Author + information
- Received September 11, 2017
- Accepted November 26, 2017
- Published online January 30, 2018.
- Alison M. Mudge, PhDa,b,∗ (, )
- Charles P. Denaro, MDa,b,
- Adam C. Scott, PhDc,
- Deborah Meyers, MDd,
- Julie A. Adsett, BPhty (Hons)e,
- Robert W. Mullins, MAppSci (Clin Ex Sci)f,
- Jessica M. Suna, MAppScia,f,
- John J. Atherton, PhDb,c,
- Thomas H. Marwick, PhDg,
- Paul Scuffham, PhDh and
- Peter O’Rourke, PhDi
- aDepartment of Internal Medicine and Aged Care, Royal Brisbane and Women’s Hospital, Brisbane, Queensland, Australia
- bUniversity of Queensland Faculty of Medicine, Brisbane, Queensland, Australia
- cDepartment of Cardiology, Royal Brisbane and Women’s Hospital, Brisbane, Queensland, Australia
- dTexas Heart Institute, Houston, Texas
- eHeart Support Service, Royal Brisbane and Women’s Hospital, Brisbane, Queensland, Australia
- fQueensland University of Technology, Brisbane, Queensland, Australia
- gBaker IDI Heart and Diabetes Institute, Melbourne, Victoria, Australia
- hCentre for Applied Health Economics, Menzies Health Institute Queensland, Griffith University, Brisbane, Queensland, Australia
- iStatistics Unit, QIMR Berghofer, Brisbane, Queensland, Australia
- ↵∗Address for correspondence:
Dr. Alison M. Mudge, Internal Medicine Research, Level 9 Ned Hanlon Building, Royal Brisbane and Women’s Hospital, Butterfield Street, Herston, Queensland 4029, Australia.
Objectives This study sought to measure the impact on all-cause death or readmission of adding center-based exercise training (ET) to disease management programs for patients with a recent acute heart failure (HF) hospitalization.
Background ET is recommended for patients with HF, but evidence is based mainly on ET as a single intervention in stable outpatients.
Methods A randomized, controlled trial with blinded outcome assessor, enrolling adult participants with HF discharged from 5 hospitals in Queensland, Australia. All participants received HF-disease management program plus supported home exercise program; intervention participants were offered 24 weeks of supervised center-based ET. Primary outcome was all-cause 12-month death or readmission. Pre-planned subgroups included age (<70 years vs. older), sex, left ventricular ejection fraction (≤40% vs. >40%), and exercise adherence.
Results Between May 2008 and July 2013, 278 participants (140 intervention, 138 control) were enrolled: 98 (35.3%) age ≥70 years, 71 (25.5%) females, and 62 (23.3%) with a left ventricular ejection fraction of >40%. There were no adverse events associated with ET. There was no difference in primary outcome between groups (84 of 140 [60.0%] intervention vs. 90 of 138 [65.2%] control; p = 0.37), but a trend toward greater benefit in participants age <70 years (OR: 0.56 [95% CI: 0.30 to 1.02] vs. OR: 1.56 [95% CI: 0.67 to 3.64]; p for interaction = 0.05). Participants who exercised to guidelines (72 of 101 control and 92 of 117 intervention at 3 months) had a significantly lower rate of death and readmission (91 of 164 [55.5%] vs. 41 of 54 [75.9%]; p = 0.008).
