ORIGINAL CONTRIBUTION

Effect of Spironolactone on Diastolic Function

and Exercise Capacity in Patients
WithHeartFailureWithPreservedEjectionFraction
The Aldo-DHF Randomized Controlled Trial
Frank Edelmann, MD
Rolf Wachter, MD
Albrecht G. Schmidt, MD
Elisabeth Kraigher-Krainer, MD
Caterina Colantonio, MD
Wolfram Kamke, MD
Andre Duvinage, MD
Raoul Stahrenberg, MD
Kathleen Durstewitz, MD
Markus Lo ffler, MD
Hans-Dirk Du ngen, MD
Carsten Tscho pe, MD
Christoph Herrmann-Lingen, MD
Martin Halle, MD
Gerd Hasenfuss, MD
Go tz Gelbrich, PhD
Burkert Pieske, MD
for the Aldo-DHF Investigators

EART FAILURE (HF) WITH

preserved ejection fraction
(EF) accounts for more
than 50% of the total HF
population.1 Community-based cohort
studies have shown that mortality
rates are similar in HF with preserved
EF compared with HF with reduced
EF,1 but data from large clinical trials
point toward a better outcome in HF
with preserved EF. This may indicate
that comorbidities that are typically
excluded in trials may contribute to
the poor prognosis in HF with preserved EF.1-6 Left ventricular diastolic
dysfunction and adverse cardiac
remodeling are considered major

For editorial comment see p 825.

Importance Diastolic heart failure (ie, heart failure with preserved ejection fraction)
is a common condition without established therapy, and aldosterone stimulation may
contribute to its progression.
Objective To assess the efficacy and safety of long-term aldosterone receptor blockade in heart failure with preserved ejection fraction. The primary objective was to determine whether spironolactone is superior to placebo in improving diastolic function and
maximal exercise capacity in patients with heart failure with preserved ejection fraction.
Design and Setting The Aldo-DHF trial, a multicenter, prospective, randomized,
double-blind, placebo-controlled trial conducted between March 2007 and April 2012
at 10 sites in Germany and Austria that included 422 ambulatory patients (mean age,
67 [SD, 8] years; 52% female) with chronic New York Heart Association class II or III
heart failure, preserved left ventricular ejection fraction of 50% or greater, and evidence of diastolic dysfunction.
Intervention Patients were randomly assigned to receive 25 mg of spironolactone
once daily (n=213) or matching placebo (n=209) with 12 months of follow-up.
Main Outcome Measures The equally ranked coprimary end points were changes
in diastolic function (E/e') on echocardiography and maximal exercise capacity (peak
V O2) on cardiopulmonary exercise testing, both measured at 12 months.
Results Diastolic function (E/e') decreased from 12.7 (SD, 3.6) to 12.1 (SD, 3.7) with spironolactone and increased from 12.8 (SD, 4.4) to 13.6 (SD, 4.3) with placebo (adjusted
mean difference, 1.5; 95% CI, 2.0 to 0.9; P.001). Peak V O2 did not significantly
change with spironolactone vs placebo (from 16.3 [SD, 3.6] mL/min/kg to 16.8 [SD, 4.6]
mL/min/kg and from 16.4 [SD, 3.5] mL/min/kg to 16.9 [SD, 4.4] mL/min/kg, respectively;
adjusted mean difference, 0.1 mL/min/kg; 95% CI, 0.6 to 0.8 mL/min/kg; P=.81).
Spironolactone induced reverse remodeling (left ventricular mass index declined; difference, 6 g/m2; 95% CI, 10 to1 g/m2; P=.009) and improved neuroendocrine activation (N-terminal probrain-type natriuretic peptide geometric mean ratio, 0.86; 95%
CI, 0.75-0.99; P=.03) but did not improve heart failure symptoms or quality of life and
slightly reduced 6-minute walking distance (15 m; 95% CI, 27 to 2 m; P=.03). Spironolactone also modestly increased serum potassium levels (0.2 mmol/L; 95% CI, 0.1 to
0.3; P.001) and decreased estimated glomerular filtration rate (5 mL/min/1.73 m2;
95% CI, 8 to 3 mL/min/1.73 m2; P.001) without affecting hospitalizations.
Conclusions and Relevance In this randomized controlled trial, long-term aldosterone receptor blockade improved left ventricular diastolic function but did not affect
maximal exercise capacity, patient symptoms, or quality of life in patients with heart
failure with preserved ejection fraction. Whether the improved left ventricular function observed in the Aldo-DHF trial is of clinical significance requires further investigation in larger populations.
Trial Registration clinicaltrials.gov Identifier: ISRCTN94726526; Eudra-CT No: 2006002605-31
JAMA. 2013;309(8):781-791
Author Affiliations are listed at the end of this
article.
A complete list of the Aldo-DHF Investigators appears in the eAppendix.

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

www.jama.com
Corresponding Author: Burkert Pieske, MD, Department of Cardiology, Medical University Graz, Auenbruggerplatz 15, A-8010 Graz, Austria (burkert.pieske
@medunigraz.at).
JAMA, February 27, 2013Vol 309, No. 8 781

SPIRONOLACTONE IN PATIENTS WITH HEART FAILURE

METHODS

Figure 1. Participant Flow

Trial Design and Oversight

795 Patients aged 50 y screened

373 Excluded
197 Dropped out of rescreening
67 Inclusion criteria not confirmed
80 Refused participation
29 Physician decision

422 Randomized

209 Randomized to placebo

209 Received placebo
as assigned

213 Randomized to spironolactone

213 Received spironolactone
as assigned

10 Discontinued study
1 Physician decision
8 Withdrew consent
1 Lost to follow-up

5 Discontinued study
4 Withdrew consent
1 Lost to follow-up

6-mo Follow-up

6-mo Follow-up

199 Remained in study

199 Attended 6-mo visit
196 Had echocardiographic
data available
187 Had spiroergometric data
available

208 Remained in study

206 Attended 6-mo visit
206 Had echocardiographic
data available
194 Had spiroergometric data
available

3 Discontinued study
1 Withdrew consent
2 Lost to follow-up

4 Discontinued study
1 Died
2 Withdrew consent
1 Lost to follow-up

12-mo Follow-up

12-mo Follow-up

196 Attended 12-mo visit

195 Had echocardiographic
data available
187 Had spiroergometric data
available

204 Attended 12-mo visit

203 Had echocardiographic
data available
187 Had spiroergometric data
available

209 Included in analysis of the

primary outcomes

213 Included in analysis of the

primary outcomes

underlying pathologies in HF with

preserved EF.7 However, pharmacotherapies tested to date have not
shown improvements in diastolic dysfunction, cardiac remodeling, or cardiovascular outcome.3-6
Mineralocorticoid receptor activation by aldosterone contributes to the
pathophysiology of HF (regardless of
EF) through several mechanisms, including sodium retention, potassium
loss, endothelial dysfunction, vascular inflammation, fibrosis, and hypertrophy.8,9 The mineralocorticoid receptor antagonists spironolactone and
eplerenone reduce total and cardiovascular mortality across the spectrum of
HF with reduced EF and in patients
with acute myocardial infarction com782

JAMA, February 27, 2013Vol 309, No. 8

plicated by left ventricular dysfunction and heart failure.10-12 Although

small preliminary studies suggest that
mineralocorticoid receptor antagonists might also be effective in hypertensive heart disease and diastolic heart
failure, no adequately powered clinical trials have been conducted to investigate the effects of chronic mineralocorticoid receptor blockade on
structural and functional end points in
HF with preserved EF.13,14
The Aldosterone Receptor Blockade in Diastolic Heart Failure (AldoDHF) trial was designed to address this
knowledge gap by evaluating the effects of spironolactone on diastolic
function and exercise capacity in patients with HF with preserved EF.15

The Aldo-DHF trial was a multicenter, randomized, placebo-controlled, double-blind, two-armed, parallel-group study that enrolled patients
from 10 trial sites in Germany (GE) and
Austria (AT). The study design has been
previously published.15 The protocol
and amendments were approved by the
institutional review board at each participating center, and the trial was conducted in accordance with the principles of the Declaration of Helsinki,
Good Clinical Practice guidelines, and
local and national regulations. Written informed consent was provided by
all patients before any study-related procedures were performed.
The trial was designed and implemented by the principal investigator,
the study coordinator, core laboratory
guarantors, and the Coordination Center for Clinical Trials Leipzig (University of Leipzig). The Coordination Center for Clinical Trials Leipzig was
responsible for all aspects related to site
monitoring, data collection, and data
management. An independent data and
safety monitoring board reviewed the
safety data on an ongoing, predefined
basis throughout the trial. Patients, the
investigator team, individuals performing the assessments, and data analysts
remained blinded to the identity of
treatment until after database lock;
analyses were performed according to
a predefined statistical analysis plan.
Participants

