 reported sei-
zures as side effect. Because A1 adenosine receptors play a
critical role in the central nervous system,33–35 seizures might
be a class effect rather than a compound-related side effect.
Although there were no exclusions for patients with previous
seizures and organic brain disease such as brain surgery or
Figure 4. Time course of plasma cystatin C concentrations after brain tumor in the SLV320 study, it is possible that the lack
administration of 5, 10, and 15 mg SLV320 IV, placebo, or 40 of seizures is due to the limited number of patients (n⫽67)
mg furosemide IV. Samples from all 111 study patients were exposed to the active drug. Further studies are clearly needed
used for this analysis. Error bars are ⫾1 SEM. *P⬍0.05, for furo-
semide versus placebo at the same time point; #P⬍0.05, for 10 to assess whether SLV320 has a better safety profile com-
mg of SLV320 versus placebo at the same time point. pared with other A1 adenosine antagonists. With respect to
safety, it is also important to note that none of the hemody-
renal failure, indicating that this biomarker of renal function namic parameters showed any safety issue (Figure 5). By
is suitable for the detection of rapid changes of kidney contrast, most of these parameters were influenced in a
function.26 –32 This SLV320 study suggests that cystatin C way—although not reaching statistical significance—that fa-
measurements offer a new and practical method for monitor- vors improvement in patients with HF. This is important to

Figure 5. Time course of pulmonary capillary wedge pressure, RAP, pulmonary vascular resistance, systemic vascular resistance, mean
arterial pressure, and mean pulmonary arterial pressure after administration of 5, 10, and 15 mg SLV320 IV, placebo, or 40 mg furo-
semide IV. All 111 patients could be used for hemodynamic assessments. Error bars are ⫾1 SEM. *P⬍0.05, for furosemide versus pla-
cebo at the same time point.

Downloaded from http://circheartfailure.ahajournals.org/ at BIBL DE L'UNIV LAVAL on July 3, 2015

530 Circ Heart Fail November 2009

note because there were suspicions based on preclinical data a favorable safety profile. Thus, SLV320 as an A1 adenosine
that a blockade of the A1 receptor might be harmful at least receptor antagonist might represent a new therapeutic strat-
in patients with coexisting coronary heart disease.36 A signif- egy for the treatment of patients with HF.
icant proportion of the patients included in the study suffered
from ischemic heart disease. These patients had the same Sources of Funding
efficacy and safety profile as those with nonischemic heart This study was supported in part by the Deutsche Forschungsge-
disease (data not shown). meinschaft (DFG; to B.H.).

