Effect of resistance exercise on body composition, muscle strength and cardiometabolic health during dietary weight loss in people living with overweight or obesity: a systematic review and meta-analysis

Abstract

Background

The prevalence of obesity has tripled over the past 35 years. Although caloric restriction reduces body fat, lean tissue is also lost. Resistance exercise may mitigate these effects. This review assesses the effects of resistance exercise on body composition and cardiometabolic health in adults with overweight or obesity undergoing dietary weight loss.

Methods

A search was performed in PubMed, Embase, CINAHL, Scopus, Web of Science and CENTRAL. Systematic searches yielded N=6934 studies of which n=25 were relevant for inclusion after screening for eligibility. We included randomised controlled trials of adults (18–65 years) (body mass index ≥25 kg/m²) comparing dietary weight-loss interventions with resistance exercise to diet-only weight-loss interventions. The Cochrane Collaboration’s risk of bias and GRADE (Grades of Recommendations Assessment, Development and Evaluation) assessment tools were used. A meta-analysis was performed, including those studies that compared dietary weight-loss plus resistance exercise interventions to diet-only weight-loss interventions.

Results

Overall, resistance exercise during diet-induced weight loss had no effect on body mass (mean difference between groups: −0.32 kg, p=0.35), with weight loss ranging from −2.5 kg to −20.9 kg in diet plus resistance exercise groups and from −0.7 kg to −20.4 kg in diet-only groups. However, resistance exercise protected against the loss of fat-free mass (between-group standardised mean difference (SMD): 0.40, p=0.0003, moderate certainty) and increased loss of fat mass (SMD: −0.36, p<0.00001, high certainty). Muscular strength was also significantly greater (SMD 2.36, p=0.00001) following the inclusion of resistance exercise (low certainty). No effects of resistance were seen in any of the other cardiometabolic markers studied.

Conclusion

In people living with obesity and overweight, the addition of resistance exercise to dietary restriction may enhance its beneficial effects. Current evidence, therefore, supports the implementation of resistance exercise during weight loss to attenuate the loss of fat-free mass (moderate certainty), increase loss of fat mass (high certainty) and improve muscle strength (low certainty).

PROSPERO registration number

CRD 42021266482.

WHAT IS ALREADY KNOWN

• Weight loss is a common goal for many people living with overweight or obesity.

• Dietary interventions are a mainstay for the treatment of obesity.

• Lean tissue, including muscle tissue, accounts for around 25% of the mass lost during dietary weight loss.

• Whether adding resistance exercise into dietary weight loss interventions to mitigate this loss of muscle remains unclear.

WHAT ARE THE NEW FINDINGS

• Although it has no effect on the change in body mass, resistance exercise during diet-induced weight loss protects against the loss of fat-free mass and increases the loss of fat mass.

• Muscular strength was also significantly greater with the inclusion of resistance exercise.

• There is a need for further research to investigate the cardiometabolic effects of resistance exercise training during weight loss, and to develop and evaluate pragmatic resistance exercise interventions.

Introduction

Obesity is a major public health problem with serious consequences for morbidity, mortality and healthcare costs.¹ Obesity increases the risk of type 2 diabetes, hypertension, coronary heart disease, dyslipidaemia and certain cancers.² While the global prevalence continues to rise, the burden of obesity is greatest in adults aged 45–59 and in women.^{3 4} Dietary interventions are a mainstay for the treatment of obesity, and a recent systematic review and meta-analysis has shown they result in significant weight loss (around 4–5 kg), that is at least partially sustained up to 12 months.⁵ This level of weight loss is associated with improvements in cardiovascular risk factors such as blood pressure, low density lipoprotein (LDL) and high-density lipoprotein (HDL) cholesterol and glycaemic control.⁵ Greater weight loss (∼10 kg) has been shown to result in full remission of type 2 diabetes in almost half of participants.⁶

One of the less desirable consequences of dietary interventions is that ∼20–30% of weight lost is fat-free mass, which includes the loss of muscle tissue.⁷,⁹ This is a concern as skeletal muscle has both functional and metabolic roles,^{10 11} and low muscle mass/strength is a contributing factor to cardiometabolic and other obesity-related diseases¹² and is associated with higher mortality and morbidity.⁷ Incorporating strategies to maintain fat-free/muscle mass and muscle strength may improve the beneficial effects of dietary weight loss interventions.