The complete eligibility criteria have

been published. 15 Briefly, men and
women aged 50 years or older were eligible to participate in the study if they
had current heart failure symptoms consistent with New York Heart Association (NYHA) class II or III, left ventricular ejection fraction (LVEF) of 50% or
greater, echocardiographic evidence of
diastolic dysfunction (grade I) or atrial
fibrillation at presentation, and maximum exercise capacity (peak V O2) of 25
mL/kg/min or less.
Major exclusion criteria included
prior documented reduced left ven-

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

SPIRONOLACTONE IN PATIENTS WITH HEART FAILURE

tricular ejection fraction (LVEF 40%),

significant coronary artery disease (current angina pectoris or ischemia on
stress tests; untreated coronary stenosis 50%), myocardial infarction or
coronary artery bypass graft surgery 3
months or less prior to enrollment,
clinically relevant pulmonary disease
(vital capacity 80% or forced expiratory volume in 1 second 80% of reference values on spirometry), significant laboratory abnormalities
(potassium 5.1 mmol/L; hemoglobin 11 g/dL; hematocrit 33%; serum creatinine 1.8 mg/dL; or estimated glomerular filtration rate
[eGFR]30 mL/min/1.73 m2, calculated using the Modification of Diet in
Renal Disease formula: 186 [serum
creatinine {in micromoles per liter}/
88.4]1.154 age [in years]0.203
1.21 [if patient is black]0.742 [if
patient is female]), known contraindications for spironolactone or known intolerance to or therapy with a mineralocorticoid receptor antagonist within
the last 3 months, concomitant therapy
with a potassium-sparing diuretic (eg,
triamterene, amiloride), or potassium
supplementation.
Study Drug Administration
and Study Procedures

Eligible patients were randomly assigned to receive either spironolactone

(25 mg/d) or matching placebo. The randomization ratio was 1:1 for spironolactone or placebo using the Pocock minimization algorithm.16 Randomization
was stratified by grade of diastolic dysfunction (I vs II-III), rhythm (sinus vs
other), and study center. The allocation
sequence was implemented remotely via
Internet/fax by the Coordination Center for Clinical Trials Leipzig. Standard
therapies for risk factor and symptom
control were at the discretion of treating physicians and required to be unchanged within the 2 weeks prior to
randomization.
Production of identical matching placebo and quality control, packaging, labeling, storage, and dispensing of both
spironolactone and placebo were performed by Allphamed PHARBIL. The

first dose of study drug was administered immediately after randomization under the supervision of the local
investigator. No further up-titration was
planned.
The study drug could be decreased
temporarily to 25 mg every other day for
a potassium level greater than 5.2 mmol/L
or in the presence of other reversible,
nonlife-threatening adverse effects. For
safety reasons, study medication was
stopped for relevant hyperkalemia (se-

rum potassium 5.5 mmol/L) and/or

hyperkalemia-associated clinical symptoms, significant renal impairment (serum creatinine 2.5 mg/dL; eGFR 20
mL/min/1.73 m2), significant breast pain
or gynecomastia, or withdrawal of
informed consent; rechallenge was
encouraged wherever possible. Adherence was assessed at all regularly scheduled visits.
Patients were followed up while receiving blinded study medication for 12

Table 1. Demographic and Clinical Baseline Characteristics a

Characteristics

Total
(n = 422)

Demographics
Age, mean (SD), y
67 (8)
Female
221 (52)
Medical history
Hospitalization for heart failure in past 12 mo
156 (37)
Coronary heart disease
170 (40)
Hypertension
387 (92)
Hyperlipidemia
273 (65)
Diabetes mellitus
70 (17)
Chronic obstructive pulmonary disease
14 (3)
Atrial fibrillation
22 (5)
Physical examination, mean (SD)
28.9 (3.6)
Body mass index b
Systolic blood pressure, mm Hg
135 (18)
Diastolic blood pressure, mm Hg
79 (11)
Heart rate, /min
65 (13)
Signs and symptoms
NYHA functional class
II
363 (86)
III
59 (14)
Peripheral edema
165 (39)
Nocturia
338 (80)
Paroxysmal nocturnal dyspnea
67 (16)
Nocturnal cough
61 (15)
Fatigue
249 (59)
Laboratory measures
Sodium, mmol/L
140.3 (3.0)
Potassium, mean (SD), mmol/L
4.2 (0.4)
Hemoglobin, mean (SD), g/dL
13.8 (1.2)
eGFR, mean (SD), mL/min/1.73 m2
79 (19)
NT-proBNP, median (IQR), ng/L
158 (83-299)
Current medications
ACE inhibitors/angiotensin receptor
325 (77)
antagonists
-Blockers
302 (72)
Diuretics
227 (54)
Calcium antagonists
105 (25)
Lipid-lowering drugs
230 (55)

Placebo Group
(n = 209)

Spironolactone
Group (n = 213)

67 (8)
110 (53)

67 (8)
111 (52)

75 (36)
78 (37)
190 (91)
143 (68)
34 (16)
3 (1)
9 (4)

81 (38)
92 (43)
197 (92)
130 (61)
36 (17)
11 (5)
13 (6)

28.9 (3.6)
135 (18)
80 (12)
64 (12)

28.9 (3.6)
135 (18)
79 (10)
66 (14)

183 (88)
26 (12)
84 (40)
168 (80)
31 (15)
31 (15)
118 (56)

180 (85)
33 (15)
81 (38)
170 (80)
36 (17)
30 (14)
131 (62)

140.3 (2.7)
4.2 (0.4)
13.8 (1.3)
78 (18)
148 (80-276)

140.3 (3.3)
4.2 (0.4)
13.8 (1.2)
79 (19)
179 (81-276)

158 (76)

167 (78)

156 (75)
109 (52)
58 (28)
118 (56)

146 (69)
118 (55)
47 (22)
112 (53)

Abbreviations: ACE, angiotensin-converting enzyme; eGFR, estimated glomerular filtration rate calculated by Modification
of Diet in Renal Disease formula: 186 (serum creatinine [in micromoles per liter]/88.4) 1.154 age (in years)
0.2031.21 (if patient is black)0.742 (if patient is female); IQR, interquartile range; NT-proBNP, N-terminal probraintype natriuretic peptide; NYHA, New York Heart Association.
a Data are expressed as No. (%) unless otherwise specified.
b Body mass index is defined as weight in kilograms divided by height in meter squared.

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

JAMA, February 27, 2013Vol 309, No. 8 783

SPIRONOLACTONE IN PATIENTS WITH HEART FAILURE

months plus an additional safety period

of 4 weeks after termination of individual study-related therapy (interim visits at 1 week and 3, 6, and 9 months).
At baseline and the 6- and 12-month follow-up visits, all patients underwent
physical examination, echocardiography, cardiopulmonary exercise testing,
6-minute walk testing, quality-of-life assessment, and blood sampling. Quality

of life was assessed by the 36-Item Short

Form Health Survey (SF-36), the Minnesota Living With Heart Failure Questionnaire, the Patient Health Questionnaire, and the Hospital Anxiety and
Depression Scale.
Study Objectives and End Points

The primary objective of the AldoDHF trial was to determine whether spi-

Table 2. Echocardiographic, Exercise Testing, and Quality-of-Life Baseline Characteristics a

Characteristics
Echocardiography
LV ejection fraction, %
LV diameter (end diastolic), mm
LV diameter (end systolic), mm
LV mass index, g/m2
Men
Women
Left atrial volume index, mL/m2
E-wave velocity, cm/s
Medial e' wave velocity, cm/s
E/e' (medial) velocity ratio
E/A velocity ratio
Isovolumic relaxation time, ms
Deceleration time, ms
Grade of diastolic dysfunction, No. (%) b
I
II
III
IV
Paulus criteria positive, No. (%)
Cardiopulmonary exercise testing
Duration of exercise, s
Peak V O2, mL/min/kg
ATV O2, mL/min/kg
V E/V CO2 slope
Borg scale
Six-minute walk test
Walk distance, m
Quality of life
Responded to questionnaire, No. (%)
Minnesota Living With Heart Failure Questionnaire,
total score
SF-36 Physical Functioning Scale score
SF-36 Global Self-Assessment score
Symptoms of depression (PHQ-9 summary score)
HADS anxiety score
HADS depression score

