J Appl Physiol 119: 726 733, 2015.

Review First published March 6, 2015; doi:10.1152/japplphysiol.00904.2014.

HIGHLIGHTED TOPIC Exploring New Concepts in the Management of Heart Failure

with Preserved Ejection Fraction: Is Exercise the Key for Improving Treatment?

Clinical outcomes and cardiovascular responses to exercise training in heart

failure patients with preserved ejection fraction: a systematic review and
meta-analysis
Gudrun Dieberg, Hashbullah Ismail, Francesco Giallauria, and X Neil A. Smart
School of Science and Technology, University of New England, Armidale, New South Wales, Australia

Dieberg G, Ismail H, Giallauria F, Smart NA. Clinical outcomes and cardiovascular

responses to exercise training in heart failure patients with preserved ejection fraction: a
systematic review and meta-analysis. J Appl Physiol 119: 726733, 2015. First published

Downloaded from http://jap.physiology.org/ by 10.220.32.246 on October 2, 2017

March 6, 2015; doi:10.1152/japplphysiol.00904.2014.Exercise training induces phys-
ical adaptations for heart failure patients with systolic dysfunction, but less is
known about those patients with preserved ejection fraction. To establish whether
exercise training produces changes in peak VO2 and related measures, quality of
life, general health, and diastolic function in heart failure patients with preserved
ejection fraction. We conducted a MEDLINE search (1985 to October 10, 2014),
for exercise-based rehabilitation trials in heart failure, using search terms exercise
training, heart failure with preserved ejection fraction, heart failure with normal
ejection fraction, peak VO2, and diastolic heart dysfunction. Seven intervention
studies were included providing a total of 144 exercising subjects and 114 control
subjects, a total of 258 participants. Peak VO2 increased by a mean difference (MD)
2.13 mlkg1min1 [95% confidence interval (CI) 1.54 to 2.71, P 0.00001] in
exercise training vs. sedentary control, equating to a 17% improvement from
baseline. The corresponding data are provided for the following exercise test
variables: VE/VCO2 slope, MD 0.85 mlkg1min1 (95% CI 0.05 to 1.65, P
0.04); maximum heart rate, MD 5.60 beats per minute (95% CI 3.95 to 7.25, P
0.00001); Six-Minute Walk Test, MD 32.1 m (95% CI 17.2 to 47.1, P 0.0001);
and indices of diastolic function: E/A ratio, MD 0.07 (95% CI 0.02 to 0.12, P
0.005); E/E= ratio MD 2.31 (95% CI 3.44 to 1.19, P 0.0001); deceleration
time (DT), MD 13.2 ms (95% CI 19.8 to 6.5, P 0.0001); and quality of life:
Minnesota Living with Heart Failure Questionnaire, MD 6.50 (95% CI 9.47 to
3.53, P 0.0001); and short form-36 health survey (physical dimension), MD
15.6 (95% CI 7.4 to 23.8, P 0.0002). In 3,744 h patient-hours of training, not one
death was directly attributable to exercise. Exercise training appears to effect
several health-related improvements in people with heart failure and preserved
ejection fraction.
exercise training; cardio-respiratory fitness; heart failure with preserved ejection
fraction

HEART FAILURE WITH PRESERVED ejection fraction (HFpEF) is
defined as an inability of the ventricles to optimally accept
blood from the atria with blunted end-diastolic volume re-
sponse by limiting the stroke volume and cardiac output.
Exercise intolerance and reduced quality of life are known as
the primary chronic symptoms in HFpEF patients. HFpEF
prevalence is higher in elderly and women and may be linked
to hypertension, diabetes mellitus, and atrial fibrillation (2, 18).
Chronic heart failure (CHF) patients with either normal or
Address for reprint requests and other correspondence: N. A. Smart, School
of Science and Technology, Univ. of New England, Armidale, NSW 2351,
Australia (e-mail: nsmart2@une.edu.au).
726 8750-7587/15 Copyright 2015 the American Physiological Society
abnormal systolic function have similar mortality rates (4). In
the United States, CHF affects 5 million individuals with
heart failure, and more than 555,000 are newly diagnosed with
CHF annually (22). Corresponding costs to the health care
systems are enormous. A review by Smart (25) estimated the
hospital-based exercise therapy treatment costs to prevent mor-
tality in one CHF patient to be approximately (U.S.) $60,000/
yr. Meta-analyses have shown exercise training to be beneficial
in HFpEF patients in terms of improved cardiorespiratory
fitness (12, 30).
In people with HFpEF, as well as systolic heart failure,
cardiorespiratory fitness (peak VO2) is impaired. Impaired peak
VO2 has been associated with increased mortality risk (10) and
http://www.jappl.org
 Review
Exercise Training in Heart Failure with Preserved Ejection Fraction Dieberg G et al. 727
decreased quality of life in heart failure patients. As measured
Records identified through
by traditional indices such as ejection fraction, systolic func-