Conclusions Supervised center-based ET was a safe, feasible addition to disease management programs with supported home exercise in patients recently hospitalized with acute HF, but did not reduce combined end-point of death or readmission. (A supervised exercise programme following hospitalisation for heart failure: does it add to disease management?; ACTRN12608000263392)
Exercise training (ET) has physiological and clinical outcome benefits in stable outpatients with heart failure (HF), with improved exercise tolerance, improved quality of life, and reduced HF-related and all-cause readmissions (1). Current guidelines recommend that patients with HF should be encouraged to undertake regular aerobic exercise (2,3), including referral to supervised ET programs (4). However, there are gaps in the evidence (1,5). Trials have enrolled patients with chronic HF stabilized on optimal medical therapy, so that the feasibility, safety, and effectiveness of ET in the early post-hospital setting remain unclear (6). ET has largely been tested as a single intervention, despite evidence that coordinated multidisciplinary HF disease management programs (HF-DMPs) reduce mortality and readmissions in recently hospitalized patients with HF (2,7). HF-DMPs offer an opportunity to provide supported home exercise advice, and may also provide an efficient setting for center-based ET (8). Limited trials have examined ET in the setting of HF-DMP (9–11), and whether ET has an additive effect to HF-DMPs is unclear. Supported home ET may have similar benefits to supervised center-based ET (12), but no trial has compared the addition of center-based ET to a supported home exercise program. Women (13), older patients (14), and patients with heart failure with preserved ejection fraction (HFpEF) (15,16) remain underrepresented in ET trials. Access to center-based ET remains limited (17), and exercise adherence is a challenge in real-world practice (18).
The aim of this pragmatic, multicenter, randomized, controlled trial was to investigate the feasibility, safety, and effectiveness of delivering supervised center-based ET as part of a HF-DMP for recently hospitalized patients with HF compared with HF-DMP alone. In recognition of current guidelines, HF-DMP included a supported home exercise program. Our primary hypothesis was that participants allocated to supervised center-based ET plus HF-DMP would have a reduction in the composite outcome of all-cause death and hospitalization over 12 months compared with HF-DMP. We also aimed to explore the benefits in women, older participants, and participants with HFpEF, and measure the impact of the combined intervention on exercise outcomes and quality of life.
The EJECTION-HF (Exercise Joins Education: Combined Therapy to Improve Outcomes in Newly-discharged Heart Failure) was a multicenter randomized controlled trial conducted in 5 hospitals in Queensland, Australia, which each had an established HF-DMP (19). Human research ethics approval was obtained from all sites, with central coordination by the Royal Brisbane and Women’s Hospital Human Research Ethics Committee.
Adult hospital inpatients with HF as a major contributing diagnosis were identified by HF-DMP staff and screened for eligibility by a research officer. Detailed inclusion and exclusion criteria have been reported previously (19); in brief, potential participants were willing and able to attend a 12-week HF-DMP, and assessed as suitable for supervised moderate intensity ET. All participants provided written informed consent.
Randomization and masking
Participants were randomized to control (HF-DMP with supported home exercise program), or intervention (HF-DMP with supported home exercise program plus supervised center-based ET). Allocation was undertaken by the project manager, after obtaining consent and baseline assessment by the research officer. Allocation was based on computerized random sequence generation by the statistician in blocks of 10, stratified by site, provided in sealed opaque envelopes to the project manager. The project manager notified the HF-DMP of participant allocation and helped to coordinate their program commencement, and was responsible for data management, but not program delivery or data collection. Participants and HF-DMP staff were aware of their treatment group. Clinical outcomes were collected by a single trained research officer masked to group assignment.
HF-DMP (all participants)
All sites offered a multidisciplinary HF-DMP including active in-hospital case finding, clinic visits, and telephone support to provide education, symptom monitoring, medication titration, and social support. The HF-DMP included 12 weeks of weekly review with disease monitoring, review and progression of the home exercise program, and weekly structured self-management education classes (HF pathophysiology, symptom monitoring, medications, dietary advice, stress management, risk factor modification, and exercise advice). The HF-DMP was delivered with a rolling start to support timely commencement. Each HF-DMP had access to a cardiologist, HF trained nursing staff, pharmacist, and exercise professional (physiotherapist and/or exercise physiologist), although programs were adapted to local resourcing. Two investigators (senior exercise professionals in one HF-DMP) developed the exercise protocols and supported training, mentoring, and support for other HF-DMP staff. An individualized home exercise program was prescribed by the local HF-DMP exercise professional for each participant (both groups), aiming to achieve moderate intensity aerobic exercise (e.g., walking at moderate pace and/or use of participants’ own stationary bike, treadmill, or other fitness equipment) for 30 min 5 days per week, plus simple resistance exercises using light weights, Theraband or body weight. Intensity was prescribed and self-monitoring taught using rating of perceived exertion on a 6- to 20-point scale. Each participant was provided with an exercise diary to assist weekly review and progression.
