Combined strength and aerobic training vs. aerobic training alone in patients with heart failure: A systematic review and meta-analysis

Highlights

• •Low cardiorespiratory fitness (CRF) and muscle strength increase the risk of morbidity and mortality in patients with heart failure with reduced ejection fraction (HFrEF).
• •Our systematic review showed that aerobic training (AT) combined with resistance training increases CRF and muscle strength more than AT alone in patients with HFrEF.
• •The superior effect of combined training on CRF was retained when matched for total exercise session duration.
• •High-intensity interval training (HIIT) combined with resistance training also showed a greater increase in CRF than HIIT alone.
• •In HFrEF, adding resistance training may be an effective strategy to enhance CRF and muscle strength.

Abstract

Background

Heart failure (HF), characterized by low exercise tolerance, progressive functional decline, and reduced health-related quality of life (HRQoL), affects more than 64 million people worldwide. This study assessed the effects of aerobic training (AT) combined with muscle strength training (combined training) on these health measures in patients with HF across the spectrum of ejection fraction (EF).

Methods

A systematic search was conducted across MEDLINE, Embase, Cochrane Central Register of Controlled Trials, CINAHL, SPORTDiscus, Scopus, and gray literature sources. A meta-analysis compared the effects of combined training and AT alone on cardiorespiratory fitness (CRF), 6-min walk test (6MWT) distance, muscle strength, HRQoL, and cardiac function in patients with HF with reduced EF (HFrEF) and those with preserved EF (HFpEF). Pooled estimates were derived using random-effects models, and pre-specified subgroup analyses examined HF classification, exercise volume and type.

Results

Of 13,965 studies screened, 15 were included (n = 526, 17% females; HFrEF: n = 466, 89%; and HFpEF: n = 60, 11%). In HFrEF, combined training increased CRF (standardized mean difference (SMD) = 0.40, 95% confidence interval (95%CI): 0.10‒0.71, p = 0.01, small-to-medium effect), 6MWT distance (mean difference (MD) = 48.4 m, 95%CI: 35.6‒61.0 m, p < 0.001), and upper body muscle strength (MD = 8.3 kg, 95%CI: 3.2 ‒13.4 kg, p = 0.02) more than AT alone. When matched for exercise session duration, combined training increased CRF more than AT alone. High-intensity interval training (HIIT) combined with muscle strength training also showed a greater increase in CRF compared to HIIT alone. There were no differences in HRQoL or cardiac function.

Conclusion

In predominantly male patients with HFrEF, combined training yielded greater improvements in CRF, 6MWT distance, and upper body strength than AT alone. Replacing part of aerobic training with muscle strength training may be an effective strategy to further enhance CRF in HFrEF. Also, HIIT may be combined with muscle strength training to induced further increases in CRF. Further evidence is needed to clarify the effects of combined training in HFpEF.

Graphical abstract

1. Introduction

Heart failure (HF) affects more than 64 million people worldwide (1%–3% in the general adult population) and its prevalence is projected to increase due to an aging population, improved survival following ischemic heart disease, and decreased HF mortality resulting from the increased availability of effective evidence-based therapies.¹ HF is a multi-faceted and life-threatening condition characterized by low exercise tolerance, progressive functional decline, reduced health-related quality of life (HRQoL), and a high risk of hospitalizations, morbidity, and mortality.1, 2, 3, 4 Exercise-based cardiac rehabilitation (CR) is a Class I recommendation to improve functional fitness and HRQoL, and to reduce hospitalization,^2,5 constituting a key pillar of HF management alongside pharmacotherapy and medical device interventions.⁶

HF is mainly categorized into 2 main types based on left ventricular ejection fraction (LVEF): HF with reduced ejection fraction (HFrEF; symptomatic HF with EF ≤ 40%) and HF with preserved ejection fraction (HFpEF; symptomatic HF with EF ≥ 50%).⁷ Randomized controlled trials (RCTs) and systematic reviews have shown that aerobic training (AT) significantly improves cardiorespiratory fitness (CRF),7, 8, 9, 10, 11 6-min walk test (6MWT) distance,11, 12, 13, 14 and HRQoL¹⁵ in both HFrEF and HFpEF. Emerging evidence has also demonstrated superior benefits of high-intensity interval training (HIIT) on CRF and cardiac function (e.g., LVEF and left ventricular end diastolic diameter) in HFrEF^10,16,17 and CRF in HFpEF^8,9 when compared to moderate-intensity continuous training (MICT). Furthermore, muscle strength training alone has been shown to improve CRF,^18,19 6MWT distance,¹⁸ HRQoL,¹⁸ and cardiac function, such as LVEF and diastolic function,¹⁹ in patients with HF.

Based on available evidence highlighting the independent benefits of AT and muscle strength training, current guidelines^5,20 recommend that patients with HF, regardless of LVEF, engage in both AT and muscle strength training. However, despite the well-established benefits of each modality, the effects of combining AT and muscle strength training (hereafter referred to as combined training) compared to the routinely recommended AT alone²⁰ remain unclear in patients with HF. While a small number of systematic reviews and meta-analyses directly comparing combined training with AT alone in patients with HF showed no differences in changes in CRF, LVEF, or HRQoL,^16,21, 22, 23, 24 these meta-analyses lacked the assessment of exercise training effects across the LVEF spectrum,^21,22 comprehensive database searching,^16,23 and analyses considering exercise volume or type.^16,21, 22, 23, 24 Pathophysiology differs between HFrEF and HFpEF.⁵ Greater exercise volume may emphasize the effects of combined training^25,26 but may also compromise exercise adherence.²⁷ Additionally, HIIT has increasingly been prescribed to patients with HF. Thus, analyses considering the HF classification, exercise volume, and exercise type are needed to identify more effective approaches to improve exercise tolerance, functional fitness, and HRQoL in patients with HF.