Study Limitations Disclosures

None.
We have to acknowledge the following study limitations: (1)
assessment of GFR was not performed by a clearance-based
References
technique such as inulin clearance or creatinine clearance; (2) 1. Boerrigter G, Burnett JC Jr. Cardiorenal syndrome in decompensated
only a limited dose range was investigated; (3) the combina- heart failure: prognostic and therapeutic implications. Curr Heart Fail
tion of furosemide and an A1 receptor antagonist was not Rep. 2004;1:113–120.
2. Haffner D, Hocher B, Müller D, Simon K, König K, Richter CM, Eggert
studied; and (4) the study was performed in patients with B, Schwarz J, Godes M, Nissel R, Querfeld U. Systemic cardiovascular
stable HF with a clinical need for being treated with furo- disease in uremic rats induced by 1,25(OH)2D3. J Hypertens. 2006;23:
semide. However, the target population for an IV formulation 1067–1075.
is more likely acutely decompensated. Because of the early 3. Hillege HL, Girbes AR, de Kam PJ, Boomsma F, de Zeeuw D, Char-
lesworth A, Hampton JR, van Veldhuisen DJ. Renal function, neuro-
phase of development, patients with stable HF were selected hormonal activation, and survival in patients with chronic heart failure.
for initial evaluation. Circulation. 2000;102:203–210.
4. Cotter G, Dittrich HC, Weatherley BD, Bloomfield DM, O’Connor CM,
Clinical Implications Metra M, Massie BM; Protect Steering Committee, Investigators, and
Coordinators. The PROTECT pilot study: a randomized, placebo-
Renal insufficiency represents an independent risk factor for controlled, dose-finding study of the adenosine A1 receptor antagonist
disease progression and mortality in patients with HF. More- rolofylline in patients with acute heart failure and renal impairment.
over, it was already shown that parameters describing kidney J Card Fail. 2008;4:631– 640.
5. Linden J. Molecular approach to adenosine receptors: receptor-mediated
function have a better prognostic prediction for outcome than mechanisms of tissue protection. Annu Rev Pharmacol Toxicol. 2001;41:
purely cardiac biomarkers in patients with HF.3 775–787.
The SLV320 data as well as other studies14,15,19 –22 suggest 6. Ren Y, Garvin JL, Carretero OA. Efferent arteriole tubuloglomerular
that a blockade of the A1 receptor might be of special interest feedback in the renal nephron. Kidney Int. 2001;59:222–229.
7. Osswald H, Gleiter C, Mühlbauer B. Therapeutic use of theophylline.
in several subgroups of patients with HF: (1) patients with Clin Nephrol. 1995;43(suppl 1):S33–S37.
diuretic resistance; (2) patients with acutely decompensated HF 8. Yamagata T, Kobayashi T, Kusaka H, Karasawa A. Diuretic effects of
with worsening of GFR; (3) patients with chronic HF requiring KW-3902, a novel adenosine A1-receptor antagonist, in anesthetized
dogs. Biol Pharm Bull. 1994;17:1599 –1603.
long-term treatment with diuretics; and (4) patients with HF with 9. Wilcox CS, Welch WJ, Schreiner GF, Belardinelli L. Natriuretic and
a pronounced cardiac remodelling or fibrosis.17,18 diuretic actions of a highly selective adenosine A1 receptor antagonist.
With respect to acutely decompensated HF, it is important J Am Soc Nephrol. 1999;10:714 –720.
that SLV320 did not alter total peripheral resistance, whereas 10. Bak M, Thomsen K. Effects of the adenosine A1 receptor inhibitor FK
838 on proximal tubular fluid output in rats. Nephrol Dial Transplant.
furosemide treatment is known to increase total peripheral 2004;19:1077–1082.
resistance37,38 as shown in the study. Increase of total periph- 11. Becker BF, Zahler S, Seligmann C, Kupatt C, Habazettl H. [Interaction of
eral resistance in patients with an acutely failing heart is adenosine with leukocytes and thrombocytes]. Z Kardiol. 1996;85:
161–170 [in German].
clearly an adverse effect of loop diuretics. However, up to 12. Raschke P, Becker BF. Adenosine and PAF dependent mechanisms lead
now, there was no alternative approach. to myocardial reperfusion injury by neutrophils after brief ischaemia.
Our study is the first to show that diuresis can be achieved Cardiovasc Res. 1995;29:569 –576.
in patients with HF without compromising total peripheral 13. Forman MB, Vitola JV, Velasco CE, Murray JJ, Dubey RK, Jackson EK.
Sustained reduction in myocardial reperfusion injury with an adenosine
resistance. receptor antagonist: possible role of the neutrophil chemoattractant
However, it is important to consider that the diuretic response. J Pharmacol Exp Ther. 2000;292:929 –938.
property of adenosine A1 receptor antagonists is much 14. Gottlieb SS, Brater DC, Thomas I, Havranek E, Bourge R, Goldman S,
Dyer F, Gomez M, Bennett D, Ticho B, Beckman E, Abraham WT.
weaker compared with loop diuretics. Thus, these drugs will BG9719 (CVT-124), an A1 adenosine receptor antagonist, protects
most likely be added to loop diuretics, especially in patients against the decline in renal function observed with diuretic therapy.
with acute HF. Additional studies are needed to analyze the Circulation. 2002;105:1348 –1353.
cardiac hemodynamic effects in patients with HF receiving 15. Gottlieb SS, Skettino SL, Wolff A, Beckman E, Fisher ML, Freuden-
berger R, Gladwell T, Marshall J, Cines M, Bennett D, Liittschwager EB.
both loop diuretics and A1 adenosine receptor antagonists. Effects of BG9719 (CVT-124), an A1-adenosine receptor antagonist, and
In conclusion, A1 adenosine receptor antagonism has no furosemide on glomerular filtration rate and natriuresis in patients with
immediate hemodynamic effects in patients with HF. How- congestive heart failure. J Am Coll Cardiol. 2000;35:56 –59.
16. Felker MG, O’Connor CM, Braunwald E; for the Heart. Failure Clinical
ever, one dose (10 mg) of SLV320 improved kidney function Research Network Investigators. Loop diuretics in acute decompensated
(this statement is based on a positive F test followed by heart failure: necessary? evil? necessary evil? Circ Heart Fail. 2009;2:
pairwise comparisons). Thus, we conclude that A1 adenosine 56 – 62.
17. Kalk P, Eggert B, Relle K, Godes M, Heiden S, Sharkovska Y, Fischer Y,
receptor antagonism by SLV320 might improve kidney func-
Ziegler D, Bielenberg GW, Hocher B. The adenosine A1 receptor antag-
tion compared with furosemide while simultaneously promot- onist SLV320 reduces myocardial fibrosis in rats with 5/6 nephrectomy
ing natriuresis and diuresis in patients with HF. SLV320 had without affecting blood pressure. Br J Pharmacol. 2007;151:1025–1032.