Supervised center-based ET (intervention group only)
Intervention group participants were encouraged to attend twice weekly ET sessions for 12 weeks followed by once weekly sessions for 12 weeks, aiming for a total of 36 sessions over 24 weeks (20). ET was individually prescribed by the HF-DMP exercise professional and undertaken in group format in a gymnasium at each site. ET was supervised by a nurse and exercise professional and included 50 min of graded moderate intensity interval and resistance training plus warm up and cool down. Participants used traditional gym-based equipment such as stationary bikes, treadmills, steps, free weights, and multistation resistance equipment. Balance activities were included if appropriate. ET was progressed weekly maintaining rating of perceived exertion of 9 to 14. Transport assistance was offered for 1 session per week.
The primary outcome was a composite of death or readmission from any cause within 12 months of randomization. Specified secondary analyses included individual outcomes (death, all-cause readmission, HF readmission), time to primary outcome, and number and total days of hospitalization in the first 12 months. All deaths (date, cause, location) occurring by December 31, 2014, were identified from the Queensland state death registry, and cross-checked with clinical records. Cause of death was classified as progressive HF, sudden or arrhythmic death, other cardiovascular cause, and noncardiovascular cause. Readmissions were identified from clinical records and the state-wide hospital information system, which includes all admissions to public hospitals in Queensland. Cause of admission was classified by the project manager and a physician unaware of treatment allocation based on discharge coding and review of discharge summaries if required. Causes were grouped as HF, possibly HF treatment related (e.g., dehydration, pre-syncope, device-related admissions), arrhythmic (atrial or ventricular arrhythmias), other cardiovascular (e.g., ischemic heart disease, stroke), or noncardiovascular. Dates of first readmission and first HF readmission and number of admissions and total days in hospital in 12 months were recorded. Nonadmitted emergency department attendances were coded separately.
Adverse events were defined as symptoms (e.g., chest pain or syncope), documented arrhythmia, and fall or other injury occurring during the center-based ET program that required medical review. Exercise outcomes included exercise capacity (6-min walk distance) and physical activity (self-report of home exercise adherence). The 6-min walk distance was evaluated at baseline, and 3 and 6 months by the research officer using recommended procedures (21), and measured to the nearest 1 m; the test was repeated at baseline to minimize learning effect. Change in 6-min walk distance was calculated at 3 and 6 months compared with the maximum baseline measure, and clinically significant improvement (21,22) defined as an increase of ≥30 m (unless the baseline distance was ≥500 m, where those with no decrease at the second reading were also classified as improvement to account for the known ceiling effect). Home exercise levels were assessed at baseline, and 3 and 6 months by self-report of minutes of exercise per week and dichotomized based on general adult exercise recommendations (150 min per week moderate or vigorous exercise). Quality of life was measured at baseline and 6 months using the Assessment of Quality of Life (AQoL-4D), a multi-attribute utility measure designed for use in public health and acute care interventions with weighted scores ranging from −0.04 (minimum) to 1.0 (maximum), with Australian population norms of 0.81 (23) .
Characteristics and process measures
Participant characteristics were obtained from structured assessment by a single research assistant before randomization, supplemented by medical record review and basic laboratory investigations. The left ventricular ejection fraction was obtained from the most recent echocardiogram performed in the previous 12 months, and dichotomized as HF with a reduced ejection fraction (left ventricular ejection fraction of ≤40%) and HFpEF (left ventricular ejection fraction of >40%). Individual attendance at education and exercise classes was monitored and recorded weekly by HF-DMP staff. Pharmacotherapy, outpatient visits, and telephone follow-up were obtained from HF-DMP clinical records, medical record review, and outpatient information systems. Transport assistance was offered for the baseline, and 3- and 6-month follow-up visits.