Identification
database searching Additional records identified
tion appears largely normal under resting baseline conditions in (n = 32) through other sources
(n = 6)
HFpEF. However, studies have shown through global assess-
ment of systolic function by other techniques, such as strain
rate imaging, systolic abnormalities do exist in HFpEF patients Records after duplicates removed
(17). Despite the preservation of systolic function at rest, (n = 19)

mortality rates in HFpEF are similar to those observed in

Screening
systolic failure (5), highlighting the clear need for effective
treatment strategies for these patients. Records screened Records excluded
Interestingly, conventional methods for treating heart failure (n = 19) (n = 11)

have proven largely ineffective for HFpEF patients. Currently,

effective therapeutic approaches for HFpEF are limited and
focus primarily on managing cardiovascular risk factors, espe- Full-text articles

Eligibility
Full-text articles
excluded, with
cially hypertension. Exercise therapy is a promising effective assessed for eligibility
(n = 8) reasons (n=1)
adjunct therapy that can delay disease progression, minimize
pharmaceutical use, and improve functional limitations and
quality of life. While, to date, 100 randomized, controlled

Downloaded from http://jap.physiology.org/ by 10.220.32.246 on October 2, 2017

trials of exercise training have been published in systolic heart
failure patients, strikingly, only 10 such trials exist examining
Studies included in

HFpEF.
In this meta-analysis, our purpose was to assess the effects
of exercise on a number of outcome measures that are com-
monly used to assess clinical status in HFpEF. First, we
evaluated the impact of exercise training on changes to exer-
cise capacity in HFPEF patients compared with sedentary
controls through examination of peak VO2, VE/VCO2, heart rate,
and the Six-Minute Walk Test (6MWT). Second, we studied
clinical measures of diastology, including early-to-late filling
ratio (E/A), E/E= (a widely accepted noninvasive surrogate of
left ventricular filling pressure) and deceleration of early ven-
tricular filling. Third, we examined whether exercise training
produces better quality of life and/or general health through use
of the Minnesota Living with Heart Failure Questionnaire
(MLHFQ) (21) and the SF-36 (which has established norms)
(3). Finally, we examined whether rates of serious events,
mortality, and hospitalization were more frequent with exercise
training in HFPEF patients.
METHODS
Search Strategy
Included
quantitative synthesis
(meta-analysis)
(n = 7)
Fig. 1. Consort statement.
Study Selection
Included were randomized controlled designs of exercise training
in heart failure patients with normal or preserved ejection fraction.
Studies of heart failure patients with abnormal systolic function were
excluded. All included studies are comparisons between exercise
training and sedentary control. Records identified 32 papers through
database searching. Six additional records from reference lists were
added. Only principal studies, with the most subjects, were included
studies for which multiple publications existed from the same dataset.
After initial screening, 19 studies were removed, which included
overlapping studies, duplicates, duplicate data, abstracts, irrelevant
articles (e.g., editorials and discussion papers). We further excluded
12 studies in which the control group received additional intervention,
nonrelevant studies, and acute exercise responses. Seven studies of
exercise training intervention were included (see Consort Statement,
Fig. 1).

Studies were identified through a MEDLINE search (1985 to Outcome Measures

October 10, 2014), Cochrane Controlled Trials Registry (1966 to
October 10, 2014), CINAHL, SPORTDiscus, and the Science Citation
Index. The search strategy included a mix of MeSH and free text
terms for key concepts related to exercise training, heart failure with
normal ejection fraction, and heart failure with preserved ejection
fraction, peak VO2, and diastolic heart failure for clinical trials of
exercise training in heart failure patients. Studies were included when
patients exhibited baseline left ventricular ejection fraction above
45%. These searches were limited to prospective, randomized, or
controlled trials and human studies, with no language restrictions on
publications. Manuscript reference lists and latest journal editions
were scrutinized for new references. Full journal articles were as-
sessed by three reviewers (N. A. Smart, G. Dieberg, and H. Ismail) for
relevance and eligibility. Methodological disagreements were re-
solved by reviewers through discussion. For two studies (9, 16),
authors were contacted and requested to provide further data about the
protocol for implementing desired exercise intensity.
We recorded the following; peak VO2 (baseline and postexercise),
VE/VCO2, maximum heart rate, 6MWT, diastolic function [E/A and
E/E= ratios, deceleration time (DT)], MLHFQ, short form-36 health
survey, participant completion rates, adverse medical events, hospi-
talization, and mortality.
Data Synthesis
We calculated patient-hours of exercise training and percentage
change in peak VO2.
Assessment of Study Quality
We assessed study quality using the 15-point TESTEX scale (29),
which is a validated study quality assessment tool specific to exercise
training studies. Median TESTEX score was 10, with two studies
scoring 8, three scoring 10, and two studies scoring 12.