Downloaded from http://circheartfailure.ahajournals.org/ at BIBL DE L'UNIV LAVAL on July 3, 2015

 Mitrovic et al A1 Adenosine Receptor Antagonist and Heart Failure 531

18. Funakoshi H, Chan TO, Good JC, Libonati JR, Piuhola J, Chen X, 27. Devarajan P. Emerging biomarkers of acute kidney injury. Contrib
MacDonnell SM, Lee LL, Herrmann DE, Zhang J, Martini J, Palmer TM, Nephrol. 2007;156:203–212.
Sanbe A, Robbins J, Houser SR, Koch WJ, Feldman AM. Regulated 28. Baas MC, Bouman CS, Hoek FJ, Krediet RT, Schultz MJ Cystatin C in
overexpression of the A1-adenosine receptor in mice results in adverse critically ill patients treated with continuous venovenous hemofiltration.
but reversible changes in cardiac morphology and function. Circulation. Hemodial Int. 2006;10(suppl 2):S33–S37.
2006;114:2240 –2250. 29. Mazur MJ, Heilman RL. Early detection of acute renal failure by serum
19. Greenberg B, Thomas I, Banish D, Goldman S, Havranek E, Massie BM, cystatin C: a new opportunity for a hepatologist. Liver Transpl. 2005;11:
Zhu Y, Ticho B, Abraham WT. Effects of multiple oral doses of an A1 705–707.
adenosine antagonist, BG9928, in patients with heart failure: results of a 30. Westhuyzen J. Cystatin C: a promising marker and predictor of impaired
placebo-controlled, dose-escalation study. J Am Coll Cardiol. 2007;50: renal function. Ann Clin Lab Sci. 2006;36:387–394.
600 – 606. 31. Villa P, Jimenez M, Soriano MC, Manzanares J, Casasnovas P. Serum
20. Givertz MM, Massie BM, Fields TK, Pearson LL, Dittrich HC; CKI-201 cystatin C concentration as a marker of acute renal dysfunction in crit-
and CKI-202 Investigators. The effects of KW-3902, an adenosine ically ill patients. Crit Care. 2005;9:R139 –R143.
A1-receptor antagonist, on diuresis and renal function in patients with 32. Ahlstrom A, Tallgren M, Peltonen S, Pettila V. Evolution and predictive
acute decompensated heart failure and renal impairment or diuretic power of serum cystatin C in acute renal failure. Clin Nephrol. 2004;62:
344 –350.
resistance. J Am Coll Cardiol. 2007;50:1551–1560.
33. Listos J, Talarek S, Fidecka S. Adenosine receptor agonists attenuate the
21. Dittrich HC, Gupta DK, Hack TC, Dowling T, Callahan J, Thomson S.
development of diazepam withdrawal-induced sensitization in mice. Eur
The effect of KW-3902, an adenosine A1 receptor antagonist, on renal
J Pharmacol. 2008;588:72–77.
function and renal plasma flow in ambulatory patients with heart failure
34. Namyar S, Mirnajafi-Zadeh J, Fathollahi Y, Zeraati M. The role of
and renal impairment. J Card Fail. 2007;13:609 – 617.
piriform cortex adenosine A1 receptors on hippocampal kindling. Can
22. Vallon V, Muhlbauer B, Osswald H. Adenosine and kidney function. J Neurol Sci. 2008;35:226 –231.
Physiol Rev. 2006;86:901–940. 35. Bekar L, Libionka W, Tian GF, Xu Q, Torres A, Wang X, Lovatt D,
23. Shlipak MG, Katz R, Sarnak MJ, Fried LF, Newman AB, Stehman-Breen Williams E, Takano T, Schnermann J, Bakos R, Nedergaard M. Adeno-
C, Seliger SL, Kestenbaum B, Psaty B, Tracy RP, Siscovick DS. Cystatin sine is crucial for deep brain stimulation-mediated attenuation of tremor.
C and prognosis for cardiovascular and kidney outcomes in elderly Nat Med. 2008;14:75– 80.
persons without chronic kidney disease. Ann Intern Med. 2006;145: 36. Solenkova NV, Solodushko V, Cohen MV, Downey JM. Endogenous
237–246. adenosine protects preconditioned heart during early minutes of reper-
24. Shlipak MG, Praught ML, Sarnak MJ. Update on cystatin C: new insights fusion by activating Akt. Am J Physiol Heart Circ Physiol. 2006;290:
into the importance of mild kidney dysfunction. Curr Opin Nephrol H441–H449.
Hypertens. 2006;15:270 –275. 37. Goldsmith SR. Current treatments and novel pharmacologic treatments
25. Shlipak MG, Katz R, Fried LF, Jenny NS, Stehman-Breen CO, Newman for hyponatremia in congestive heart failure. Am J Cardiol. 2005;95:
AB, Siscovick D, Psaty BM, Sarnak MJ. Cystatin-C and mortality in 14B–23B.
elderly persons with heart failure. J Am Coll Cardiol. 2005;45:268 –271. 38. Carr DB, Gavrila D, Brateng D, Easterling TR. Maternal hemodynamic
26. Trof RJ, Di Maggio F, Leemreis J, Groeneveld AB. Biomarkers of acute changes associated with furosemide treatment. Hypertens Pregnancy.
renal injury and renal failure. Shock. 2006;26:245–253. 2007;26:173–178.