The primary aim of this systematic review was to compare the effects of combined training and AT alone on predictors of cardiovascular mortality,28, 29, 30 including CRF measured through peak exercise testing and functional fitness (i.e., 6MWT distance and muscle strength) in patients with HF. The secondary aim was to compare the effects of combined training and AT alone on other important health indicators, including HRQoL and cardiac function. Comparisons were made while accounting for HF classifications (i.e., HFrEF or HFpEF), total exercise duration (i.e., equivalent or non-equivalent exercise session duration between combined and AT alone), and exercise type (i.e., HIIT or MICT). We hypothesized that combined training would be superior to AT alone in improving CRF and functional fitness.

2. Materials and methods

2.1. Study design

The protocol for this systematic review and meta-analysis was prospectively registered with the international Prospective Register of Systematic Reviews (PROSPERO) on December 18, 2024 (registration number: CRD42024625955). This study was conducted in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) 2020 ³¹ and implementing Prisma in Exercise, Rehabilitation, Sport medicine and SporTs science (PERSiST) guidance.³²

2.2. Search strategy

A peer-reviewed search strategy³³ was conducted on December 1, 2024 in MEDLINE, Embase, Cochrane Central Register of Controlled Trials, CINAHL, SPORTDiscus, and Scopus (Supplementary Tables 1–6). No limits to language or publication date were applied, but a search filter for RCTs was employed where applicable.^34,35 The main search concepts comprised terms related to muscle strength exercise and HF, and they were informed by previously conducted systematic searches.²⁵ For gray literature, we searched ProQuest Dissertations and Theses, Europe PubMed Central (PMC) for pre-prints published within the previous 2 years, and ClinicalTrials.gov and the World Health Organization’s International Clinical Trials Registry Platform. The search was updated in June 2025 (Supplementary Tables 7–10). Conference abstract results in Embase were limited by publication date to within 2 years of the searches, and a supplemental hand search was conducted in December 2024 (and updated in June 2025) on a short list of relevant conferences (Supplementary Table 11) for any content not already indexed in Embase.

2.3. Eligibility criteria

2.3.1. Participants

Studies including adult patients (≥18 years old) diagnosed with HFrEF or HFpEF were eligible. For HFrEF, only studies with mean EF < 40% without implantable cardiac devices were included. There was no exclusion based on the New York Heart Association (NYHA) classification, additional cardiovascular conditions, or medical conditions.

2.3.2. Interventions and comparator

Given high study heterogeneity in existing exercise trials, we aimed to include homogenous studies to elucidate the effects of combined vs. AT alone. We included RCTs assigning participants into combined training or AT alone with a duration of at least 4 weeks (i.e., a required period for physiological adaptations to occur in patients with cardiovascular disease^36,37). Exercise training could be combined with other interventions, such as nutritional counseling, smoking cessation, vocational counseling, psychological counseling, and social work counseling. However, due to their potential synergistic effects, studies with additional exercise interventions (e.g., inspiratory muscle training (IMT)) were excluded. Studies that prescribed inconsistent AT between combined and AT alone were also excluded.

AT was defined as any structured, purposeful activity involving large muscle groups in a continuous and rhythmic manner.³⁸ Muscle strength training was defined as any structured strength-developing exercise that encompasses free weights, machines, body weight, bands/tubing, or any other objects that require one to exert force against a resistance³⁸ using any major muscle group. Studies with any intensity were included. When possible, exercise intensity was classified according to the American College of Sports Medicine (ACSM) guidelines.³⁸

2.4. Outcomes

Eligible studies reported at least one of the following primary or secondary outcomes. Primary outcomes included: CRF (i.e., peak oxygen consumption (VO_2peak) or peak metabolic equivalent (MET)) and functional fitness (i.e., 6MWT distance and muscle strength). Secondary outcomes included: HRQoL assessed using validated outcomes and cardiac functions (e.g., EF, end diastolic volume, end systolic volume, left ventricular diastolic diameter, left ventricular systolic diameter, stroke volume, and cardiac output). We also explored additional outcomes including adherence as defined by the number of sessions attended by participants, dropouts and their reasons, and adverse events.

2.5. Selection process

Titles and abstracts of studies identified in our search strategies were uploaded into the Covidence Systematic Review Software (Veritas Health Innovation, Melbourne, VIC, Australia). Duplicate articles were automatically removed by the software. The title and abstract of each identified study were independently reviewed by 2 of the 4 authors (GK, TN, TK, and TT). Disagreements on study eligibility were resolved by discussion involving TT. Full texts of selected studies were uploaded to the Covidence Software. Each full text was independently reviewed by 2 of the 4 authors (GK, TN, TK, and TT) for eligibility. TT acted as an arbiter if consensus was not reached between the 2 primary reviewers.