Total
Placebo Group Spironolactone
(n = 422)
(n = 209)
Group (n = 213)
67 (8)
46.5 (6.2)
25.5 (6.4)
109 (28)
117 (31)
101 (23)
28.0 (8.4)
73 (19)
5.9 (1.3)
12.8 (4.0)
0.91 (0.33)
89 (26)
243 (63)

68 (7)
46.9 (6.0)
25.8 (6.7)
109 (27)
118 (29)
102 (22)
27.8 (7.7)
74 (21)
6.0 (1.4)
12.8 (4.4)
0.92 (0.34)
88 (25)
247 (66)

67 (8)
46.2 (6.4)
25.2 (6.2)
108 (29)
116 (33)
100 (23)
28.2 (9.1)
72 (17)
5.9 (1.3)
12.7 (3.6)
0.90 (0.31)
89 (26)
239 (60)

307 (77)
86 (21)
4 (1)
3 (1)
220 (52)

151 (75)
44 (22)
2 (1)
3 (2)
109 (52)

156 (78)
42 (21)
2 (1)
0
111 (52)

540 (176)
16.4 (3.5)
11.6 (3.2)
30.3 (5.2)
5.4 (3.7)

545 (176)
16.4 (3.5)
11.4 (3.0)
30.7 (5.8)
5.4 (2.1)

535 (176)
16.3 (3.6)
11.9 (3.4)
30.0 (4.6)
5.4 (4.8)

530 (87)

531 (86)

529 (88)

388 (92)
22 (16)

194 (93)
21 (15)

194 (91)
22 (16)

63 (22)
3.4 (0.6)
5.6 (4.1)
5.3 (3.8)
4.7 (3.6)

63 (23)
3.4 (0.6)
5.7 (4.3)
5.32 (3.8)
4.7 (3.6)

62 (22)
3.4 (0.6)
5.6 (3.9)
5.28 (3.7)
4.8 (3.6)

Abbreviations: A, peak atrial transmitral ventricular filling velocity; ATV O2, oxygen consumption at anaerobic threshold; e',
early diastolic tissue Doppler velocity; E, peak early transmitral ventricular filling velocity; HADS, Hospital Anxiety and Depression Scale; LV, left ventricular; PHQ-9, 9-item depression scale of the Patient Health Questionnaire; SF-36, 36-Item
Short Form Health Survey; V CO2, volume of expired carbon dioxide; V E, expired volume per unit time; V O2, oxygen consumption.
a Data are expressed as mean (SD) unless otherwise specified. Higher values indicate better performance for LV ejection
fraction, medial e wave velocity, duration of exercise, peak V O2, ATVO2, Borg scale, walk distance, SF-36 Physical Functioning Scale, and SF-36 global self assessment. Lower values indicate better performance for left atrial volume index,
E/e' (medial) velocity ratio, grade of diastolic dysfunction, V E/V CO2 slope, Minnesota Living With Heart Failure Questionnaire total score, symptoms of depression (PHQ-9 summary score), HADS anxiety score, and HADS depression score.
b Data not measured because of presence of atrial fibrillation: placebo, n=9 (4%); spironolactone, n=13 (6%).

784

JAMA, February 27, 2013Vol 309, No. 8

ronolactone is superior to placebo in

improving diastolic function and maximal exercise capacity in patients with
HF with preserved EF.
The change in E/e' (ie, the relation
of peak early transmitral ventricular filling velocity to early diastolic tissue Doppler velocity as an echocardiographic
estimate of filling pressure) at 12
months and the change in maximum
exercise capacity (peak V O2 on cardiopulmonary exercise testing) at 12
months compared with baseline were
coequal primary end points.
Prespecified major and additional
secondary end points included changes
in echocardiographic measures of cardiac function and remodeling, measures of submaximal and maximal exercise capacity, serum biomarkers, and
quality of life. Clinical tolerability was
assessed as the safety end point. Morbidity and mortality (all-cause and cardiovascular-specific) were also predefined exploratory end points.
End Point Assessments

Echocardiography. Detailed echocardiography was performed as described

previously.15 All echocardiographic data
were reviewed and confirmed at a blinded
core laboratory. A standard operating procedure for obtaining all echocardiographic measurements was released
before recruitment began, and all participating investigators were trained and
certified by the core laboratory staff. Stability of the coprimary end point E/e'
(variation 20% between screening and
baseline visits) was a required inclusion
criterion. Diastolic dysfunction was prospectively identified and graded by a prespecified algorithm defined in the study
protocol as previously described,15 and
diagnostic criteria for HF with normal EF
were used according to current European Society of Cardiology recommendations.17 Detailed definitions can be
found in the eAppendix (available at http:
//www.jama.com).
Cardiopulmonary Exercise Testing.
Standardized cardiopulmonary exercise testing was performed as described
previously.15 All data including validity
criteria and protocol adherence were re-

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

SPIRONOLACTONE IN PATIENTS WITH HEART FAILURE

viewed and confirmed at a blinded core

laboratory. An standard operating procedure was released before recruitment
began, and all participating investigators were trained and certified by the cardiopulmonary exercise testing core laboratory staff. Peak V O2 was prospectively
defined as the maximum value of the last
three 10-second averages during exercise. The variation in peak V O2 was required to be 15% or less between screening and baseline.
Quality of Life. The following validated self-rating scales were used for assessing quality of life: the SF-36,18 the
Minnesota Living With Heart Failure
Questionnaire,19 the Patient Health
Questionnaire,20 and the Hospital Anxiety and Depression Scale.21 Minimal
clinically important differences have not
been established previously for most of
these instruments, and published reports show large variations in methods and suggested minimal clinically
important differences, even for identical subscales of the SF-36.22,23 Hence,
the clinical importance of changes in
these instruments is subject to future
research.
Laboratory Measurements. Venous
blood samples were drawn under standardized conditions after 20 minutes of
rest in the supine position. Samples
were immediately cooled, centrifuged,
and processed for storage at 80C
(112F). N-terminal probrain-type
natriuretic peptide (NT-proBNP) was
analyzed with the Elecsys NT-proBNP
immunoassay (Roche Diagnostics).
Statistical Analyses

For sample size calculation, a gain in

peak V O2 of 2 mL/min/kg and a decrease in E/e' of 1.2 by spironolactone
treatment were assumed. Withingroup standard deviations were expected to be 5 mL/min/kg for V O2 and
3 for E/e'. Thus, expected mean differences were 0.4 SDs for both primary end
points. Global type I and II error rates
were set at .05 and .1, respectively (ie,
.025 and .05 for each of the 2 primary
tests). Thus, 380 evaluable patients (190
in each treatment group) were needed.
Assuming a dropout rate of 10%, re-

cruitment of 420 patients was planned.

Analyses of the primary end points
were carried out using 2-sided MannWhitney U tests. All analyses were
based on the intention-to-treat principle. Patients who died were assigned
the worst rank in both tests, reflecting
the worst case of functional loss. A missing value in one of the end points did
not preclude analysis of the other. Sensitivity analyses examining the possible effect of missing data on the primary results were performed using last
observation carried forward and multiple imputation. Quantitative effects of
the intervention were assessed by analysis of covariance with the follow-up
value as the dependent variable, treatment as a factor, and the baseline value
as the covariate. SPSS version 20 (SPSS
Inc) was used as statistical software.
Prespecified secondary end points
were changes in anaerobic threshold,
slope of expired volume per unit time
to volume of expired carbon dioxide
E/V
CO2), and Borg dyspnea scale dur(V
ing cardiopulmonary exercise testing,
E and e' velocities, grade of diastolic
dysfunction, left ventricular mass index and left atrial volume index on echocardiography, 6-minute walk distance, NT-proBNP level, and the
patient-reported Minnesota Living With
Heart Failure Questionnaire total score,
SF-36 Physical Functioning scale score,
and symptoms of depression measured by the summary score on the
9-item depression scale of the Patient
Health Questionnaire. Changes in all
other variables, including changes in
baseline characteristics, were analyzed in an exploratory manner. No adjustments for multiple comparisons
were planned, except for the primary
end point.
Analyses were carried out by analysis of covariance, binary, or ordinal logistic regression with the follow-up
value as the dependent variable, treatment as a factor, and the baseline value
as the continuous or categorical covariate, as appropriate for quantitative, binary, or ordinal categorical variables.
Between-group comparisons are presented as mean differences or odds ra-