CLINICAL PERSPECTIVE
Heart failure (HF) is still one of the most common reasons for hospitalization in developed countries with an aging
population worldwide. A substantial proportion of patients with acute HF show significant renal dysfunction. Treatment
of congestive HF itself is limited by worsening renal function, despite persistent volume overload. This connection between
HF and renal dysfunction has been termed the cardiorenal syndrome and has made treatment of patients with stable and
unstable HF difficult. Blocking the tubuloglomerular feedback mechanism with adenosine A1 receptor antagonists such as
SLV320 seems to improve diuresis and sodium excretion without compromising glomerular filtration rate in patients with
HF. This study confirmed existing clinical data in patients with chronic HF, however, by use of, for the first time in a
hemodynamic chronic HF clinical study, a biomarker for glomerular filtration rate: cystatin C. By using this biomarker,
we demonstrated a fast and sustained improvement of GFR after SLV320 infusion. Thus, cystatin C might be a biomarker
even suitable to describe acute renal effect of new drugs in clinical research. This study also analyzed, for the first time,
direct cardiac effects of A1 adenosine receptor antagonists. We demonstrated that SLV320 has no direct effects on heart
rate, blood pressure, pulmonary capillary wedge pressure, mean pulmonary arterial pressure, systemic vascular resistance,
right atrial pressure, and cardiac output. We concluded that SLV320 infusion shows no immediate effects on cardiac
hemodynamics. SLV320 might improve glomerular filtration rate while simultaneously promoting natriuresis and diuresis.
It remains to be demonstrated that these immediate renal effects translate into clinical benefits on the long run.

Downloaded from http://circheartfailure.ahajournals.org/ at BIBL DE L'UNIV LAVAL on July 3, 2015

Cardio-Renal Effects of the A1 Adenosine Receptor Antagonist SLV320 in Patients With
Heart Failure
Veselin Mitrovic, Petar Seferovic, Slobodan Dodic, Mirjana Krotin, Aleksander Neskovic,
Kenneth Dickstein, Hanka de Voogd, Christiane Böcker, Dieter Ziegler, Michael Godes,
Roumen Nakov, Hans Essers, Cees Verboom and Berthold Hocher

Circ Heart Fail. 2009;2:523-531; originally published online September 24, 2009;
doi: 10.1161/CIRCHEARTFAILURE.108.798389
Circulation: Heart Failure is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX
75231
Copyright © 2009 American Heart Association, Inc. All rights reserved.
Print ISSN: 1941-3289. Online ISSN: 1941-3297

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circheartfailure.ahajournals.org/content/2/6/523

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation: Heart Failure can be obtained via RightsLink, a service of the Copyright Clearance Center, not
the Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints

Subscriptions: Information about subscribing to Circulation: Heart Failure is online at:

http://circheartfailure.ahajournals.org//subscriptions/

Downloaded from http://circheartfailure.ahajournals.org/ at BIBL DE L'UNIV LAVAL on July 3, 2015