J Appl Physiol doi:10.1152/japplphysiol.00904.2014 www.jappl.org

Review
728 Exercise Training in Heart Failure with Preserved Ejection Fraction Dieberg G et al.

Table 1. Patient and training characteristics for randomized control trials included in the meta-analysis on exercise training
studies with HFpEF patients
Participants Included in the Final
Study Country Sessions Attended, % Analysis Training Characteristics Outcome Measures

Alves et al. (1) Portugal 100 Total patients N 98
Exercise (55% LVEF): n 20, 22
m/9 f, mean age: 62.9. Control:
n 11
Exercise (45%54% LVEF): n
23, 24 m/9 f, mean age: 63.6.
Control: n 10
Exercise (45% LVEF): n 22, 27
m/7 f, mean age: 62.0. Control:
n 12
NYHA class I/II/III/IV
Edelmann et al. (6) Germany 34 participated in 90%, Total patients: N 64
52 in 7090% and 14 in Exercise: n 44, 24 m/20 f, mean
6 mo of interval exercise training. LVEF
First month, 3 sessions per Diastolic function
week, and 15 min at 7075%
of maximal heart rate.
Following 5 mo, 3 sessions per
week, and 35 min at 7075%
of maximal heart rate
32 sessions of continuous LVEF
exercise training. Weeks 14, 2 Peak VO2

Downloaded from http://jap.physiology.org/ by 10.220.32.246 on October 2, 2017

70% of the exercise age: 64 sessions per week, 2040 min Heart rate
sessions. Control: n 20, 12 m/8 f, mean at 5060% of peak VO2. Week 6MWT
age: 65 5 onward, 3 sessions per week MLHF
NYHA class II/III at 70% of peak VO2 and SF-36
resistance training, 15 reps at
6065% 1 RM
Gary et al. (9) USA 100 Total patients N 28 12 wk of continuous exercise 6MWT
Exercise: n 15, 15 f, mean training (walking). 3 sessions MLHF
age: 67 per week, 2040 min at 40%
Control: n 13, 13f, mean age: 69 60% of the maximal heart
NYHA class II/III rate
Karavidas et al. (15) Greece 100 Total patients N 30 6 wk of functional electrical MLHF
Exercise: n 15, 6 m/9 f, mean stimulation (FES) training. 5 KCCQ
age: 69.4 sessions per week, 30 min at BDI
Control: n 15, 6 m/9 f, mean age: 25 Hz for 5 s followed by 5-s 6MWT
68.5 rest Diastolic function
NYHA class II/III BNP
Kitzmann et al. (16) USA 86, final testing; Total patients N 63 4 mo (16 wk of continuous LVEF
88, exercise training Exercise: n 24, 23 m/9 f, mean exercise training. 3 sessions per Peak VO2
age: 70 week, 60 min at 4070% HRR VE/VCO2
Control: n 30, 25 m/6 f, mean Heart rate
age: 70 6MWT
NYHA class II/III Diastolic function
EDV and ESV
SBP and DBP
MLHF
SF-36
Palau et al. (20) Australia 100 Total patients N 26 12 wk of interval exercise LVEF
Exercise: n 14, 7 m/7 f, mean training. 2 sessions per week, Peak VO2
age: 68 20 min. Subjects started VE/VCO2
Control: n 12, 6 m/6 f, mean breathing at a resistance equal Heart rate
age: 74 to 2530% MIP for 1 wk, and 6MWT
NYHA class II/III/IV each subsequent session was Diastolic function
adjusted to 2530% MIP. NT-proBNP
MLHF
Smart et al. (28) Australia 87.6 Total patients N 25 16 wk of interval exercise LVEF
Exercise: n 12, 7 m/5 f, mean training. 3 sessions per week, Peak VO2
age: 67 30 min at 6070% peak VO2 VE/VCO2
Control: n 13, 6 m/7 f, mean Heart rate
age: 61 Diastolic function
NYHA class I/II MLHF
Peak VO2, maximal oxygen consumption; FES, functional electrical stimulation; peak HR, maximal heart rate; MHR, maximum heart rate; HRR, heart
rate reserve; m/f, male/female; MIP, inspiratory muscle trainer; 1 RM, 1 repetitive maximum; 6MWT, Six-Minute Walk Test; LVEF, left ventricular
ejection fraction; EDV, end diastolic volume; ESV, end systolic volume; MLHF, Minnesota Living with Heart Failure Questionnaire; SF-36, short form-36
health survey; VE/VCO2, ventilatory equivalent for carbon dioxide; KCCQ, Kansas City Cardiomyopathy Questionnaire; BDI, Beck depression inventory;
SBP, systolic blood pressure; DBP, diastolic blood pressure; BNP, brain natriuretic peptide; and NT-proBNP, N-terminal prohormone of brain natriuretic
peptide.