All analyses were by intention to treat, that is, all participants were included in the analysis (according to their allocation group) regardless of the intervention received. Participant characteristics and processes of care were summarized and compared between groups using the chi-square test for categorical variables and 1-way analysis of variance or the Mann–Whitney U tests for continuous variables. The primary outcome was analyzed by comparing the proportion of control and intervention participants experiencing death or readmission within 12 months of randomization using the chi-square test. Power calculations assumed a reduction in the primary outcome (death or readmission) from 60% to 45%, alpha 0.05 and 80% power, with planned sample size of 175 participants per group (350 total). Time to death or readmission was investigated using Kaplan–Meier curves and the log-rank test. Days per admitted patient and days alive out of hospital were compared between groups using the Mann–Whitney U test.
Pre-planned subgroup analyses for the primary outcome included age (<70 years vs. ≥70 years), gender, and HF with reduced ejection fraction versus HFpEF, using interaction testing for randomization status*subgroup status in logistic regression models for the primary outcome. Good versus poor adherence was conceptualized in 3 ways. Attendance at group-based HF-DMP sessions was dichotomized for both groups at median attendance (6 education classes). Self-reported home-based exercise at 3-month follow-up was dichotomized into <150 min of moderate exercise per week versus ≥150 min per week. In an exploratory analysis within the intervention group, we compared the primary outcome by quartiles of ET attendance.
Change in 6-min walk distance was compared using the Student t test. The proportion of participants improving on the 6-min walk test, and the proportion adhering to exercise guidelines, were compared using the chi-square test. Utility scores were generated using a weighted algorithm supplied by the developers and scores at 6 months compared between groups using the Student t test.
Analyses were conducted using SPSS version 22 (SPSS, Inc., Chicago, Illinois). A p < 0.05 was considered significant in all analyses.
Between May 2008 and July 2013, 2,551 inpatients with HF were screened, 615 (24.1%) were eligible, and 278 (45.2% of eligible) consented to participate and were randomized (Figure 1). Online Table 1 shows screening and recruitment at each site. The median time from hospital discharge to randomization visit was 30 days (interquartile range [IQR]: 21 to 42 days). The median time from hospital discharge to program commencement was 43 days (IQR: 29 to 58 days). Delays beyond 6 weeks were due to illness and readmission, service closure during holidays, space limitations in classes, and individual participant issues. Three participants crossed over from control to intervention. There were no serious adverse events reported. No participant formally withdrew from the study, and all participants were included in the primary outcome analysis. The 3-month follow-up visit was attended by 101 control and 117 intervention participants; the 6-month follow-up visit was attended by 92 control and 105 intervention participants.
Participant clinical and treatment characteristics are shown in Tables 1 and 2. Control and intervention groups were well-matched. Ninety-eight participants (35.3%) were aged ≥70 years, 71 (25.5%) were female, and 62 (23.3%) had HFpEF. Medication use was similar between the 2 groups (Table 1). Participants in both groups attended a median of 6 education sessions (IQR: 2 to 8) (Table 2). Only 2 participants attended 36 center-based exercise sessions, 28 (20.0%) attended ≥24, 60 (42.9%) attended ≥18, and 86 (61.4%) attended ≥12. Intervention participants had fewer HF clinic visits, but similar cardiology clinic and telephone follow-up (Table 2).
In the 12 months after randomization, 174 participants (62.6%) died or were readmitted to hospital, with a nonsignificant trend to reduction in the intervention group (odds ratio [OR]: 0.80; 95% CI: 0.49 to 1.30) (Table 3). There were significantly fewer deaths within 12 months in the intervention compared with control group (3 [2.1%] vs 10 [7.2%]; p = 0.04). Causes of death in the intervention group included 1 HF, 1 other cardiovascular, and 1 noncardiovascular death; in the control group there were 4 HF, 3 arrhythmic, 1 other cardiovascular, and 2 noncardiovascular deaths. There was no difference in all-cause or HF readmissions between the control and intervention groups (Table 3). In the first 12 months, there were 404 hospital readmissions: 82 (20.2%) due to HF, 18 (4.4%) possibly HF related, 33 (8.2%) due to arrhythmia, 74 (18.3%) due to other cardiovascular causes, and 197 (48.7%) due to noncardiovascular causes. Time to first readmission or death is shown in Figure 2, with no difference between groups (log-rank test p = 0.31). There was no difference in the 6-month utility scores between groups (intervention mean 0.60 ± 0.28 vs. control 0.61 ± 0.28; p = 0.63), with mean scores substantially lower than population norms (23).