Resistance exercise is the most effective method to maintain or increase muscle mass and strength. It has also been shown to improve blood lipids, reduce blood pressure and glycaemic control, increase cardiorespiratory fitness^{13 14} and improve muscle strength, power and endurance.¹⁵ Individual randomised controlled trials have shown that resistance exercise can attenuate lean mass loss during dietary weight loss, with one systematic review having examined the effects of resistance exercise training during weight loss.¹⁶ This review included children and adolescents, did not specifically isolate resistance exercise effects nor comprehensively examine body composition, cardiometabolic and muscle strength outcomes. To our knowledge, no prior systematic review with meta-analysis has specifically examined resistance exercise during dietary weight loss in people living with overweight or obesity across a broad range of outcomes. Therefore, the aim of the current systematic review was to examine the impact of resistance exercise on body weight/composition, including lean mass loss, muscle strength and markers of cardiometabolic health, in adults living with overweight or obesity undergoing dietary weight loss interventions.

Method

A systematic review and meta-analysis of RCTs was conducted and reported according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2021 updated guidelines¹⁷ (online supplemental table 1). The study was registered with PROSPERO (registration ID: CRD 42021266482). One deviation from our registered PROSPERO protocol should be acknowledged. Muscle strength was included in our PROSPERO protocol under additional outcomes as ‘muscle function’, but prior to beginning analysis was elevated to a main outcome for inclusion in the meta-analysis rather than being treated as an exploratory additional outcome.

Review questions

The review was designed to address the following questions with the PICOS detailed in table 1.

Table 1. PICOS.

Review questions	Inclusion criteria	Exclusion criteria
Population (P)	Adults (18–65 years of age) who are overweight/obese, overweight (BMI 25–29.9 kg/m2) or obese (BMI ≥30). For the age criteria, we used the mean age of the study sample±1 SD for inclusion.	Anyone <18 years or >65 years with obesity, overweight; any population without obesity or overweight
Intervention (I)	Dietary weight loss intervention (with no minimum caloric deficit required), including caloric restriction, meal replacements and VLCDs + all resistance exercise interventions (free weights, machines, resistance bands, body weight, gym, lab or home-based, mHealth/digitally delivered interventions).	No weight loss intervention
Comparator (C)	Dietary weight loss intervention + Any comparator including no intervention or non-resistance exercise interventions.	Resistance exercise in all comparator arms
Outcomes (O)	Empirical evidence of measured changes in cardiometabolic health, body weight/composition and muscle strength due to any resistance exercises.	Other outcomes
Study design (S)	Experimental studies: RCTs,	Systematic reviews and meta-analyses, quasi-experimental designs, observational studies: cross-sectional study; Single-case studies, case control studies, discussion articles, non-randomised studies, documents for reviews, cohort studies
Other	Published peer-reviewed articles in scientific journals, in an English language, human subjects.	–

BMI, body mass index; RCTs, randomised controlled trials; VLCDs, very low calorie diets.

1. Does the addition of resistance exercise to a dietary weight loss intervention impact changes in body weight/composition or markers of cardiometabolic health in people living with overweight or obesity compared with dietary weight loss interventions?

2. Does the addition of resistance exercise to a dietary weight loss intervention affect changes in physical function and strength in people living with overweight or obesity compared with dietary weight loss interventions?