J Appl Physiol doi:10.1152/japplphysiol.00904.2014 www.jappl.org


Exercise Training in Heart Failure with Preserved Ejection Fraction
Statistical Analyses
Revman 5.1 software (The Nordic Cochrane Centre, Copenhagen,
Denmark) was used for data analysis. Continuous data were reported
as means and SD. Revman 5.1 enabled calculation of postintervention
change from baseline for SD, using change in mean values, number of
subjects, and P value or preferably 95% CIs. In many cases in which
exact P values were not provided, we used default values e.g., P
0.05 became P 0.049. Mean difference (MD) in these data from
baseline was analyzed. We used a 5% level of significance and a 95%
CI to report change in outcome measures. Egger plots were produced
to identify sources of publication bias (7).
RESULTS
Included Studies
Seven studies met selection criteria (1, 6, 9, 15, 16, 20, 28),
providing a total of 144 exercising subjects and 114 control
subjects, a total of 258 participants (Table 1). Total patient-
hours of exercise training reported were 3,744 h.
J Appl Physiol doi:10.1152/japplphysiol.00904.2014 www.jappl.org
Review
Dieberg G et al. 729
Fig. 2. Cardiorespiratory variables. A:
change in peak VO2 for exercise training
studies with heart failure patients with pre-
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on October 2, 2017
served ejection fraction (HFpEF) patients. B:
change in VE/VCO2 for exercise training
studies with HFpEF patients. C: change in
maximum heart rate for exercise training
studies with HFpEF patients. D: change in
Six-Minute Walk Test (6MWT) for exercise
training studies with HFpEF patients.
Exercise Training Parameters
Program length for high-intensity training varied from 6
to 26 wk and frequency from two to five sessions weekly.
Three studies used walking and cycle training, one used
only walking, one used cycling only, one used functional
electrical stimulation, and one inspiratory muscle training.
Outcome Measures
Change in peak VO2. Data from four studies showed MD for
peak VO2 was 2.13 mlkg1min1 [95% confidence interval (CI)
1.54 to 2.71, P 0.00001] in exercise vs. control (Fig. 2A).
Change in peak VE/VCO2 slope. Data from three studies
for VE/VCO2 slope showed MD was 0.85 units higher (95%
CI 0.05 to 1.65, P 0.04) in exercise vs. control (see
Fig. 2B).
Change in heart rate. Data from four studies showed
MD for maximum heart rate was 5.60 bpm (95% CI 3.95
Review
730 Exercise Training in Heart Failure with Preserved Ejection Fraction
Fig. 3. Diastolic function. A: change in E/A
ratio for exercise training studies with HFpEF
patients. B: change in E/E= ratio for exercise
training studies with HFpEF patients. C:
change in deceleration time for exercise train-
ing studies with HFpEF patients.
to 7.25, P 0.00001) in exercise vs. control (see
Fig. 2C).
Change in 6MWT. Data from five studies for 6MWT re-
sulted in a MD of 32.1 m (95% CI 17.20 to 47.05, P 0.0001)
with exercise vs. control (see Fig. 2D).
Change in Diastolic Function
Change in E/A ratio. Data from three studies showed the
MD for diastolic function (E/A ratio) was 0.07 (95% CI 0.02 to
0.12, P 0.005) with exercise vs. control (Fig. 3A).
Change in E/E= ratio. Data from four studies showed the MD
for diastolic function (E/E= ratio) was 2.31 (95% CI 3.44 to
1.19, P 0.0001) with exercise vs. control (Fig. 3B).
Change in deceleration time. Data from three studies for
deceleration time (DT) showed a MD of 13.2 ms (95% CI
19.8 to 6.5, P 0.0001) with exercise vs. control (Fig. 3C).
Change in Quality of Life
Data from six studies showed MD for MLHFQ was 6.50
(95% CI 9.47 to 3.53, P 0.0001) with exercise vs.
control (see Fig. 4A).
Change in the Short Form-36 Health Survey
Data from two studies showed MD for the physical dimension
of the short form-36 health survey was 15.6 (95% CI 7.35 to 23.8,
P 0.0002) with exercise vs. control (see Fig. 4B).
Adverse Events
All studies reported adverse events (deaths, hospitalizations,
and cardiovascular events), see Table 2. There were no deaths
reported from exercise training or control groups in any of the
included studies. One hospital admission was reported from an
J Appl Physiol doi:10.1152/japplphysiol.00904.2014 www.jappl.org
Dieberg G et al.
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on October 2, 2017
exercise training patient. Overall, there were 3,744 patient-
hours of exercise training reported. There were insufficient
adverse event data to justify analyses.
Study Quality
Median TESTEX score was 10 (maximum 15) for all studies
(see Table 3). Funnel (Egger) plots of the analysis showed
minimal evidence of publication bias.
Heterogeneity
Only the analyses of VE/VCO2 slope and 6MWT showed
high heterogeneity.
DISCUSSION
This meta-analysis indicates that, in HFpEF patients, the
magnitude of gain in cardiorespiratory fitness is similar to that
seen in systolic heart failure patients exercising at moderate
intensity (13, 14). In HFpEF, exercise training elicits improve-
ments in cardiac (diastolic) function, health-related quality of
life, general health, VE/VCO2 slope, maximum heart rate, and
6-min walk distance, which complement improvements seen in
cardiorespiratory fitness.
Change in peakVO2,VE/VCO2 Slope, Maximum Heart Rate,
and 6-Minute Walk Test
The improvements in peak VO2 observed with exercise
training are complemented with changes in maximum heart
rate, but not in VE/VCO2 slope. This current analysis produced
peak VO2 changes of similar effect size seen previously (30).
Although one trial has established a 4.4 unit improvement in
VE/VCO2 slope (28), our analysis does not identify an improve-
ment with exercise training. A previous meta-analysis did not
 Review
Exercise Training in Heart Failure with Preserved Ejection Fraction Dieberg G et al. 731