Pre-planned subgroup analysis showed a borderline interaction (p = 0.05) between allocation groups and age in regression modeling, with participants aged <70 years more likely to benefit from ET than older participants (OR: 0.56 [95% CI: 0.30 to 1.02; p = 0.06] vs. OR: 1.56 [95% CI: 0.67 to 3.64; p = 0.31]). Kaplan–Meier curves (Figure 2) show a significantly decreased risk of death or readmission in younger participants allocated to the intervention group throughout 12-month follow-up (log-rank test p = 0.026), but no difference between groups in participants age >70 years. Participants with HFpEF had higher death and readmission overall (OR: 1.99; 95% CI: 1.02 to 3.88; p = 0.04), but there was no interaction with allocation group. There was no difference in outcomes or treatment effect by gender.
The 6-min walk test distance is shown in Table 4. About one-half of participants in both groups showed a clinically significant improvement in 6-min walk distance at 3 and 6 months, with no difference between groups. Adherence to the supported home exercise program was high in both groups. At baseline, 61 of 138 control (44.2%) and 60 of 140 intervention (42.9%) participants reported exercising 150 min per week. At 3 months, 72 of 101 control participants (71.3%) attending follow-up and 92 of 117 (78.6%) intervention participants reported meeting home exercise guidelines (p = 0.21); at 6 months 64 of 92 control participants (69.6%) and 69 of 105 intervention participants (65.7%) reported meeting guidelines (p = 0.57). Self-reported exercise at 3 and 6 months was strongly associated with death and readmission. The primary outcome occurred in 91 of 164 participants (55.5%) meeting exercise guidelines at 3 months versus 41 of 54 (75.9%) not meeting guidelines (p = 0.008), and 73 of 133 (54.9%) in those meeting exercise guidelines at 6 months versus 47 of 64 (73.4%) not meeting guidelines (p = 0.012). There was no significant interaction of this effect with allocation group. Better attendance at HF-DMP group education sessions was not associated with death or readmission, which occurred in 80 of 121 (66.1%) of those attending <6 sessions and 94 of 157 (59.9%) of those attending ≥6 (p = 0.29). There was no association with greater ET attendance (<8 sessions 21 of 30 [70.0%], 8 to 15 sessions 28 of 43 [65.1%], 16 to 23 sessions 19 of 39 [48.7%], ≥24 sessions 16 of 28 [57.1%]; p = 0.27).
We undertook a pragmatic, multicenter, randomized, controlled trial to evaluate the feasibility, safety, and effectiveness of delivering supervised center-based ET as part of HF-DMP in patients with HF recently discharged from the hospital. Program delivery was feasible, with one-half of participants commencing their allocated treatment within 6 weeks; a previous study aiming for a 2-week time frame found this was not feasible (6). Improvements in the 6-min walk test were modest, and did not differ between the groups. Three-quarters of participants reported meeting exercise guidelines at 3 months, and 68% at 6 months, demonstrating the effectiveness of the HF-DMP in supporting home exercise in both groups. However, center-based ET adherence was poor, with only 43% of intervention participants attending one-half or more of their scheduled sessions. No adverse events were related to ET, consistent with existing safety data in stable outpatients enrolled in ET trials (1,20). There was no reduction in the primary endpoint of death and readmission at 12 months. Mortality was lower in the intervention group; this was based on a small number of events and should be interpreted with caution.