Search strategy and study selection

Systematic searches were conducted in the following databases: PubMed, Embase, CINAHL, Scopus, Web of Science and the Cochrane Central Register of Controlled Trials. First, keyword and categorical searches were performed for (1) obesity or overweight, (2) weight loss, (3) resistance exercise or resistance training, (4) body weight or body composition or metabolic syndrome or muscle function (online supplemental table 2 Keywords and search terms). The categories were then combined using ‘and’. We restricted our search to humans and papers published in English, with no restriction on the publication period. Search dates were from July 2021 to September 2021 and searches were re-run before the final analysis from November 2022 to January 2023. A screening process was carried out by two independent reviewers (AB and AD) using Rayyan software,¹⁸ with a third reviewer (SG) consulted to reach agreement when required.

Data extraction and risk of bias assessment

Data were independently extracted by AB and AD into a predesigned data extraction form. Extracted information included bibliometric data (study title, funding), study characteristics (sample size, identified limitations of the study), participants (age, sex, body mass index), intervention characteristics (type of exercise, frequency, intensity, dietary weight loss strategy), control characteristics (dietary weight loss strategy) and outcomes (body weight/composition—body mass, fat mass, fat-free mass and muscle mass; cardiometabolic health—cardiorespiratory fitness, lipid profile, blood pressure and glycaemic control; and muscle strength and function). Where required, we contacted corresponding authors to request additional data.

Risk of bias was independently assessed by AB and AD using the Cochrane risk of bias tool.¹⁹,²² Each study was classified as high, low or unclear risk of bias based on the following five bias domains: (selection, performance, detection, attrition and reporting). The results were entered into Review Manager (RevMan) software V.5.4.^{1 23} Any disagreement during the process was resolved through discussion with SG.

In addition, the Grades of Recommendations Assessment, Development and Evaluation (GRADE) assessment (with GRADE PRO software (https://gdt.gradepro.org)) was used to assess the quality of evidence for outcomes reported.^{22 24} Because all included studies were RCTs, their GRADE scores started high, but were downgraded due to limitations regarding risk of bias, inconsistency, indirectness, imprecision or publication bias.^{20 22}

Data synthesis

Given the potential benefits of resistance exercise in improving body composition, cardiometabolic health and physical function in people living with obesity or overweight, we aimed to specifically investigate the impact of adding resistance exercise to dietary weight loss interventions on body weight/composition (body mass, fat mass, fat-free mass and muscle mass), muscle strength and function and cardiometabolic health (cardiorespiratory fitness, lipid profile, blood pressure and glycaemic control). Therefore, we chose to include only studies that compared resistance exercise plus dietary weight loss interventions to dietary weight loss-only interventions in our meta-analysis, to allow for a more focused analysis of the effects of resistance exercise on these outcomes.²⁵

For this review, eligible dietary interventions were those specifically designed for weight loss, including caloric restriction (with no minimum energy deficit required), meal replacements and very low calorie diets (VLCDs). Resistance exercise interventions involving free weights, machines, resistance bands, body weight, gym, lab or home-based, mHealth/digitally delivered interventions were included. This allowed us to examine the specific effects of resistance exercise training during dietary weight loss interventions.

Meta-analyses were conducted in order to compare the effectiveness of intervention (resistance exercise plus diet and diet only) on these outcomes in Review Manager V.5.4.²³ Subgroup analysis was carried out, stratifying studies by the duration of intervention (short duration (≤ 5 months) or long duration (≥6 months)) for the primary body composition variables only, due to insufficient number of studies for other variables. For studies reporting outcomes at multiple time points, only data from the final assessment were included in the meta-analysis to avoid statistical dependency issues. Based on a random-effects analysis, the standardised mean difference (SMD) was calculated for fat mass, fat-free mass, muscle strength and insulin sensitivity and mean difference (MD) for body weight and cardiometabolic health outcome measures between intervention groups. Mean changes (final—baseline) were used, as well as SDs. A request for data was made to the authors where data were not available. Missing SDs were calculated and median to mean conversions conducted using established methods.²⁶

Results

As shown in figure 1, the initial search yielded 6934 unique results. After title, abstract and full text screening, 25 RCT studies met the inclusion criteria and were included in meta-analysis.