Fig. 4. Quality of life. A: change in Minne-

sota Living with Heart Failure Questionnaire
(MLHFQ) for exercise training studies with
HFpEF patients. B: change in the physical
dimension of the short form-36 health survey
for exercise-training studies with HFpEF pa-
tients.

Downloaded from http://jap.physiology.org/ by 10.220.32.246 on October 2, 2017

analyze change in VE/VCO2 slope in HFpEF patients (30). A
previous report established a strong prognostic relationship
between peak VO2, VE/VCO2 slope, and mortality in heart
failure patients with reduced systolic function (23). Our anal-
ysis showed maximum heart rate to be higher after exercise
training; the resultant increase in maximal cardiac output
would at least partially explain why peak VO2 improved with
training. Although VE/VCO2 slope did not decrease in our
analysis, a reduction would imply improved ventilatory effi-
ciency and would likely contribute to improved cardiorespira-
tory fitness. A mean 10-12 bpm increase in maximum heart
rate after training has been reported in systolic heart failure and
HFpEF in previous trials (26, 28).
Diastolic Function
Our analysis of E/A ratio is the first to identify that exercise
training may significantly improve this aspect of diastolic
function. Neither individual studies, nor pooled analyses pub-
lished previously, have shown a postexercise training benefit.
Our analysis of E/E= confirms the finding of Edelmann et al.
(6), but not the findings of the pooled analysis of Taylor et al.
(30). E/E= is a surrogate for filling pressure, and our analysis
suggests a small reduction (improvement) is elicited with
exercise training in HFpEF patients. We suspect that these
changes in E/E= do not account for all of the improvement in
peak VO2. Therefore, we acknowledge that some changes may
be due to improved endothelial function (11). Deceleration
time (DT) was also significantly reduced in our analysis; this is
the first analysis to demonstrate such a benefit. Together, these
three measures of diastolic function have shown a trend toward
normalization after exercise training. While improved diastolic
function due to exercise training has been previously demon-
strated in healthy people (19), previous work in HFpEF has
failed to show a trend toward improved E/A and DT in people
with HFpEF (20). Unfortunately, study-level (as opposed to
patient-level) meta-analyses do not allow examination of the
relationship between changes in noninvasive measures of car-
diac function to other comprehensive measures of systole/
diastole generated by catheterization, e.g., pressure-volume
loops. Further well-designed and appropriately powered stud-
ies of HFpEF are required to provide clarification of how
noninvasive and catheter-based techniques interrelate to avoid
speculation regarding the probable long-term consequences of
chronically increased left ventricular pressures (8).
Quality of Life and General Health
Our analysis showed HFpEF patients exhibited reductions
(improvements) in MLHFQ scores of similar magnitude to
those seen in patients with systolic heart failure (27). Our
analyses also showed HFpEF patients exhibited increased (im-

Table 2. Study withdrawals and adverse events in the included exercise training studies with HFpEF patients
Withdrawals Adverse Events

Study Exercise Control Exercise Control

Alves et al. (1) 2 3 0 deaths, hospitalizations, or cardiac events 0 deaths, hospitalizations, or cardiac events
Edelmann et al. (6) 2 1 0 deaths or hospitalizations, 5 cardiac events 0 deaths, hospitalizations, or cardiac events
(palpitations or dyspnea)
Gary et al. (9) 1 3 0 deaths, hospitalizations, or cardiac events 0 deaths, hospitalizations, or cardiac events
Karavidas et al. (15) 0 0 0 deaths, hospitalizations, or cardiac events 0 deaths, hospitalizations, or cardiac events
Kitzmann et al. (16) 8 1 0 deaths, 1 hospitalization, and 0 cardiac events 0 deaths, hospitalizations, or cardiac events
Palau et al. (20) 2 2 0 deaths, hospitalizations, or cardiac events 0 deaths, hospitalizations, or cardiac events
Smart et al. (28) 4 1 0 deaths, hospitalizations, or cardiac events 0 deaths, hospitalizations, or cardiac events

J Appl Physiol doi:10.1152/japplphysiol.00904.2014 www.jappl.org

Review
732 Exercise Training in Heart Failure with Preserved Ejection Fraction Dieberg G et al.