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

tios. N-terminal proBNP was analyzed

on the logarithmic scale, and the result was transformed back by the exponential function, leading to a geometric mean ratio instead of a mean
difference.
Safety end points were all-cause
death, cardiac and noncardiac hospitalizations, worsening heart failure
(worsening dyspnea and worsening or
new edema), coronary heart disease
(myocardial infarction, revascularization, or new symptoms of angina pectoris), significant renal impairment (decrease of eGFR to 30 mL/min/1.73 m2
or decrease of eGFR 15 mL/min/
1.73 m2 vs baseline) and anemia (de
novo, according to World Health Organization criteria, or decrease of hemoglobin levels 1 g/dL in anemic patients), increases in serum potassium
to higher than 5 mmol/L, occurrence
of gynecomastia, and self-reported intolerance of the study medication. Comparisons were carried out using the t test
for quantities and the Fisher exact test
for binary variables.
Prespecified subgroup analyses were
performed by analysis of covariance with
the follow-up value of the end point as
the dependent variable; respective baseline variable as covariate; and treatment, subgroup variable, and their interaction term as factor variables. Patients
were categorized into subgroups by each
of the following variables at baseline (median split for continuous variables): age,
sex, body mass index, systolic blood pressure, heart rate, NYHA class (II or III),
grade of diastolic function (I vs all other),
criteria for diastolic heart failure according to European Society of Cardiology
criteria (Paulus positive or negative),17
and eGFR.
RESULTS
Patients

Of 795 patients screened from March

2007 to April 2011, 422 were included and randomized to receive spironolactone or placebo (FIGURE 1). The
mean length of follow-up was 11.6
months (95% CI, 11.4-11.8 months)
and the mean daily dose of spironolactone was 21.6 mg (95% CI, 20.8-22.3
JAMA, February 27, 2013Vol 309, No. 8 785

SPIRONOLACTONE IN PATIENTS WITH HEART FAILURE

mg). Baseline characteristics were similar between treatment groups (TABLE 1,

TABLE 2, and eTable 1).
Primary End Points

Spironolactone significantly improved diastolic function compared

with placebo (FIGURE 2A). E/e' significantly declined with spironolactone

from 12.7 (SD, 3.6) to 12.1 (SD, 3.7)
and increased in the placebo group from
12.8 (SD, 4.4) to 13.6 (SD, 4.3)
(P .001 for difference between
groups). As shown in Figure 2A, this

effect was evident at 6 months and was

maintained at 12 months.
Peak V O2 increased from 16.3 (SD,
3.6) mL/min/kg to 16.8 (SD, 4.6) mL/
min/kg in the spironolactone group and
from 16.4 (SD, 3.5) mL/min/kg to 16.9
(SD, 4.4) mL/min/kg in the placebo

Figure 2. Equally Ranked CoPrimary End Points of Peak Early Transmitral Ventricular Filling Velocity to Early Diastolic Tissue Doppler Velocity
(E/e') and Peak Oxygen Consumption (V O2) According to Assigned Study Treatment

A E/e medial velocity ratio

P <.001

Placebo

P = .57

P = .81

6 mo

12 mo

Change in Peak VO2,

mL/min/kg

P <.001

Change in E/e

B Peak VO2

Spironolactone

1
Baseline

6 mo

12 mo

Baseline

Time Since Randomization

Time Since Randomization

Placebo
No. of patients
Mean E/e (95% CI)

209
12.8 (12.2-13.4)

196
13.2 (12.4-13.9)

195
13.6 (13.0-14.2)

Spironolactone
No. of patients
Mean E/e (95% CI)

213
12.7 (12.2-13.2)

206
11.9 (11.5-12.4)

203
12.1 (11.6-12.6)

Placebo
No. of patients
209
Mean peak VO2 (95 % CI), mL/min/kg 16.4 (15.9-16.9)
Spironolactone
No. of patients
213
Mean peak VO2 (95 % CI), mL/min/kg 16.4 (15.9-16.8)

187
16.4 (15.8-17.0)

187
16.9 (16.2-17.9)

194
16.0 (15.5-16.6)

187
16.8 (16.2-17.5)

Error bars indicate 95% CI. P values describe comparisons of the changes in the placebo or spironolactone group at the respective time point vs baseline. No further
improvement by spironolactone occurred between the 6-month and 12-month visits (P=.39 for E/e').

Table 3. Echocardiography Results After 12 Months

Spironolactone Placebo b
Measurements
Primary echocardiographic end point
E/e' (medial) velocity ratio
Secondary end points
E-wave velocity, cm/s

Placebo Group
(n = 195) a

Spironolactone Group
(n = 203) a

Difference (95% CI)

P Value

13.6 (13.0-14.2)

12.1 (11.6-12.6)

1.5 (2.0 to 0.9)

.001

73.6 (70.6-76.7)

70.5 (68.3-72.7)

2.5 (4.9 to 0.2)

.03

Medial e' wave velocity, cm/s

5.86 (5.65-6.06)

6.16 (5.94-6.37)

0.36 (0.13 to 0.60)

.002

E/A velocity ratio

0.96 (0.91-1.01)

0.91 (0.87-0.96)

0.04 (0.09 to 0.01)

Isovolumic relaxation time, ms

Deceleration time, ms
Grade of diastolic dysfunction, No. (%) c
No diastolic dysfunction

88 (85-92)

86 (82-90)

238 (229-247)

241 (232-249)

2 (1)

4 (2)

129 (68)

143 (75)

II

57 (30)

43 (23)

III
IV

0
2 (1)

6 (6 to 18)

0.63 (0.37 to 1.09) d

.08
.28
.32

.10

1 (1)
0

LV ejection fraction, %

65.9 (64.7-67.0)

67.2 (66.1-68.3)

LV mass index, g/m2

106 (102-110)

100 (96-103)

Left atrial volume index, mL/m2

27.6 (26.5-28.6)

Other variables
LV diameter (end diastolic), mm
LV diameter (end systolic), mm

3 (8 to 2)

1.6 (0.1 to 3.1)

.04

6 (10 to 1)

.009

27.5 (26.2-28.8)

0.4 (1.5 to 0.8)

.51

46.4 (45.5-47.3)

44.6 (43.7-45.4)

1.4 (2.5 to 0.3)

.01

26.5 (25.6-27.4)

25.5 (24.6-26.4)

0.6 (1.8 to 0.5)

.26

Abbreviations: A, peak atrial transmitral ventricular filling velocity; e', early diastolic tissue Doppler velocity; E, peak early transmitral ventricular filling velocity; LV, left ventricular.
a Data are expressed as groupwise mean (95% CI) unless otherwise indicated. Higher values indicate better performance for LV ejection fraction and medial e' wave velocity. Lower
values indicate better performance for left atrial volume index, E/e' (medial) velocity ratio, and grade of diastolic dysfunction.
b Between-group differences are from analysis of covariance, adjusting for baseline.
c Data not measured because of presence of atrial fibrillation: placebo, n=5; spironolactone, n=12.
d Odds ratio (95% CI).

786

JAMA, February 27, 2013Vol 309, No. 8

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

SPIRONOLACTONE IN PATIENTS WITH HEART FAILURE

Figure 3. Main Secondary End Points According to Assigned Study Treatment

A Left ventricular mass index
5

0.1

Placebo

Change in Log10
NT-proBNP, ng/L

Change in LVMI, g/m2

B Log10 NT-proBNP
P = .009

P = .16

5
Spironolactone

10

15

P = .09

P = .03

6 mo

12 mo

0.1

0.2
Baseline

6 mo

12 mo

Baseline

Time Since Randomization

Placebo
No. of patients
209
Mean LVMI (95% CI), g/m2 109 (106-113)
Spironolactone
No. of patients
212
Mean LVMI (95% CI), g/m2 108 (104-112)

195
109 (105-113)

193
106 (102-110)

206
105 (101-109)

201
100 (96-103)

Time Since Randomization

Placebo
No. of patients
195
Mean log10 NT-proBNP (95% CI), ng/L 2.19 (2.13-2.25)
Spironolactone
No. of patients
204
Mean log10 NT-proBNP (95% CI), ng/L 2.22 (2.15-2.28)

190
2.17 (2.11-2.24)

191
2.22 (2.15-2.28)

204
2.15 (2.08-2.22)

201
2.18 (2.11-2.25)

LVMI indicates left ventricular mass index; NT-proBNP, N-terminal probrain-type natriuretic peptide. Error bars indicate analysis of covariance estimates of treatment
effects within subgroups. P values describe comparisons of the changes in the placebo or spironolactone group at the respective time point vs baseline. No further
change by spironolactone occurred between the 6-month and 12-month visits (P=.16 for LVMI and P=.87 for log10 NT-proBNP).

group, without a difference between

groups (P=.81) (Figure 2B).
Results regarding primary end points
were consistent across prespecified subgroups (eFigure 1).
Secondary Echocardiographic
End Points