2.6. Data extraction

For each study, 2 of the 3 authors (GK, TN, and TT) independently extracted: general study information (author, country, publication year), participants characteristics (sex, HF and NYHA classifications, LVEF), study design, sample size, characteristics of combined training and AT alone (exercise frequency, intensity, duration, type, and progression), adherence to the prescribed training, dropout rates, and the number and types of adverse events. All extracted data were verified for consistency by TT. Any inconsistencies were resolved by consensus. For continuous variables, the mean and standard deviation (SD) at baseline, follow-up, and changes from baseline were extracted. When not reported, mean changes were calculated by subtracting the pre-intervention value from the post-intervention value. The SD of change scores were calculated from pre-SD and post-SD values.³⁹ When 95% confidence interval (95%CI) or median and interquartile ranges (IQR) were reported, mean and SD were estimated using the previously described methods.^40,41 Multiple treatment groups from a single study were combined to create a single pairwise comparison.⁴² Data presented in figures⁴³ were extracted using WebPlotDigitizer.⁴⁴

2.7. Statistical analysis

Statistical analyses were performed with Review Manager Software (Version 9.15.2; RevMan Web, Cochrane Collaboration, Copenhagen, Denmark) using an inverse variance method. Because of heterogeneity in the methodology of the included studies, a random effects model was used for pooled analyses. Heterogeneity was assessed using the I² statistic with the estimates of <50% and ≥50% representing below moderate and substantial heterogeneity, respectively.³⁹ The Hartung-Knapp-Sidik-Jonkman method was used to calculate 95%CI based on the number of studies included and presence of heterogeneity;⁴⁵ however, for 6MWT, stroke volume, end diastolic volume, and left ventricular diastolic diameter, the Wald-type method was used due to the small number of studies (n = 2). Continuous outcomes were analyzed using weighted mean differences (MD) with a 95%CI or standardized mean differences (SMD) with 95%CI if different measurement scales were used. SMDs of 0.2, 0.5, and 0.8 were categorized as small, medium, and large effect sizes, respectively.⁴⁶ Subgroup analyses were performed to assess the effects of: (a) combined training compared to AT alone in HFrEF and HFpEF, (b) combined training without reducing AT volume and with reduced AT volume (i.e., to match exercise session duration) when compared with AT alone, and (c) the effects of combined training compared to AT alone for HIIT and MICT separately.

Due to a variety of methods of reporting HRQoL, SMD was assessed by pooling different instruments. Given that not all studies reported data to allow their inclusion in this quantitative meta-analysis, results reporting was supplemented by a vote-counting approach to categorize results of all trials as positive (combined better than AT alone, p < 0.05), negative (AT alone better than combined, p < 0.05), or neutral (no difference between combined and AT alone, p > 0.05).⁴⁷ Where more than 2 trials reported the same validated HRQoL measures and domains (e.g., Minnesota Living with HF Questionnaire (MLWHF)), continuous outcomes were pooled separately by each scale and reported as MD and 95%CI. The level of significance was set at p < 0.05.

2.8. Study risk of bias assessment

Study quality and reporting were assessed by the Tool for the assEssment of Study qualiTy and reporting in EXercise (TESTEX), a reliable tool specifically designed to facilitate a comprehensive review of exercise training trials.⁴⁸ TESTEX does not penalize design constraints inherent to exercise trials, such as blinding of participants, and includes criteria appropriate to exercise trials, such as reporting quantifiable measures of exercise volume, adherence, and compliance. The quality of the studies was classified as high (≥12 points), good (7–11 points), or low (≤6 points). Two of 3 reviewers (TT, TN, and GK) independently scored each domain. Agreement was reached by consensus involving TT.

2.9. Assessment of the certainty and strength of evidence

The certainty and strength of the evidence were assessed using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach.⁴⁹ Evidence was downgraded from high certainty when there were concerns regarding study limitations, inconsistency of results, indirectness of evidence, and imprecision. For CRF only, publication bias was assessed by visually inspecting funnel plots as fewer than 10 studies were included in the other meta-analyses.^50,51 One reviewer (TT) assessed the certainty of the evidence, and the second reviewer (TK) verified the assessment for accuracy. Any disagreement was resolved by consensus.

3. Results

3.1. Study selection

Our search identified 25,700 studies. The full text of 62 studies were screened; of those, 47 studies were excluded. Specific reasons for exclusion were: no indication of HF classification,⁵² mean EF > 40%,⁵³ EF not available,⁵⁴ included IMT,^55,56 inconsistent AT between the groups (i.e., Nordic walking vs. traditional CR;⁵⁷ MICT vs. HIIT;⁵⁸ addition of balance and coordination exercise to AT alone;⁵⁹ and addition of chair-based range-of-motion, stretching, and flexibility exercises to AT alone⁶⁰), a secondary analysis⁶¹ of another study⁶² with no additional data to extract, and the study provided questionable values (e.g., age = 8.54 ± 193.7, LVEF: 5.42 ± 433.6).⁶³ In 1 study that included 4 groups (combined training, combined training and IMT, AT, and AT and IMT),⁶⁴ combined training and AT alone groups without IMT were included. One study combined callisthenics with AT for the first 2 weeks of 12 weeks of exercise training.⁶⁴ This study was included as no additional exercise was performed for the remaining 10 weeks. As a result, 15 studies (n = 526, HFrEF: n = 466, 89% and HFpEF: n = 60, 11%) were included in this review (Fig. 1).