Overall TESTEX, total out of 15 points. #Three points possible: 1 point if adherence 85%, 1 point if adverse events are reported, and 1 point if exercise attendance is reported. *Two points possible:
proved) SF-36 scores. The effect size was of a similar magni-

TESTEX
Overall

10
10
8
8
12
10
12
tude to improvements seen previously in heart failure patients
with normal (HFpEF) and abnormal systolic function (26).

Exercise Volume
and Energy
Limitations

Expended

NO
NO
NO
NO
NO
NO
NO
The major limitation of this analysis is that only seven
datasets currently exist, and associated sample sizes were
generally small. In terms of study quality, the median TESTEX
Intensity
Exercise
Relative

Review

score for the included studies indicated of good study quality

YES
YES
YES
YES
YES
YES
YES
and comprehensive reporting. There were insufficient data to
warrant analysis of study withdrawal, adverse events, hospi-
Control Group
Monitoring in

talization, and mortality rates.

Activity

Heterogeneity scores indicated the majority of our analyses

NO
NO
NO
NO
NO
NO
NO
were justified, but those of VE/VCO2 slope and 6MWT may
exhibit heterogeneity at levels too high to justify these analy-
ses.
Variability Reported*
Point Measures and

Meta-analysis of continuous data is problematic; we ad-

Measures of

Downloaded from http://jap.physiology.org/ by 10.220.32.246 on October 2, 2017

YES (2)
YES (2)
YES (2)
YES (2)
YES (2)
YES (2)
YES (2)

justed for baseline difference in primary outcomes between

allocation groups by measuring preintervention vs. postinter-
vention change. Often, we were accurately able to calculate
change in SD, but in cases in which exact P values were not
provided by study reports, we used default values e.g., P
Reporting of Between-

0.05 or P 0.001 in our calculations, which may introduce

Group Statistical
Comparisons

errors.
YES
YES
YES
YES
YES
YES
YES

Our funnel plots appear to suggest negligible risk of publi-

cation bias. We suggest that unpublished datasets with perhaps
negative results do not exist.
Finally, we acknowledge that factors related to volume of
exercise may explain some of the outcomes reported; for
Intention to

Analysis

example, intuitively one suspects longer study duration varia-

YES
YES
YES
YES
YES

YES
Treat

NO

tions may yield better results, but the small number of included
studies precluded subanalyses of exercise volume parameters.
Table 3. Study quality assessment of included studies using TESTEX scale

Previous work by one of our authors shows that people with

Outcome Measures
Assessed 85% of

Unclear (0)
Participants#

HFpEF typically tend to be five or more years older and more

YES (2)
YES (2)

YES (3)
YES (2)
YES (3)

likely to be female than their systolic dysfunction counterparts

1 point if primary outcome is reported and 1 point if all other outcomes are reported.
YES

(26). While the precise volume of screening to identify people

with HFpEF for this work were not published, it can be
revealed that screening of more than 4,000 echocardiograms
Assessors

yielded less than 20 exercise training participants. The expla-

Blinded

YES
YES
NO
NO
NO
NO

NO

nation of this low yield is that for the purposes of scientific

rigor, trials of HFpEF need to demonstrate participants do not
have ischemic heart disease. Isolation from any hint of isch-
Similar at
Baseline
Groups

YES
YES
YES
YES
YES
YES
YES

emic heart disease is rare in HFpEF and, therefore, makes

suitable trial participants extremely difficult to find. This is
possibly the primary explanation why over 100 randomized
Allocation
Concealed

exercise training trials exist in systolic heart failure patients,

YES
NO
NO
NO
NO
NO
NO

while only 7 exist to date in HFpEF.

In terms of expected health benefits for people with HFpEF
Participants

who undertake exercise, one may expect extended survival in

Randomly
Allocated

YES
YES
YES
YES
YES
YES
YES

one person for approximately every 17 people treated for one

year with exercise (24). These data were generated from
systolic heart failure studies, but as the expected, improvement
Eligibility

Specified

in peak VO2 from exercise training is similar in HFpEF, one

Criteria

YES
YES
YES
YES
YES
YES
YES

would expect the survival benefit to be comparable.

Reference Study

Conclusions
Ref. 1
Ref. 6
Ref. 9
Ref.15
Ref.16
Ref.20
Ref.28

Exercise training does yield improvements in cardiorespira-

tory fitness, diastolic function, quality of life, and general
health in heart failure patients with preserved ejection fraction.