Spironolactone improved major measures of cardiac function and remodeling (TABLE 3). Left ventricular ejection
fraction increased while left ventricular
end-diastolic diameter and left ventricular mass index decreased significantly in
the spironolactone group compared placebo (FIGURE 3A). Other measures of
diastolic function or cardiac structure
did not differ between groups (Table 3
and eTable 2A), but NT-proBNP levels significantly decreased with spironolactone (Figure 3B and TABLE 4).
Secondary Exercise Performance
End Points

Results are shown in T ABLE 5 and

CO2 slightly
E/V
eTable 2B. The slope of V
increased, but there were no other significant differences between groups.
Secondary Clinical Outcome
End Points

Compared with placebo, spironolactone significantly reduced systolic blood

pressure (Table 4 and eFigure 2). Heart

failure symptoms assessed by NYHA class
(Table 4), quality of life, and depressive
symptoms did not differ between groups,
whereas 6-minute walking distance
slightly decreased in the spironolactone group (eFigure 3). Further clinical
variables are shown in eTable 2C.
Safety and Adherence

Safety and adherence data are shown in

TABLE 6. A mild increase in potassium
levels and a decrease in eGFR occurred
in the spironolactone group. These
changes were evident after 1 week of
therapy and remained stable thereafter
(Table 4, eFigure 4, and eFigure 5). A significantly greater proportion of patients randomized to spironolactone experienced potassium serum levels greater
than 5.0 mmol/L (Table 6), but there was
no difference between groups in the incidence of serious hyperkalemia (5.5
mmol/L) and no patients were hospitalized for hyperkalemia.
Of all adverse events, only gynecomastia (P=.003 for intolerance; P=.03 for
dose reduction) and an increase of potassium serum levels to greater than 5.0
mmol/L (P .001) during follow-up
were significantly associated with subjective intolerance or reduction or stopping of the study medication.

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

Loss to Follow-up

Analyses to examine the possible effects of missing data (eg, last observation carried forward or multiple imputation) did not alter the results on the
primary end points (eTable 4).
COMMENT
We evaluated the effect of adding spironolactone to recommended standard risk factor control in patients with
HF with preserved EF. We found that
left ventricular end-diastolic filling,24 left
ventricular remodeling, and neurohumoral activation were improved with
spironolactone, whereas maximal exercise capacity and quality-of-life measures remained unchanged.
Similar reverse remodeling effects of
spironolactone have been detected in patients with HF with reduced EF,25,26 a
condition in which spironolactone treatment also reduces all-cause mortality
and heart failure hospitalizations.10,12 In O2
terestingly, it does not improve peak V
or quality of life in this population.10,12,27,28 In a recent meta-analysis of
1575 patients with HF with reduced EF,
aldosterone antagonists improved NYHA
functional status by only 0.13 class, an
effect size our study was not powered
to detect.29 Whether mineralocorticoid
receptor antagonists may also improve
JAMA, February 27, 2013Vol 309, No. 8 787

SPIRONOLACTONE IN PATIENTS WITH HEART FAILURE

prognosis in HF with preserved EF is

currently being investigated in the international TOPCAT trial.30
Activation of the mineralocorticoid
receptor system by aldosterone promotes hypertension, endothelial dysfunction, left ventricular hypertrophy, and myocardial fibrosis. 8 , 9
Angiotensin-converting enzyme inhibi-

tors and angiotensin receptor blockers inhibit angiotensin IImediated aldosterone release, but despite optimized
therapy, a large proportion of patients
with heart failure have elevated aldosterone plasma levels (aldosterone
escape). 31 Spironolactone has been
shown to decrease extracellular matrix turnover and myocardial collagen

content, mechanisms known to influence the progression of heart failure.26,32,33

Aldo-DHF is the first large, multicenter trial to demonstrate structural reverse cardiac remodeling in patients
with symptomatic HF with preserved
EF treated in addition to substantial antineuroendocrine background therapy

Table 4. Clinical and Laboratory Results and Quality of Life After 12 Months
Spironolactone Placebo b
Measurements
Clinical secondary end point
Six-minute walk distance, m
Other clinical variables
NYHA class, No. (%)
I
II
III
Peripheral edema, No. (%)
Systolic blood pressure, mm Hg
Diastolic blood pressure, mm Hg
Heart rate, /min
Laboratory secondary end point
NT-proBNP, median (IQR), ng/L
Other laboratory measurements
Sodium, mmol/L
Potassium, mmol/L
Hemoglobin, g/dL
eGFR, mL/min/1.73 m2
Quality-of-life secondary end points
Minnesota Living With Heart Failure Questionnaire total score
SF-36 Physical Functioning Scale score
SF-36 Global Self-Assessment score
Symptoms of depression (PHQ-9 summary score)
HADS anxiety score
HADS depression score

Placebo Group
(n = 196) a
(n = 185)
536 (521-550)

Spironolactone Group
(n = 204) a
(n = 186)
517 (504-531)

11 (6)
172 (88)
13 (7)
71 (36)
137 (135-139)
80 (79-82)
65 (63-66)

8 (4)
178 (87)
18 (9)
61 (30)
128 (126-130)
77 (75-78)
66 (65-68)

165 (82-314)

152 (77-307)

140.4 (140.0-140.8)
4.14 (4.08-4.20)
13.8 (13.6-14.0)
74 (71-77)
(n = 187)
21 (18-23)
66 (63-69)
3.3 (3.2-3.4)
5.6 (5.0-6.2)
5.0 (4.5-5.6)
4.7 (4.2-5.3)

139.5 (139.0-139.9)
4.38 (4.32-4.43)
13.6 (13.4-13.7)
69 (66-71)
(n = 194)
21 (19-24)
64 (61-68)
3.3 (3.2-3.4)
5.5 (4.9-6.1)
4.7 (4.2-5.3)
4.4 (3.8-4.9)

Difference (95% CI)

P Value

15 (27 to 2)

.02

1.30 (0.70 to 2.40) c

.41

0.69 (0.42 to 1.14) c

8 (11 to 5)
3 (5 to 2)
1 (1 to 3)

.15
.001
.001
.56

0.86 (0.75 to 0.99) d

.03

0.9 (1.4 to 0.5)

0.2 (0.1 to 0.3)
0.2 (0.4 to 0.1)
5 (8 to 3)

.001
.001
.003
.001

0.0 (2 to 2)
1 (2 to 4)
0.0 (0.1 to 0.1)
0.1 (0.7 to 0.5)
0.4 (0.9 to 0.1)
0.5 (1.0 to 0.0)

.97
.62
.79
.72
.14
.07

Abbreviations: eGFR, estimated glomerular filtration rate calculated by the Modification of Diet in Renal Disease formula: 186(serum creatinine [in micromoles per liter]/88.4)
1.154age (in years)0.2031.21 (if patient is black)0.742 (if patient is female); HADS, Hospital Anxiety and Depression Scale; NT-proBNP, N-terminal probrain-type natriuretic peptide; NYHA, New York Heart Association; PHQ-9, 9-item depression scale of the Patient Health Questionnaire; SF-36, 36-Item Short Form Health Survey.
a Data are expressed as groupwise mean (95% CI) unless otherwise indicated. Higher values indicate better performance for 6-minute walk distance, NYHA class, eGFR, SF-36
Physical Functioning Scale, and SF-36 Global Self-Assessment. Lower values indicate better performance for NT-proBNP, Minnesota Living With Heart Failure Questionnaire
total score, symptoms of depression (PHQ-9 summary score), HADS anxiety score, and HADS depression score.
b Between-group differences are from analysis of covariance, adjusting for baseline.
c Odds ratio (95% CI).
d Geometric mean ratio (95% CI).

Table 5. Cardiopulmonary Exercise Testing Results After 12 Months

Spironolactone Placebo b
Placebo Group, Mean
(95% CI) (n = 187) a

Spironolactone Group,
Mean (95% CI) (n = 187) a

Difference (95% CI)

P Value

Primary spiroergometric end point

Peak V O2, mL/min/kg

16.9 (16.2-17.5)

16.8 (16.2-17.5)

0.1 (0.6 to 0.8)

.81

Secondary end points

ATV O2, mL/min/kg
V E/V CO2 slope

12.1 (11.6-12.6)

11.9 (11.3-12.4)

0.3 (1.0 to 0.4)

.39

31.5 (30.8-32.2)

31.8 (31.2-32.5)

0.8 (0 to 1.5)

.04

4.6 (4.3-4.9)

4.6 (4.3-4.9)

0.1 (0.3 to 0.5)

.40

547 (520-574)

540 (512-567)

10 (8 to 28)

.27

Measurements

Borg scale
Other variables
Duration of exercise, s

Abbreviations: ATV O2, oxygen consumption at anaerobic threshold; V CO2, volume of expired carbon dioxide; V E, expired volume per unit time; V O2, oxygen consumption.
a Higher values indicate better performance for duration of exercise, peak V
O2, and ATV O2. Lower values indicate better performance for V E/V CO2 slope and Borg scale.
b Between-group differences are from analysis of covariance, adjusting for baseline.