Fig. 1.

Preferred reporting items for systematic review and meta-analysis flow diagram with the number of included and excluded studies. CENTRAL = Cochrane Central Register of Controlled Trials; ICTRP = International Clinical Trials Registry Platform; PMC = PubMed Central.

3.2. Description of the included studies

Detailed descriptions of the included studies are summarized in Supplementary Table 12. A total of 15 (14 single-center^43,62,65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76 and 1 multi-center⁶⁴) RCTs published between 2000 and 2025 were included (Supplementary Table 12). Most studies included patients with HFrEF (n = 14).^43,62,64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75 Only 1 study in patients with HFpEF⁷⁶ met our inclusion criteria. In patients with HFrEF, LVEF ranged from 23% ± 4%⁴³ to 37.8% ± 13.6% in combined training and ranged from 22.9% ± 10.7%⁷¹ to 34.9% ± 10.5%⁷⁵ in AT alone.

Twelve studies were performed in Europe (n = 427 patients, 81%),^43,62,64, 65, 66, 67, 68^,72, 73, 74, 75, 76 of which 7 were conducted in Greece (164 patients, 31%).^{62,64,66,68,72,74,75} Of the remaining 3 studies, 1 was completed in Canada (28 patients, 5%),⁶⁹ 1 in the US (21 patients, 4%),⁷¹ and another in Iran (50 patients, 10%).⁷⁰ Four studies included exclusively males.^62,67,71,73 There were 57 females with HFrEF (12%) and 31 females with HFpEF (52%). Most studies (n = 13, 87%) provided supervised training sessions in hospitals or CR centers.^43,64, 65, 66, 67^,69, 70, 71, 72, 73, 74, 75, 76 The presence of supervision was unclear in 2 studies.^62,68

Seven studies used continuous aerobic exercise^43,64,65,67,69, 70, 71 and 8 used HIIT.^62,66,68,72, 73, 74, 75, 76 In combined training, 3 studies added muscle strength training to AT, resulting in a longer total exercise duration compared to AT alone.^69,71,76 The other studies reduced the duration of AT to add muscle strength training.^43,62,64, 65, 66, 67, 68^,70,72, 73, 74, 75 Nine out of 15 studies indicated that AT was performed before muscle strength training,^43,62,65, 66, 67^,70,71,73,75 and the remaining studies also suggested that AT was performed first (e.g., combined training included 20 min of AT plus 20 min of muscle strength training).^{64,68,69,72,74,76} Most studies used weight machines to train both upper and lower body.^43,55,62,65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76 Frequently performed muscle strength exercises included: chest press, shoulder press, biceps curl, triceps extension, latissimus pull, horizontal row, knee extension/flexion, and seated leg press. The duration of exercise training ranged from 7 weeks to 6 months (12 ± 4 weeks, mean ± SD), with most studies (10 studies, 67%) lasting 12 weeks.^59,62,66, 67, 68, 69^,72,74, 75, 76 Exercise training was performed twice a week in 1 study⁷⁶ and 3 times per week for all remaining studies.

3.3. Study quality and risk of bias and certainty and strength of evidence

Overall, the quality of the included RCTs was good (median TESTEX score of 9.0 points, IQR: 9.0–10.0 points; range: 6–14 points). Three studies (20%) were classified as high quality (i.e., ≥12 points),^65,70,76 11 (73%) as good (i.e., 7–11 points),^{43,62,64,66,67,69,}71, 72, 73, 74, 75 and 1 (7%) as low.⁶⁸ Of the 15 studies, 14 reported the eligibility criteria (93%),^43,62,64, 65, 66, 67, 68, 69, 70, 71, 72^,74, 75, 76 8 specified the method used to randomize participants (53%),^{64,66,68,70,72,}74, 75, 76 5 stated that they concealed allocation (33%),^{64,65,69,70,76} 14 reported similar participant characteristics at baseline between the groups (93%),^43,62,64, 65, 66, 67^,69, 70, 71, 72, 73, 74, 75, 76 and 8 unambiguously stated that an assessor was blinded to group allocation (53%).^43,62,64, 65, 66^,70,74,75 One study used clustered randomization, where the first 30 participants were assigned to combined and the next 30 to AT alone.⁶⁵

For study reporting, 13 studies reported adverse effects (87%)^43,62,64, 65, 66, 67^,70, 71, 72, 73, 74, 75, 76 and 9 reported exercise attendance (60%),^62,65,67,69,71, 72, 73, 74, 75 although 1 study did not provide quantifiable measures (i.e., excellent attendance).⁶⁵ Eight studies reported that >85% of participants completed the study (53%).^{43,62,65,67,71,73,74,76} Point measures, measures of variability, and between-group comparison were reported in all studies.^43,62,64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76 Only 3 studies (20%) clearly stated that they conducted an intention-to-treat analysis.^69,70,76 Exercise intensity was adjusted to maintain the same relative exercise intensity in 11 studies (73%),^43,62,64, 65, 66^,71, 72, 73, 74, 75, 76 and exercise volume was available in all studies.^43,62,64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76 Funnel plots for CRF and ejection fraction showed no publication bias (Supplementary Fig. 1).