Meta-analyses of ET in HF have shown a reduction in HF and all-cause readmissions (1,24). Several factors may explain why our study did not. First, participants were enrolled in a comprehensive HF-DMP, which reduces avoidable readmissions, so the potential for further reduction in readmission may have been low. Second, all participants received an individualized, supported, home-based exercise program, with most participants from both groups reporting that they met home exercise guidelines, which likely reduced the incremental benefits from center-based ET. A no-exercise comparison was no longer standard of care when we designed our trial. Third, more than one-third of these participants reported meeting exercise guidelines at baseline, which may reduce the potential to benefit from additional exercise. Fourth, adherence to center-based ET was modest, suggesting that high adherence reported in single site trials with highly selected participants may not be achievable in a real-life setting. Finally, we enrolled a more representative HF population with greater multimorbidity than many previous trials, and most readmissions were for causes other than HF.
Subgroup analyses suggested that younger participants (age <70 years) may have benefited more from the intervention than older participants. Previous meta-analysis of trials in older patients with HF undertaking ET (14) showed improvements in exercise capacity and quality of life, but not mortality or readmissions, although a recent multisite Italian trial showed a reduction in hospitalizations in stable older outpatients undertaking ET (25). Their trial had a self-selected sample and high adherence to center-based ET delivered 3 times per week. It is possible that our training stimulus was not sufficient for the older participant subgroup. Older patients may have less central adaptation to ET (26), and experience more impairment in muscle function and greater deconditioning with acute HF (27). Our sample of recently hospitalized patients had high rates of comorbidity, including diabetes, chronic lung disease, and musculoskeletal disorders, which may limit exercise prescription and progression.
Our study is unique in providing contemporary, multidisciplinary HF-DMP including supported home exercise advice in both arms, consistent with current HF guidelines. We recognize that this design likely reduced the impact of the intervention, because both groups improved exercise capacity and physical activity to a similar extent. A previous trial comparing a specialist nurse–led HF-DMP with and without supported home exercise did not show any difference in death, HF readmission, or quality of life between the groups (10). Their trial provided less intensive home exercise support, and achieved lower exercise adherence at 6 months (54%) than our study. A review of home-based ET in HF studies showed better exercise outcomes and quality of life when home-based ET is compared with usual care, and no difference in exercise outcomes between home-based and center-based exercise rehabilitation (12). We are not aware of any previous study reporting the incremental benefit of center-based ET in addition to a supported home exercise program.
Both groups reported increased physical activity at 3 and 6 months, and adherence to exercise guidelines was associated with significantly better clinical outcomes. This reflects use of strategies for exercise adherence, including individual goal setting, regular review and feedback, and the group HF-DMP setting (18,28). Although the center-based ET offered more frequent feedback, supervision, and transport assistance, these strategies did not significantly enhance physical activity in the intervention group. Few studies have examined the impact of center-based ET on physical activity. The HF-ACTION trial (20) provided individualized ET, commencing with 36 sessions of supervised center-based ET and transitioning to home exercise with telephone support. Unlike our study, this intervention was not integrated into HF-DMP, and only 30% of their intervention group met exercise goals at 3 and 6 months (120 min of exercise per week). The intervention group in HF-ACTION showed similar changes in 6-min walk distance to our study, and secondary analysis showed a significant association between physical activity and all-cause hospitalization and mortality (29).
Other strengths of our study include multiple centers, a pragmatic exercise program delivered by clinical rather than research staff, and broad enrollment criteria, enhancing the generalizability of our findings. Our study also has several weaknesses. It was underpowered for the assumed effect size (only 80% of planned sample size recruited due to slower recruitment than anticipated), which may have contributed to the negative outcome. Only 11% of screened patients were recruited, similar to other post-hospital ET and DMP trials, reflecting the challenge of recruiting unwell patients to complex interventions (9,10,30), and undersampling older, more complex patients. Good baseline levels of self-reported physical activity suggest selection of motivated participants. Although subgroups were pre-specified, women, older patients and patients with HFpEF remained under-represented despite an inclusive recruitment strategy. This limits the confidence in subgroup findings, and has relevance to the potential reach of center-based ET programs. Participants and program staff could not be blinded. Exercise adherence was measured by self-report and only available on participants who attended follow-up. Although the multiple secondary outcomes, including subgroup analyses and exercise outcomes, help in the understanding of this complex intervention, these secondary findings should be considered exploratory because of the risks of multiple comparisons.