788

JAMA, February 27, 2013Vol 309, No. 8

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

SPIRONOLACTONE IN PATIENTS WITH HEART FAILURE

for risk factor control. Earlier pharmacologic interventions in HF with preserved EF have failed to improve diastolic dysfunction, the major underlying
cardiac pathophysiology in HF with
preserved EF. Angiotensin receptor
blockers34 and -blockers35 have not induced either functional or clinical outcome improvements3,4 in randomized
trials. Hence, spironolactone is the first
drug to show an improvement in diastolic function among patients with HF
with preserved EF in a randomized,
double-blind, placebo-controlled clinical trial.36 However, we did not observe a reduction in left atrial size as in
another recent trial of patients with HF
with preserved EF.37 This may be explained by the mild symptoms and only
mildly dilated left atria as well as by the
low prevalence of atrial fibrillation in
our study. Another explanation could
be that functional and structural
changes induced by spironolactone
need more time to affect left atrial size.
In addition, a mild potassium-sparing
diuretic effect may have contributed to
our findings.
Compared with placebo, spironolactone resulted in a substantial blood
pressure reduction, which could explain structural and functional cardiac effects. However, after adjustment for baseline and follow-up blood
pressure values, the effects of spironolactone on diastolic function (E/e',
1.1; 95% CI, 1.6 to 0.5; P.001)
and left ventricular mass index (4.8
g/m 2 ; 95% CI, 9.4 to 0.3 g/m 2 ;
P = .04) remained statistically significant, suggesting that the reverse remodeling effects of spironolactone are
independent of blood pressure reduction. However, blood pressure is an indirect measure of left ventricular afterload and correction for blood pressure
reduction may not be adequate to refute a primary effect of changes in systolic and diastolic load. Although the
mean blood pressure reduction by spironolactone in the Aldo-DHF trial was
5 mm Hg lower than that observed in
the VALIDD study with valsartan, spironolactone treatment lowered E/e',
neuroendocrine activation, and left ven-

Table 6. Adverse Events and Adherence

Events/Adherence

Placebo Group,
No. (%)

Adverse events
Death

Spironolactone
Group, No. (%)

P Value

1 (1)

.99

50 (24)

60 (28)

.38

Cardiac hospitalization

15 (7)

21 (10)

.38

Noncardiac hospitalization

37 (18)

47 (22)

.27

32 (15)

33 (15)

.99

Hospitalization

Worsening dyspnea

New or worsening edema

44 (21)

35 (16)

.26

Worsening coronary heart disease a

29 (14)

33 (15)

.68

Myocardial infarction
Worsening renal function b
eGFR 30 mL/min/1.73 m2 at last visit

3 (1)

5 (2)

.72

43 (21)

77 (36)

.001

3 (1)

.62

New or worsening anemia c

18 (9)

34 (16)

.03

Anemia at last visit

20 (10)

33 (15)

.08

Serum potassium level

Ever increased 5.0 mmol/L

22 (11)

44 (21)

.005

5.0 mmol/L at last visit

7 (3)

13 (6)

.25

Ever increased 5.5 mmol/L

3 (1)

4 (2)

.99

1 (1)

1 (1)

.99

1 (1)

9 (4)

5.5 mmol/L at last visit

Gynecomastia
Adherence to study medication
Overall adherence rate, mean (95% CI), %

1 (1)

.02

93 (91-96)

92 (89-95)

Ever reported intolerance

13 (6)

36 (17)

.001

.44

Ever reduced dose or stopped

30 (14)

48 (23)

.03

Abbreviation: eGFR, estimated glomerular filtration rate calculated by Modification of Diet in Renal Disease formula:
186 (serum creatinine [in micromoles per liter]/88.4) 1.154 age (in years) 0.203 1.21 (if patient is black)
0.742 (if patient is female).
a Worsening coronary heart disease is defined as myocardial infarction, revascularization, or occurrence of angina pectoris at follow-up.
b Worsening renal function is defined as worsening as reported by the physician, decrease of eGFR to below 30 mL/
min/1.73 m2, or decrease of eGFR by more than 15 mL/min/1.73 m2 vs baseline.
c New or worsening anemia is defined as newly diagnosed anemia according to World Health Organization criteria or
worsening of hemoglobin levels in anemic patients by 1 g/dL or more during follow-up.

tricular mass index, whereas valsartan

had no overall effect on these measures. 34 Similarly, enalapril 38 and
nebivolol35 decreased blood pressure
without effects on diastolic function or
structural remodeling in HF with preserved EF.
The beneficial effects of spironolactone on diastolic function were not associated with any clinical improvement. Our study population may have
been too young or too healthy, or the
treatment period may have been too
short, for observing a translation of improved diastolic function into a clinical benefit. The low event rate in the
Aldo-DHF trial may indicate that the
study population likely represented
early-stage HF with preserved EF, and
longer follow-up may have been needed
to fully evaluate the potential effects of
spironolactone on symptomatic or clinical outcome end points. Also, the observed 2.8% reduction in submaximal

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

exercise capacity in the spironolactone group warrants consideration. Although this decline appears clinically
irrelevant, it could be explained by the
modest decrease in renal function, the
yet unexplained decrease in hemoglobin levels, or external factors such as
assessment technique, patient motivation, or statistical chance. The neutral
effect of spironolactone on peak V O2
and the small negative effect on 6-minute walking distance could possibly also
be explained by a reduction in filling
pressures. However, the known antiandrogenic action of spironolactone,
with adverse effects on skeletal muscle
function and strength independent of
myocardial function and left ventricular remodeling, might also have contributed to the lack of symptomatic improvement in our cohort.39 However,
whether a more specific mineralocorticoid receptor blockade using, for instance, eplerenone or canrenone may
JAMA, February 27, 2013Vol 309, No. 8 789

SPIRONOLACTONE IN PATIENTS WITH HEART FAILURE

result in different outcomes is currently unknown.13 Nevertheless, the

lack of meaningful beneficial effects of
spironolactone on various measures of
exercise capacity and quality of life suggests that in addition to left ventricular structural and functional remodeling, other mechanisms also contribute
to impaired functional capacity in HF
with preserved EF.40
Spironolactone increased potassium levels by an average of 0.2 mmol/L.
Of note, clinically relevant hyperkalemia (5.5 mmol/L) was rare, potassium levels never exceeded 5.8 mmol/L,
and no hospitalizations for hyperkalemia occurred. Observational data after the publication of the RALES study
in HF with reduced EF have shown that
inappropriately high doses of spironolactone and less rigorous control of risk
factors for hyperkalemia in daily practice may shift the risk-benefit ratio of
spironolactone toward harm.38 Thus,
adherence to recommended exclusion
criteria, dosing guidelines, and regular monitoring of potassium levels and
renal function is recommended.
The present results should be interpreted in the context of several limitations. The Aldo-DHF study population
consisted of stable patients with moderate heart failure symptoms. The results
may not apply to patients with more severe disease and more comorbidities.40-42 However, at later stages, patients with HF with preserved EF more
often die of noncardiovascular causes.
Therefore, an intervention with an effect on cardiac structure and function at
an earlier phase of the disease as tested
in Aldo-DHF seems attractive.42-44 However, the relatively stable study population may have precluded translation of
cardiac functional and structural improvements into better exercise tolerance and quality of life. Importantly,
Aldo-DHF was not powered to evaluate
the effect of spironolactone on heart failure hospitalizations or mortality.
Additionally, in contrast to large outcome trials, NT-proBNP was not selected as a specific inclusion criterion
in the Aldo-DHF trial. The reason for
this was that NT-proBNP is not a sen790