3.4. Certainty and strength of evidence

The overall risk of bias in included studies was judged to be serious. Four of 15 studies had an attrition rate >15%,^64,69,72,75 2 had significant baseline differences,^73,75 and 3 did not blind assessors (Supplementary Table 13).^68,71,74 For CRF, 4 of 13 studies (31%) showed high attrition bias,^64,69,72,75 and the intention-to-treat analytical approaches were insufficiently described in 11 studies (85%).^{43,62,64‒66,68,69,72‒75} Thus, the overall certainty of evidence was downgraded by 1 level. The certainty of evidence for CRF was categorized as moderate. The certainty of evidence for 6MWT was downgraded by 2 levels and deemed to be low due to the combination of serious risk of bias and risk of imprecision. One of 2 studies showed a high dropout rate and neither study provided a clear indication of intention-to-treat analysis. The risk of imprecision was due to the small number of included studies. Both upper and lower body muscle strength were downgraded by 2 levels based on a serious risk of bias and imprecision (Supplementary Table 13).

3.5. CRF

Cardiorespiratory fitness was measured in 13 of the 15 studies.^43,62,64, 65, 66^,68, 69, 70^,72, 73, 74, 75, 76 Overall, CRF increased significantly more following combined training compared to AT alone (SMD = 0.44, 95%CI: 0.16‒0.71, p = 0.005, small-to-medium effect, I² = 30%, p(I²) = 0.08; Fig. 2A). In HFrEF, CRF increased significantly more following combined training (SMD = 0.40, 95%CI: 0.10‒0.71, p = 0.01, small-to-medium effect, I² = 33%, p(I²) = 0.08; Fig. 2A). Subgroup comparisons showed a greater increase in CRF following combined training when the AT duration was reduced to match the exercise session duration (SMD = 0.44, 95%CI: 0.13‒0.76, p = 0.01, small-to-medium effect, I² = 31%, p(I²) = 0.09, Fig. 2B). Sub-analyses by exercise type showed that HIIT combined with muscle strength training increased CRF more than HIIT alone (SMD = 0.68, 95%CI: 0.31‒1.05, p = 0.003, medium-to-large effect, I² = 0%, p(I²) = 0.20; Fig. 2C). In all sub-analyses, the study by Tzanis et al.⁶² displayed noticeable variance in the direction of effects. This study included the smallest number of participants (n = 6 in AT alone and n = 7 in combined), which may have contributed to greater variability and an increased risk of chance findings. When this study was excluded from the analysis, the overall I² was reduced to 22% (p = 0.25).

Fig. 2.

Changes in cardiorespiratory fitness following combined training and aerobic training alone by (A) heart failure classification, (B) exercise volume, and (C) exercise type. As the study by Delagardelle⁷³ did not report SD of pre and post measures, we used SD reported by Agapitou⁶⁸ that had the same sample size and similar CRF. ^a 95%CI calculated by Hartung-Knapp-Sidik-Jonkman method, ^bτ² calculated by Restricted Maximum-likelihood method. 95%CI = 95% confidence interval; HFpEF = heart failure with preserved ejection fraction; HFrEF = heart failure with reduced ejection fraction; HIIT = high-intensity interval training; IV = inverse variance; MICT = moderate-intensity continuous training.

3.6. 6MWT distance

Two studies compared the effects of combined training and AT alone on 6MWT distance in HFrEF.^64,67 There was a significantly greater increase in 6MWT distance following combined training compared to AT alone (MD = 48.35 m, 95%CI: 35.66‒61.04 m, p < 0.0001; I² = 0%, p(I²) = 0.64; Fig. 3). Both studies used MICT and reduced the AT volume to match training durations.

Fig. 3.

Changes in 6-min walk test distance in patients with heart failure with reduced ejection fraction following combined training and aerobic training alone. Both studies used moderate-intensity continuous training and reduced the aerobic training volume to match training durations. ^a 95%CI calculated by Hartung-Knapp-Sidik-Jonkman method, ^bτ² calculated by Restricted Maximum-likelihood method. 95%CI = 95% confidence interval; IV = inverse variance.

3.7. Muscle strength

After excluding 1 study with no indication of the muscles involved for the muscle strength assessment,⁷² 5 studies were included in the lower body muscle strength analysis^{62,65,68,69,71} and 3 for upper body muscle strength.^65,69,71 All these studies assessed muscle strength in patients with HFrEF. There was a significantly greater improvement in upper body muscle strength following combined training compared to AT alone (MD = 8.29 kg, 95%CI: 3.20‒13.38 kg, p = 0.02, I² = 0%, p(I²) = 0.45; Fig. 4A). No sub-analysis by exercise type was performed for upper body strength as all included studies used MICT. There were no differences in lower body muscle strength (MD = 5.73 kg, 95%CI: –0.62 to 12.07 kg, p = 0.07, I² = 57%, p(I²) = 0.06; Fig. 5).

Fig. 4.

Changes in upper body muscle strength in patients with heart failure with reduced ejection fraction following (A) combined training and aerobic training alone, and (B) subgroup analysis by exercise volume. ^a 95%CI calculated by Hartung-Knapp-Sidik-Jonkman method, ^bτ² calculated by Restricted Maximum-likelihood method. 95%CI = 95% confidence interval; IV = inverse variance.