J Appl Physiol doi:10.1152/japplphysiol.00904.2014 www.jappl.org


Exercise Training in Heart Failure with Preserved Ejection Fraction
GRANTS
This work received no financial support and has no relationship to industry.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the authors.
AUTHOR CONTRIBUTIONS
Author contributions: G.D., H.I., F.G., and N.A.S. conception and design of
research; G.D., H.I., F.G., and N.A.S. analyzed data; G.D., H.I., F.G., and
N.A.S. interpreted results of experiments; G.D., F.G., and N.A.S. prepared
figures; G.D., H.I., F.G., and N.A.S. drafted manuscript; G.D., H.I., F.G.,
and N.A.S. edited and revised manuscript; G.D., H.I., F.G., and N.A.S.
approved final version of manuscript.
REFERENCES
1. Alves AJ, Ribeiro F, Goldhammer E, Rivlin Y, Rosenschein U, Viana
JL, Duarte JA, Sagiv M, Oliveira J. Exercise training improves diastolic
function in heart failure patients. Med Sci Sports Exerc 44: 776 785,
2012.
2. Bhatia RS, Tu JV, Lee DS, Austin PC, Fang J, Haouzi A, Gong Y, Liu
PP. Outcome of heart failure with preserved ejection fraction in a popu-
lation-based study. N Engl J Med 355: 260 269, 2006.
3. Brazier JE, Harper R, Jones NM, OCathain A, Thomas KJ, Usher-
wood T, Westlake L. Validating the SF-36 health survey questionnaire:
new outcome measure for primary care. BMJ 305: 160 164, 1992.
4. Burkhoff D. Mortality in heart failure with preserved ejection fraction: an
unacceptably high rate. Eur Heart J 33: 1718 1720, 2012.
5. Bursi F, Weston SA, Redfield MM, Jacobsen SJ, Pakhomov S, Nkomo
VT, Meverden RA, Roger VL. Systolic and diastolic heart failure in the
community. JAMA 296: 2209 2216, 2006.
6. Edelmann F, Gelbrich G, Dungen HD, Frhling S, Wachter R, Stah-
renberg R, Binder L, Tpper A, Lashki DJ, Schwarz S, Herrmann-
Lingen C, Lffler M, Hasenfuss G, Halle M, Pieske B. Exercise training
improves exercise capacity and diastolic function in patients with heart
failure with preserved ejection fraction: results of the Ex-DHF (Exercise
training in Diastolic Heart Failure) pilot study. J Am Coll Cardiol 58:
1780 1791, 2011.
7. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-
analysis detected by a simple, graphical test. BMJ 315: 629 634, 1997.
8. Fontes-Carvalho R, Leite-Moreria A. Heart failure with preservd ejec-
tion fraction: fighting misconceptions for a new approach. Arq Bras
Cardiol 96: 504 514, 2011.
9. Gary RA, Sueta CA, Dougherty M, Rosenberg B, Cheek D, Preisser J,
Neelon V, McMurray R. Home-based exercise improves functional
performance and quality of life in women with diastolic heart failure.
Heart Lung 33: 210 218, 2004.
10. Guazzi M, Myers J, Peberdy MA, Bensimhon D, Chase P, Pinkstaff S,
Arena R. Echocardiography with tissue Doppler imaging and cardiopul-
monary exercise testing in patients with heart failure: a correlative and
prognostic analysis. Int J Cardiol 143: 323329, 2010.
11. Hambrecht R, Hilbrich L, Erbs S, Gielen S, Fiehn E, Schoene N,
Schuler G. Correction of endothelial dysfunction in chronic heart failure:
additional effects of exercise training and oral L-arginine supplementation.
J Am Coll Cardiol 35: 706 713, 2000.
12. 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 57: 1676 1686, 2011.
13. Ismail H, McFarlane JR, Dieberg G, Smart NA. Exercise training
program characteristics and magnitude of change in functional capacity of
heart failure patients. Int J Cardiol 171: 6265, 2014.
14. Ismail H, McFarlane JR, Nojoumian AH, Dieberg G, Smart NA.
Clinical outcomes and cardiovascular responses to different exercise
training intensities in patients with heart failure: a systematic review and
meta-analysis. JACC Heart Fail 1: 514 522, 2013.
J Appl Physiol doi:10.1152/japplphysiol.00904.2014 www.jappl.org
Review
Dieberg G et al. 733
15. Karavidas A, Driva M, Parissis JT, Farmakis D, Mantzaraki V,
Varounis C, Paraskevaidis I, Ikonomidis I, Pirgakis V, Anastasiou-
Nana M, Filippatos G. Functional electrical stimulation of peripheral
muscles improves endothelial function and clinical and emotional status in
heart failure patients with preserved left ventricular ejection fraction. Am