JAMA, February 27, 2013Vol 309, No. 8

sitive diagnostic marker of the disease, is influenced by typical comorbidities such as renal dysfunction and
obesity, and is not elevated beyond diagnostic cutoff values even in many patients included in large multicenter
trials such as I-Preserve.3 The lack of
considerably elevated NT-proBNP levels in Aldo-DHF may indicate a relatively stable HF population, which accounts for the low rate of cardiovascular
events observed in our cohort. Because of our study design, typical comorbidities were underrepresented in
Aldo-DHF.
To date, there is no accepted minimal clinically important difference in
peak V O2 or E/e' that should be reached
in view of altering prognosis in HF with
preserved EF. Future studies, while remaining stringent from a pathophysiologic point of view, need to establish
the effect of changes in our echocardiographic and clinical end points on
morbidity and mortality.
In conclusion, Aldo-DHF showed
that compared with placebo, spironolactone treatment in patients with HF
with preserved EF improved diastolic
function and left ventricular remodeling but did not alter maximal exercise
capacity. The lack of accepted minimal clinically important differences in
E/e' or peak V O2 in HF with preserved
EF warrants additional prospective, randomized, adequately powered studies
to further evaluate the effect of improving diastolic function on symptomatic, functional, and clinical end
points.
Author Affiliations: Department of Cardiology and
Pneumology, Heart Center (Drs Edelmann, Wachter,
Duvinage, Stahrenberg, Durstewitz, and Hasenfuss),
German Center for Cardiovascular Research (Drs Edelmann, Wachter, Herrmann-Lingen, and Hasenfuss),
and Department of Psychosomatic Medicine and Psychotherapy (Dr Herrmann-Lingen), University of
Go ttingen, Go ttingen, Germany; Department of Cardiology, Medical University Graz (Drs Schmidt,
Kraigher-Krainer, Colantonio, and Pieske) and Ludwig- Boltzmann Institute for Translational Heart Failure Research (Drs Kraigher-Krainer and Pieske), Graz,
Austria; Department of Internal Medicine, MediClin
Rehabilitation Center Spreewald, Spreewald, Germany (Dr Kamke); Institute of Medical Informatics, Statistics, and Epidemiology (Dr Lo ffler) and Clinical Trial
Center, University of Leipzig (Dr Gelbrich), Leipzig, Germany; Department of Internal MedicineCardiology,
Charite Campus Virchow-Klinikum (Dr Du ngen), and
Department of Cardiology and Pneumology, Charite
Campus Benjamin Franklin, Universita tsmedizin (Dr

Tscho pe), Berlin, Berlin, Germany; Department of Preventive and Rehabilitative Sports Medicine, Technical University Munich, Munich, Germany (Dr Halle);
and Institute of Clinical Epidemiology and Biometry,
University of Wu rzburg, Wu rzburg, Germany (Dr
Gelbrich).
Author Contributions: Dr Gelbrich had full access to
all of the data in the study and takes responsibility for
the integrity of the data and the accuracy of the data
analysis. Drs Edelmann and Wachter contributed
equally to the article and share first authorship.
Study concept and design: Edelmann, Wachter,
Schmidt, Lo ffler, Hermann-Lingen, Halle, Hasenfuss,
Gelbrich, Pieske.
Acquisition of data: Edelmann, Wachter, Schmidt,
Kraigher-Krainer, Colantonio, Kamke, Duvinage,
Stahrenberg, Durstewitz, Du ngen, Tscho pe, Halle,
Pieske.
Analysis and interpretation of data: Edelmann,
Wachter, Schmidt, Kraigher-Krainer, Colantonio,
Durstewitz, Dungen, Halle, Hasenfuss, Gelbrich, Pieske.
Drafting of the manuscript: Edelmann, Wachter, Loffler,
Pieske.
Critical revision of the manuscript for important intellectual content: Edelmann, Wachter, Schmidt,
Kraigher-Krainer, Colantonio, Kamke, Duvinage,
Stahrenberg, Durstewitz, Lo ffler, Du ngen, Tscho pe,
Hermann-Lingen, Halle, Gelbrich.
Statistical analysis: Edelmann, Du ngen, Gelbrich.
Obtained funding: Edelmann, Lo ffler, HermannLingen, Halle, Hasenfuss, Gelbrich, Pieske.
Administrative, technical, or material support:
Edelmann, Wachter, Kamke, Duvinage, Durstewitz,
Halle.
Study supervision: Edelmann, Wachter, Schmidt,
Lo ffler, Tscho pe, Hermann-Lingen, Halle, Hasenfuss,
Pieske.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure
of Potential Conflicts of Interest. Dr Edelmann reports having been an investigator, consultant, or
speaker for Berlin Chemie, Novartis, Pfizer, Servier,
Bayer, Gilead, CVRx, Relypsa, Sanofi, and AstraZeneca. Dr Wachter reports having been, since 2003,
an investigator, consultant, or speaker for Bayer, Berlin Chemie, Boehringer Ingelheim, Boston Scientific,
CVRx, Gilead, Johnson & Johnson, Medtronic, Novartis, Pfizer, Relypsa, Sanofi, and Servier. Dr HermannLingen reports receiving speakers honoraria from
Pfizer, Servier, and Berlin-Chemie. He has received royalties from Verlag Hans Huber and Deutscher A rzteverlag and participates in institutional research cooperation with KKH-Allianz. Dr Halle reports receiving
speakers honoraria from Berlin-Chemie, Merck Sharpe
& Dohme, Bristol-Myers Squibb, and Sanofi-Aventis
and receiving honoraria as board member (SanofiAventis), for expert testimony (Health Insurance Company, TK Germany), and as consultant (BMW). Dr
Hasenfuss reports receiving honorarium for presentations from CVRx, Impulse Dynamics, and Servier and
serving as a consultant for Novartis and Servier. Dr Gelbrich reports receiving remuneration from Robert Bosch
Health Care for board membership. Dr Pieske reports
receiving speaker honoraria from Bayer Healthcare,
Boehringer Ingelheim, Servier, Medtronic, BristolMyers-Squibb, and Menarini and serving as a consultant and/or steering committee member for Bayer
Healthcare, Menarini, and Novartis. He participated
in research cooperations (institutional) with Bayer
Healthcare and Medtronic. No other conflicts of interest were reported.
Funding/Support: This work was supported by the German-Austrian Heart Failure Study Group and the German Competence Network of Heart Failure. AldoDHF was funded by the Federal Ministry of Education
and Research Grant 01GI0205 (clinical trial program
Aldo-DHF [FKZ 01KG0506]). The University of Go ttingen was the formal sponsor.

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

SPIRONOLACTONE IN PATIENTS WITH HEART FAILURE

Role of the Sponsor: The sponsor and supporters of
this study had no role in the design and conduct of
the study; in the collection, management, analysis, and
interpretation of the data; or in the preparation, review, or approval of the manuscript.
Online-Only Material: The eAppendix, eTables 1
through 4, and eFigures 1 through 5 are available at
http://www.jama.com.

REFERENCES
1. Owan TE, Hodge DO, Herges RM, Jacobsen SJ,
Roger VL, Redfield MM. Trends in prevalence and outcome of heart failure with preserved ejection fraction.
N Engl J Med. 2006;355(3):251-259.
2. Campbell RT, Jhund PS, Castagno D, Hawkins NM,
Petrie MC, McMurray JJ. What have we learned about
patients with heart failure and preserved ejection
fraction from DIG-PEF, CHARM-Preserved, and
I-PRESERVE? J Am Coll Cardiol. 2012;60(23):
2349-2356.
3. Massie BM, Carson PE, McMurray JJ, et al; IPRESERVE Investigators. Irbesartan in patients with
heart failure and preserved ejection fraction. N Engl J
Med. 2008;359(23):2456-2467.
4. Cleland JG, Tendera M, Adamus J, Freemantle N,
Polonski L, Taylor J; PEP-CHF Investigators. The Perindopril in Elderly People With Chronic Heart Failure (PEPCHF) study. Eur Heart J. 2006;27(19):2338-2345.
5. Ahmed A, Rich MW, Love TE, et al. Digoxin and
reduction in mortality and hospitalization in heart failure: a comprehensive post hoc analysis of the DIG trial.
Eur Heart J. 2006;27(2):178-186.
6. Yusuf S, Pfeffer MA, Swedberg K, et al; CHARM
Investigators and Committees. Effects of candesartan in patients with chronic heart failure and preserved left-ventricular ejection fraction: the CHARMPreserved Trial. Lancet. 2003;362(9386):777-781.
7. Zile MR, Baicu CF, Gaasch WH. Diastolic heart
failureabnormalities in active relaxation and passive stiffness of the left ventricle. N Engl J Med. 2004;
350(19):1953-1959.
8. Borlaug BA, Paulus WJ. Heart failure with preserved ejection fraction: pathophysiology, diagnosis,
and treatment. Eur Heart J. 2011;32(6):670-679.
9. Edelmann F, Tomaschitz A, Wachter R, et al. Serum aldosterone and its relationship to left ventricular structure and geometry in patients with preserved
left ventricular ejection fraction. Eur Heart J. 2012;
33(2):203-212.
10. Zannad F, McMurray JJ, Krum H, et al; EMPHASISHF Study Group. Eplerenone in patients with systolic
heart failure and mild symptoms. N Engl J Med. 2011;
364(1):11-21.
11. Pitt B, Remme W, Zannad F, et al; Eplerenone PostAcute Myocardial Infarction Heart Failure Efficacy and
Survival Study Investigators. Eplerenone, a selective
aldosterone blocker, in patients with left ventricular
dysfunction after myocardial infarction. N Engl J Med.
2003;348(14):1309-1321.
12. Pitt B, Zannad F, Remme WJ, et al; Randomized
Aldactone Evaluation Study Investigators. The effect
of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med. 1999;
341(10):709-717.
13. Deswal A, Richardson P, Bozkurt B, Mann DL. Results of the Randomized Aldosterone Antagonism in
Heart Failure With Preserved Ejection Fraction trial
(RAAM-PEF). J Card Fail. 2011;17(8):634-642.
14. Mottram PM, Haluska B, Leano R, Cowley D,
Stowasser M, Marwick TH. Effect of aldosterone antagonism on myocardial dysfunction in hypertensive
patients with diastolic heart failure. Circulation. 2004;
110(5):558-565.
15. Edelmann F, Schmidt AG, Gelbrich G, et al. Ra-