Fig. 5.

Changes in lower body muscle strength in patients with heart failure with reduced ejection fraction following (A) combined training and aerobic training alone, and subgroup analyses by (B) exercise volume and (C) exercise type. ^a 95%CI calculated by Hartung-Knapp-Sidik-Jonkman method, ^bτ² calculated by Restricted Maximum-likelihood method. 95%CI = 95% confidence interval; HIIT = high-intensity interval training; IV = inverse variance; MICT = moderate-intensity continuous training.

3.8. Health-related quality of life

Five trials were included in the pooled analysis.^{43,64,67,69,76} There was no difference between combined training and AT alone in HRQoL (Fig. 6A and 6B). No sub-analysis by exercise type was performed as only 1 study used HIIT. There was no difference between combined training and AT alone in HRQoL as measured by MLWHF (Fig. 6C). Across the 6 trials, 2 reported statistically superior effects of combined training compared to AT alone,^65,67 while the remaining 4 showed no difference (Supplementary Table 14).^43,64,69,76

Fig. 6.

Changes in health-related quality of life following combined training and aerobic training alone: (A) pooled analysis by HF classification including different HRQoL measures, (B) pooled analysis by exercise volume including different HRQoL measures, and (C) HRQoL measured by Minnesota Living with Heart Failure Questionnaire (MLWHF) in patients with heart failure with reduced ejection fraction. ^a 95%CI calculated by Hartung-Knapp-Sidik-Jonkman method, ^bτ² calculated by Restricted Maximum-likelihood method. 95%CI = 95% confidence interval; HF = heart failure; HFpEF = heart failure with preserved ejection fraction; HFrEF = heart failure with reduced ejection fraction; HRQoL = health-related quality of life; IV = inverse variance.

3.9. Cardiac function

Changes in EF, end systolic volume, end diastolic volume, and left ventricular diastolic diameter in HFrEF are summarized in Supplementary Figs. 2–5. There were no differences between combined training and AT alone. No sub-analyses were performed due to the small number of included studies.

3.10. Adherence, dropout, and adverse events

Of the 15 studies included, 9 reported adherence (60%).^62,65,67,69,71, 72, 73, 74^,76 Among 7 studies reporting adherence by groups, 1 study described adherence as excellent for both groups,⁶⁵ 4 had the adherence of 100% as missed sessions were added to the end of the program,^62,72, 73, 74 and the others reported >90%⁶⁷ and >83%⁷⁶ for both groups. The group-specific dropout rates were reported in 7 studies.^{43,64,65,70,72,73,76} Two studies had no dropouts in either group.^43,73 The dropout rates were the same in 1 study,⁷⁰ higher in AT in 2 studies (5.0% vs. 2.5%⁷⁶ and 24% vs. 10%⁷²) and lower in AT in 2 studies (0% vs. 7%⁶⁵ and 9% vs. 18%⁶⁴).

Adverse events of the AT alone included: hyperglycemia (n = 2),⁶⁹ gout (n = 1),⁶⁹ prolonged severe fatigue (n = 1),⁶⁹ severe hypotension (n = 1),⁷⁰ severe knee pain (n = 1),⁷⁰ atrial fibrillation (n = 1),⁴³ severe hypotension (n = 1),⁷³ and worsening HF (n = 3).⁶⁴ Adverse events of the combined training included: ischialgia, (n = 1),⁶⁵ atrial fibrillation (n = 1),⁶⁹ undiagnosed chest pain (n = 1),⁶⁹ angina during walking (n = 2),⁷⁰ knee pain (n = 2),⁷⁰ depression (n = 1),⁶⁴ arrhythmia (n = 1),⁶⁴ pulmonary infection (n = 1),⁶⁴ and death due to respiratory infection (n = 1).⁶⁴

4. Discussion

4.1. Overall effects of combined training

We found that combined training increased CRF, 6MWT distance, and upper body muscle strength more than AT alone in patients with HFrEF. The superior effects of combined training on CRF and 6MWT were retained when exercise session duration was matched between the 2 exercise modalities, indicating that combined training has a greater impact on these measures independent of total exercise duration. Our findings also showed that, when combined with muscle strength training, HIIT further increased CRF. No differences were found between combined and AT alone in lower body muscle strength or HRQoL. Only 1 study met our inclusion criteria for patients with HFpEF,⁷⁶ highlighting the need for further research to clarify the effects of combined training on HFpEF.

4.2. CRF

It has been proposed that AT and muscle strength training have an additive or synergistic benefit on CRF.⁷⁷ However, this was not supported by previous reviews.^16,21 The smaller number of studies included in previous reviews compared to ours (n = 4¹⁶ and 11²¹ vs. n = 15), combined with our inclusion strategy to reduce heterogeneity may explain the discrepancy. Meeting either AT or muscle strength training recommendations is associated with an approximately 20%–35% reduced risk of cardiovascular disease mortality, whereas meeting both guidelines is associated with a 50% reduced risk of cardiovascular disease mortality.⁷⁸ Because higher or improved CRF is associated with lower cardiovascular disease mortality,²⁹ engaging in both AT and muscle strength training may reduce the risk by emphasizing the increase in CRF.