Heart J 166: 760 767, 2013.
16. Kitzman DW, Brubaker PH, Herrington DM, Morgan TM, Stewart
KP, Hundley WG, Abdelhamed A, Haykowsky MJ. Effect of endur-
ance exercise training on endothelial function and arterial stiffness in older
patients with heart failure and preserved ejection fraction: a randomized,
controlled, single-blind trial. J Am Coll Cardiol 62: 584 592, 2013.
17. Kraigher-Krainer E, Shah AM, Gupta DK, Santos A, Claggett B,
Pieske B, Zile MR, Voors AA, Lefkowitz MP, Packer M, McMurray
JJ, Solomon SD, PARAMOUNT Investigators. Impaired systolic func-
tion by strain imaging in heart failure with preserved ejection fraction. J
Am Coll Cardiol 63: 447456, 2014.
18. Lam CS, Donal E, Kraigher-Krainer E, Vasan RS. Epidemiology and
clinical course of heart failure with preserved ejection fraction. Eur J
Heart Fail 13: 18 28, 2011.
19. Obert P, Mandigout S, Vinet A, NGuyen LD, Stecken F, Courteix D.
Effect of aerobic training and detraining on left ventricular dimensions and
diastolic function in prepubertal boys and girls. Int J Sports Med 22:
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on October 2, 2017
90 96, 2001.
20. Palau P, Dominguez E, Nunez E, Schmid JP, Vergara P, Ramn JM,
Mascarell B, Sanchis J, Chorro FJ, Nez J. Effects of inspiratory
muscle training in patients with heart failure with preserved ejection
fraction. Eur J Prev Cardiol 21: 14651473, 2014.
21. Rector TS, Cohn JN. Assessment of patient outcome with the Minnesota
Living with Heart Failure questionnaire: reliability and validity during a
randomized, double-blind, placebo-controlled trial of pimobendan. Pimo-
bendan Multicenter Research Group. Am Heart J 124: 10171025, 1992.
22. Roger VL, Go AS, Lloyd-Jones DM, Benjamin EJ, Berry JD, Borden
WB, Bravata DM, Dai S, Ford ES, Fox CS, Fullerton HJ, Gillespie C,
Hailpern SM, Heit JA, Howard VJ, Kissela BM, Kittner SJ, Lackland
DT, Lichtman JH, Lisabeth LD, Makuc DM, Marcus GM, Marelli A,
Matchar DB, Moy CS, Mozaffarian D, Mussolino ME, Nichol G,
Paynter NP, Soliman EZ, Sorlie PD, Sotoodehnia N, Turan TN, Virani
SS, Wong ND, Woo D, Turner MB, American Heart Association
Statistics Committee and Stroke Statistics Subcommittee. Heart dis-
ease and stroke statistics2012 update: a report from the American Heart
Association. Circulation 125: e2e220, 2012.
23. Sarullo FM, Fazio G, Brusca I, Fasullo S, Paterna S, Licata P, Novo
G, Novo S, Di Pasquale P. Cardiopulmonary exercise testing in patients
with chronic heart failure: prognostic comparison from peak VO2 and
VE/VCO2 slope. Open Cardiovasc Med J 4: 127134, 2010.
24. Smart N. What is the number needed to treat when exercise training heart
failure patients? J Exer Physiol online, 12: 15, http://epublications.bond-
.edu.au/hsm_pubs/142.
25. Smart N. Exercise training for heart failure patients with and without
systolic dysfunction: an evidence-based analysis of how patients benefit.
Cardiol Res Pract pii. 837238, 2011.
26. Smart N, Haluska B, Jeffriess L, Marwick TH. Exercise training in
systolic and diastolic dysfunction: effects on cardiac function, functional
capacity, and quality of life. Am Heart J 153: 530 536, 2007.
27. Smart N, Haluska B, Jeffriess L, Case C, Marwick TH. Cardiac
contributions to exercise training responses in patients with chronic heart
failure: a strain imaging study. Echocardiography 23: 376 382, 2006.
28. Smart NA, Haluska B, Jeffriess L, Leung D. Exercise training in heart
failure with preserved systolic function: a randomized controlled trial of
the effects on cardiac function and functional capacity. Congest Heart Fail
18: 295301, 2012.
29. Smart NAWM, Ismail H, Giallauria F, Vigorito C, Cornelissen V,
Dieberg G. Validation of a new tool for the assessment of study quality
and reporting in exercise training studies:TESTEX. Int J Evid Based
Healthc 13: 9 18, 2015.
30. Taylor RS, Davies EJ, Dalal HM, Davis R, Doherty P, Cooper C,
Holland DJ, Jolly K, Smart NA. Effects of exercise training for heart
failure with preserved ejection fraction: A systematic review and meta-
analysis of comparative studies. Int J Cardiol 162: 6 13, 2012.