tionale and design of the Aldosterone Receptor Blockade in Diastolic Heart Failure trial: a double-blind, randomized, placebo-controlled, parallel group study to
determine the effects of spironolactone on exercise
capacity and diastolic function in patients with symptomatic diastolic heart failure (Aldo-DHF). Eur J Heart
Fail. 2010;12(8):874-882.
16. Pocock SJ, Simon R. Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics. 1975;31(1):103115.
17. Paulus WJ, Tscho pe C, Sanderson JE, et al. How
to diagnose diastolic heart failure: a consensus statement on the diagnosis of heart failure with normal left
ventricular ejection fraction by the Heart Failure and
Echocardiography Associations of the European Society of Cardiology. Eur Heart J. 2007;28(20):25392550.
18. Ware JE Jr, Sherbourne CD. The MOS 36-item
short-form health survey (SF-36), I: conceptual framework and item selection. Med Care. 1992;30(6):
473-483.
19. Rector TS, Kubo SH, Cohn JN. Validity of the Minnesota Living With Heart Failure questionnaire as a
measure of therapeutic response to enalapril or placebo.
Am J Cardiol. 1993;71(12):1106-1107.
20. Spitzer RL, Kroenke K, Williams JB. Validation and
utility of a self-report version of PRIME-MD: the PHQ
primary care study. JAMA. 1999;282(18):17371744.
21. Zigmond AS, Snaith RP. The hospital anxiety and
depression scale. Acta Psychiatr Scand. 1983;67
(6):361-370.
22. Cook CE. The Minimal Clinically Important Change
Score (MCID): a necessary pretense. J Man Manip Ther.
2008;16(4):E82-E83.
23. Wyrwich KW, Tierney WM, Wolinsky FD. Further evidence supporting an SEM-based criterion for
identifying meaningful intra-individual changes in
health-related quality of life. J Clin Epidemiol. 1999;
52(9):861-873.
24. Ommen SR, Nishimura RA, Appleton CP, et al.
Clinical utility of Doppler echocardiography and tissue Doppler imaging in the estimation of left ventricular filling pressures: a comparative simultaneous
Doppler-catheterization study. Circulation. 2000;
102(15):1788-1794.
25. Tsutamoto T, Wada A, Maeda K, et al. Effect of
spironolactone on plasma brain natriuretic peptide and
left ventricular remodeling in patients with congestive heart failure. J Am Coll Cardiol. 2001;37(5):
1228-1233.
26. Zannad F, Alla F, Dousset B, Perez A, Pitt B; Rales
Investigators. Limitation of excessive extracellular matrix turnover may contribute to survival benefit of spironolactone therapy in patients with congestive heart
failure: insights from the Randomized Aldactone Evaluation Study (RALES). Circulation. 2000;102(22):
2700-2706.
27. Chan AK, Sanderson JE, Wang T, et al. Aldosterone receptor antagonism induces reverse remodeling when added to angiotensin receptor blockade in
chronic heart failure. J Am Coll Cardiol. 2007;
50(7):591-596.
28. Vizzardi E, DAloia A, Giubbini R, et al. Effect of
spironolactone on left ventricular ejection fraction and
volumes in patients with class I or II heart failure. Am
J Cardiol. 2010;106(9):1292-1296.
29. Phelan D, Thavendiranathan P, Collier P, Marwick
TH. Aldosterone antagonists improve ejection fraction and functional capacity independently of functional class: a meta-analysis of randomised controlled trials. Heart. 2012;98(23):1693-1700.
30. Desai AS, Lewis EF, Li R, et al. Rationale and design of the treatment of preserved cardiac function
heart failure with an aldosterone antagonist trial: a ran-

2013 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a Chulalongkorn University User on 08/03/2015

domized, controlled study of spironolactone in patients with symptomatic heart failure and preserved
ejection fraction. Am Heart J. 2011;162(6):966972, e10.
31. Bomback AS, Klemmer PJ. The incidence and implications of aldosterone breakthrough. Nat Clin Pract
Nephrol. 2007;3(9):486-492.
32. Bauersachs J, Heck M, Fraccarollo D, et al. Addition of spironolactone to angiotensin-converting enzyme inhibition in heart failure improves endothelial
vasomotor dysfunction: role of vascular superoxide anion formation and endothelial nitric oxide synthase
expression. J Am Coll Cardiol. 2002;39(2):351358.
33. Lacolley P, Safar ME, Lucet B, Ledudal K, Labat
C, Benetos A. Prevention of aortic and cardiac fibrosis by spironolactone in old normotensive rats. J Am
Coll Cardiol. 2001;37(2):662-667.
34. Solomon SD, Janardhanan R, Verma A, et al; Valsartan in Diastolic Dysfunction (VALIDD) Investigators.
Effect of angiotensin receptor blockade and antihypertensive drugs on diastolic function in patients with
hypertension and diastolic dysfunction: a randomised trial. Lancet. 2007;369(9579):2079-2087.
35. Conraads VM, Metra M, Kamp O, et al. Effects
of the long-term administration of nebivolol on the
clinical symptoms, exercise capacity, and left ventricular function of patients with diastolic dysfunction: results of the ELANDD study. Eur J Heart Fail. 2012;
14(2):219-225.
36. Holland DJ, Kumbhani DJ, Ahmed SH, Marwick
TH. Effects of treatment on exercise tolerance, cardiac function, and mortality in heart failure with preserved ejection fraction: a meta-analysis. J Am Coll
Cardiol. 2011;57(16):1676-1686.
37. Solomon SD, Zile M, Pieske B, et al; Prospective
comparison of ARNI with ARB on Management of
Heart Failure With Preserved Ejection Fraction (PARAMOUNT) Investigators. The angiotensin receptor neprilysin inhibitor LCZ696 in heart failure with preserved ejection fraction: a phase 2 double-blind
randomised controlled trial. Lancet. 2012;380(9851):
1387-1395.
38. Kitzman DW, Hundley WG, Brubaker PH, et al.
A randomized double-blind trial of enalapril in older
patients with heart failure and preserved ejection fraction: effects on exercise tolerance and arterial
distensibility. Circ Heart Fail. 2010;3(4):477-485.
39. Iellamo F, Rosano G, Volterrani M. Testosterone
deficiency and exercise intolerance in heart failure:
treatment implications. Curr Heart Fail Rep. 2010;
7(2):59-65.
40. Edelmann F, Stahrenberg R, Gelbrich G, et al. Contribution of comorbidities to functional impairment is
higher in heart failure with preserved than with reduced ejection fraction. Clin Res Cardiol. 2011;
100(9):755-764.
41. Juurlink DN, Mamdani MM, Lee DS, et al. Rates
of hyperkalemia after publication of the Randomized
Aldactone Evaluation Study. N Engl J Med. 2004;
351(6):543-551.
42. Ather S, Chan W, Bozkurt B, et al. Impact of noncardiac comorbidities on morbidity and mortality in a
predominantly male population with heart failure and
preserved vs reduced ejection fraction. J Am Coll
Cardiol. 2012;59(11):998-1005.
43. Anand IS, Rector TS, Cleland JG, et al. Prognostic value of baseline plasma amino-terminal probrain natriuretic peptide and its interactions with irbesartan treatment effects in patients with heart failure
and preserved ejection fraction: findings from the IPRESERVE trial. Circ Heart Fail. 2011;4(5):569577.
44. Henkel DM, Redfield MM, Weston SA, Gerber
Y, Roger VL. Death in heart failure: a community
perspective. Circ Heart Fail. 2008;1(2):91-97.

JAMA, February 27, 2013Vol 309, No. 8 791