Considering that the total exercise time could emphasize the effects of combined training (i.e., combined training group performing a greater total duration of exercise and thus more beneficial) as observed in patients with coronary artery disease,^25,26 there was a need to assess the effects of combined training independent of total exercise time. Our results highlight that, when prescribing exercise for a fixed time frame, allocating time to both AT and muscle strength training may be a more effective strategy for improving CRF in patients with HFrEF. Additionally, because HIIT may improve CRF more effectively^16,79 while requiring less time⁸⁰ than MICT in patients with HFrEF, combining HIIT with muscle strength training may offer a time-efficient approach to improve CRF. Our finding showing that HIIT combined with muscle strength training improves CRF more than HIIT alone aligns with that of a previous systematic review.⁸¹

While improvements in cardiac function support an increase in CRF, our results showed no differences in cardiac function between combined training and AT alone. Because patients with HFrEF exhibit multiple skeletal muscle abnormalities that impair oxygen uptake and utilization, including reduced lean mass, oxidative fiber percentage, capillarity, oxidative enzyme capacity, and mitochondrial volume, it has been suggested that targeting peripheral, non-cardiac pathways may improve exercise tolerance.^82,83 The greater increase in CRF observed with combined training may reflect a more prominent increase in skeletal muscle oxygen utilization in addition to a non-attenuated increase in cardiac function compared to AT alone. Our finding is in line with a previous systematic review demonstrating that muscle strength training increases CRF without changing LVEF.⁸⁴

Among our included studies, 14 out of 15 (93%) implemented 3 exercise sessions per week, with exercise session durations ranging from 30 to 60 min. For the combined training group in which AT duration was reduced, the mean AT duration was 23.5 ± 7.9 min per session compared to 43.0 ± 8.5 min per session in the AT alone group. Thus, despite the nearly 50% lower aerobic exercise volume, combined training resulted in greater improvements in CRF than AT alone. It remains unclear why adding muscle strength training to AT (without reducing AT duration) did not yield a greater increase in CRF compared to AT alone. However, because only 2 studies (1 in HFpEF) added muscle strength exercise to AT (n = 84), whereas 11 studies reduced the volume of AT to incorporate muscle strength training (n = 396), the absence of an observed benefit in the former subgroup may reflect insufficient sample size rather than a lack of effect.

4.3. Functional fitness

Consistent with the greater improvements in CRF, 6MWT distance increased more following combined training compared to AT alone in patients with HFrEF. Given that both included studies reduced AT volume to match overall exercise session duration, this finding supports the importance of incorporating muscle strength training even when it requires reducing AT volume. However, due to the limited number of included studies and low certainty of evidence, further research is needed to clarify the effects of combined training on 6MWT distance.

Preservation of skeletal muscle strength may be clinically important for counteracting functional decline associated with catabolic states and cachexia common in HF.⁸⁵ Our findings of greater increases in upper body muscle strength following combined training compared to AT alone extend the evidence from a previous meta-analysis showing that muscle strength training significantly improves upper extremity muscle strength compared to control.¹⁸ Instrumental activities of daily living, such as household tasks, shopping, and carrying objects, rely heavily on upper body muscle strength and endurance.^86,87 However, patients diagnosed with HF demonstrate a compromised upper body functional capacity to perform necessary activities of daily living.^86,88 Additionally, upper body muscle mass and strength measured via mid-upper arm circumference⁸⁹ and handgrip strength,⁹⁰ respectively, are inversely associated with the risk of mortality in patients with HF. Thus, a greater increase in upper body muscle strength following combined training has important clinical implications for reducing risk of physical disability⁸⁸ and mortality in patients with HF.

In contrast to upper body muscle strength, we found no difference in lower body muscle strength. Four of the 5 included studies used stationary bikes for AT,^62,65,68,69 which primarily engaged the lower limbs. Consequently, participants in the AT alone group may have improved lower body muscle strength similar to those in the combined training group. Similarly, 2 of the 3 studies assessing upper body muscle strength used stationary bikes for AT.^65,69 Because stationary bikes minimally engage the upper body, combined training that included targeted upper body strength exercises may have enhanced upper body muscle strength more than AT alone. Considering that most studies used the same percentage of 1 repetition maximum (%1RM) and number of repetitions for upper and lower body muscle strength training, it is possible that a higher muscle strength exercise intensity is required to improve lower body muscle strength more than AT alone.

Most of the included studies used an aerobic-to-muscle strength exercise ratio of 1:1 and prescribed aerobic exercise before muscle strength exercise. Evidence from animal models suggests that the last bout of exercise dictates the molecular responses that induce muscle adaptations.⁹¹ In humans, studies have demonstrated either greater muscle hypertrophy when aerobic exercise was performed prior to muscle strength exercise compared to muscle strength training alone,⁹² or they have shown no negative effects of concurrent aerobic and muscle strength training on muscle hypertrophy or maximal strength development⁹³ regardless of the order of aerobic and muscle strength exercise.⁹⁴ Thus, in the included studies, we speculate that combined training did not compromise the anticipated muscle adaptations to muscle strength training and emphasized an increase in cardiorespiratory fitness compared to aerobic training alone.