Our study shows that both supervised center-based ET and a supported home exercise program can be delivered safely and effectively as part of an HF-DMP within a few weeks of discharge after hospitalization for acute HF for selected patients, with good improvements in adherence to exercise guidelines despite modest improvements in 6-min walk distance. There was no decrease in the primary outcome of death or readmission with center-based ET, but there may have been a benefit for participants age <70 years. Adherence to exercise guidelines was a strong predictor of outcomes in both groups. Further investigation of factors affecting exercise adherence, and those participants most likely to participate and benefit, would inform more effective and efficient ways to incorporate ET into HF-DMP.
COMPETENCY IN MEDICAL KNOWLEDGE: ET is safe and feasible to deliver as part of a HF disease management program within a few weeks of hospitalization for selected patients with acute HF. Patients recently hospitalized with HF who are suitable for ET should be encouraged to meet exercise guidelines.
TRANSLATIONAL OUTLOOK: A better understanding of exercise adherence patterns (in center and at home) may help to direct rehabilitation resources. Future studies should focus on supporting greater adherence to exercise guidelines in patients with HF using flexible programs, and investigate how best to adapt exercise support programs for different patient subgroups. Integrating ET into HF-DMP may be a practical strategy for delivering exercise rehabilitation for patients with HF.
The authors thank Diane Bookless for data collection, staff from the participating Heart Failure Disease Management Programs for assisting with trial recruitment and conduct, and all study participants.
The trial was funded by Australian National Health and Medical Research Council Project Grant number 498403 with contributions from the Royal Brisbane Hospital Foundation, Brisbane, Queensland, Australia (including a Patricia Dukes Fellowship award to Dr. Mudge) and the Prince Charles Hospital Foundation, Brisbane, Queensland, Australia. The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- exercise training
- heart failure
- heart failure disease management program
- heart failure with preserved ejection fraction
- interquartile range
- odds ratio
- Received September 11, 2017.
- Accepted November 26, 2017.
- 2018 American College of Cardiology Foundation
- Sagar V.A.,
- Davies E.J.,
- Briscoe S.,
- et al.
- Ponikowski P.,
- Voors A.A.,
- Anker S.D.,
- et al.
- Yancy C.W.,
- Jessup M.,
- Bozkurt B.,
- et al.
- National Clinical Guidelines Centre
- Fleg J.L.,
- Cooper L.S.,
- Borlaug B.A.,
- et al.
- Houchen-Wolloff L.,
- Watt A.,
- Schreder S.,
- Singh S.
- Takeda A.,
- Taylor S.J.,
- Taylor R.S.,
- Khan F.,
- Krum H.,
- Underwood M.
- Reed S.D.,
- Whellan D.J.,
- Li Y.,
- et al.
- Jolly K.,
- Taylor R.S.,
- Lip G.Y.,
- et al.
- Zwisler A.D.,
- Norton R.J.,
- Dean S.G.,
- et al.
- Pina I.L.,
- Bittner V.,
- Clare R.M.,
- et al.
- Chen Y.M.,
- Li Y.
- Chan E.,
- Giallauria F.,
- Vigorito C.,
- Smart N.A.
- Pandey A.,
- Parashar A.,
- Kumbhani D.J.,
- et al.
- Conraads V.M.,
- Deaton C.,
- Piotrowicz E.,
- et al.
- Holland A.E.,
- Spruit M.A.,
- Troosters T.,
- et al.
- Antonicelli R.,
- Spazzafumo L.,
- Scalvini S.,
- et al.
- Acanfora D.,
- Scicchitano P.,
- Casucci G.,
- et al.
- Butrous H.,
- Hummel S.L.
- Keteyian S.J.,
- Leifer E.S.,
- Houston-Miller N.,
- et al.
- Stewart S.,
- Carrington M.J.,
- Marwick T.H.,
- et al.