4.4. HRQoL

While combined training led to significantly greater improvements in CRF, 6MWT distance, and upper body muscle strength compared to AT alone, no significant difference was observed in changes in HRQoL. Notably, in patients with HFrEF, 2 of 6 studies reported superior effects of combined training on HRQoL compared to AT alone. However, this result should be interpreted with caution as vote-counting does not account for sample size or the magnitude of the effect. Because of the heterogeneity in HRQoL assessment methods and the limited number of studies, further research is needed to clarify the effects of combined training on HRQoL.

4.5. Adherence, dropout, and adverse events

In this systematic review, we found that overall adherence to combined training was high or comparable to that of AT alone. Similarly, dropout rates in the combined training group were comparable to those in AT alone, and no notable differences in the risk of adverse events were observed. Considering that 13 of 15 included studies provided supervised exercise sessions, the addition of muscle strength training to AT does not negatively affect adherence or attrition if exercise sessions are supervised and the training duration is approximately 12 weeks. Additionally, our results highlight that, under appropriate supervision, patients with HFrEF can safely engage in muscle strength training using weight machines. These findings support the recent American Heart Association scientific statement indicating that combined training is safe for clinically stable patients with HF.⁸³ While the availability of weight machines may differ among cardiac rehabilitation centers, muscle strength training requiring minimal (e.g., resistance bands) to no equipment (bodyweight exercise) can be as effective as weight machines and may reduce complexity when developing more accessible muscle strength training programs.⁸³

4.6. Limitations

Our study has limitations. First, we identified only 1 study in HFpEF meeting our inclusion criteria. The prevalence of HFpEF is steadily increasing and may become the most common form of HF in the future.¹ However, guidelines highlighted lack of studies in HFpEF and the need for appropriately powered RCTs to assess the efficacy and safety of CR in HFpEF.⁵ In patients with HFpEF, a recent multi-center RCT showed that combined training improves CRF (MD = 1.3 mL/kg/min, 95%CI: 0.4‒2.1 mL/kg/min) compared to usual care.⁹⁵ Because reduced CRF is a hallmark clinical feature of HFpEF,⁹⁶ combined training may play a valuable role in its management. Further exercise studies in HFpEF are needed to address the imbalance in available evidence on the effects of exercise in HFrEF and HFpEF. Second, the smaller representation of females with HFrEF (12%) limits the generalizability of our findings to female patients. Because HFrEF is more prevalent in males and HFpEF in females,⁹⁷ the low proportion of females in HFrEF may reflect the smaller pool of eligible female patients. Alternatively, it may also represent the commonly reported challenge of enrolling females in CR programs.⁹⁸ Females with HFrEF exhibit lower exercise capacity than males, largely due to attenuated increase in peak stroke volume and cardiac output.⁹⁹ While some evidence highlights greater improvements in CRF and 6MWT distance in females with HF in response to exercise-based cardiac rehabilitation compared to males,¹² it remains unclear whether adding muscle strength training that primarily targets peripheral adaptation⁸³ leads to similar benefits in females. Future trials should explicitly aim for randomization stratified by sex to determine whether the CRF adaptations to combined training differ between males and females with HFrEF. Third, most studies (>70%) were conducted in Europe. This may limit the generalizability of the findings to patients in other geographic regions and ethnicities. Although the level of detail in muscle strength training recommendations differs across global geographic regions, all European nations provide detailed guidance.¹⁰⁰ This may have facilitated the inclusion of muscle strength training in European studies. In contrast, the lack of established, detailed recommendations in some countries may have posed a barrier to prescribing muscle strength training for patients with HF. Fourth, while exercise duration was taken into consideration, this did not necessarily match the exercise volume (i.e., caloric expenditure) between combined training and AT alone. Nevertheless, given that lack of time is an often-cited barrier to participating in exercise,¹⁰¹ our results underscore a more efficient use of time to improve CRF and muscle strength. Lastly, our findings are specific to skeletal muscle training and do not encompass other types of muscle strength exercise, such as respiratory muscle interventions.

5. Conclusion

This systematic review and meta-analysis showed that combined training elicits greater improvements in CRF, 6MWT distance, and upper body muscle strength compared to AT alone in predominantly male patients with HFrEF. The improvements in CRF and 6MWT distance were greater in combined training even when matched for total exercise duration. Based on the included studies, when matched for total exercise session duration, allocating half of each exercise session to muscle strength exercise may be more beneficial than performing AT alone. Considering the absence of notable differences in adherence or adverse events, our findings support the implementation of combined training as a preferable exercise modality for patients with HFrEF. The small representativeness of HFpEF limits the ability to draw meaningful conclusions with respect to HFpEF.

Authors’ contributions

TT drafted the manuscript, developed the selection and data extraction criteria, developed the search strategy, screened studies for inclusion, extracted information on adherence and adverse events, completed the risk of bias assessments, and is the guarantor; TK screened studies for inclusion, extracted information on adherence and adverse events, completed the risk of bias assessments, and critically reviewed the manuscript; TN screened studies for inclusion and critically reviewed the manuscript; GK screened studies for inclusion, extracted information on adherence and adverse events, and critically reviewed the manuscript; RST, JLR, and KK contributed to the development of the selection and data extraction criteria and critically reviewed the manuscript. All authors have read and approved the final version of the manuscript, and agree with the order of presentation of the authors.

Declaration of competing interest

The authors declare that they have no competing interests.