Figure 1. PRISMA flow diagram showing the process of study selection. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses; RT, resistance training.

Study characteristics

The characteristics of the 25 eligible studies are summarised in online supplemental table 3. In total, they included 1608 participants.²⁷,⁵¹ Two studies included people with obesity and type 2 diabetes,³⁷ cardiovascular disease and/or metabolic syndrome.³⁰ One study included only male students,⁵⁰ 15 studies had only female participants (with 9/15 focusing on the premenopausal or postmenopausal period)2729 32 33 35 36 38 39 42,^{46 48 51} and the remainder included both sexes.^{28 30 31 34 37 40 41 47 49} Eight studies had two experimental groups testing diet and resistance exercise against diet alone.^{27 28 33 34 36 37 45 48} 10 studies had three experimental groups, 8 of which tested diet and resistance exercise against diet and aerobic exercise or against diet alone,^{3032 39},^{42 44 50} and the remaining 2 limited comparison to diet and resistance exercise against diet alone or against a control group.^{38 43} Across the remaining seven studies, four to five experimental groups were included: resistance exercise+weight loss diet, aerobic exercise+weight loss diet, resistance plus aerobic exercise+weight loss diet, weight loss diet alone or control.^{29 31 35 46 47 49 51} 18 studies were of interventions that were delivered for a shorter time period (2 months–5 months).2829 31 32 34,^{36 38 40} In seven studies, the interventions lasted for at least 6 months (6 months–18 months), with some having follow-up of 3 years.^{27 30 33 37 39 47 49}

Resistance exercise intervention characteristics

Most included (n=23) studies employed supervised resistance exercise sessions using traditional weight training machines or free weights.²⁷,⁴⁴⁴⁷ Resistance exercise was most often performed three times per week, but some studies had two⁴³ or four^{30 35} sessions weekly. Sessions lasted between 30 and 60 min on average, involving 8–10 exercises targeting major muscle groups and generally including leg extension, leg press, chest press, shoulder press, lateral pull-down and arm exercises to target the major muscle groups.

Dietary weight loss intervention characteristics

Included studies employed various dietary weight loss approaches, with caloric restriction being the common element. Most studies (n=15) used a moderate caloric restriction of 500–1200 kcal/day below estimated requirements,2829 31 33 37 38 41 43,^{45 47} while others (n=10) implemented VLCDs (≤800–900 kcal/day).^{2730 32 34},36 39 40 42 46 Macronutrient distribution generally followed standard guidelines, with most diets providing 50–60% of calories from carbohydrates, 20–30% from fat and 15–25% from protein. Several studies (n=8) used liquid meal replacements, either exclusively or partially, particularly those implementing very low calorie interventions.²⁷³⁴,36 39 40 42 48 Dietary adherence was monitored through various methods including weekly weigh-ins, food diaries and regular meetings with dietitians. Most dietary interventions (n=18) provided nutritional counselling, either individually or in group sessions, focusing on topics such as healthy eating behaviours, portion control and long-term dietary changes.²⁷,3133 37 38 41 43

Adherence and supervision characteristics

Adherence to resistance training interventions was inconsistently reported across studies. Specifically, 14 out of 25 studies provided clear information regarding adherence to resistance training, with most reporting high levels of compliance,²⁷,2931 33 34 37 38 41 43 particularly in short duration interventions and trials where training sessions were fully supervised. Adherence rates commonly exceeded 85%, with several studies requiring participants to attend a minimum percentage of sessions (≥90%) to be included in the final analyses. Dietary adherence was not generally reported quantitatively but was described in 22 studies.²⁷,³¹³³ Methods used to assess dietary adherence included attendance at diet education sessions,^{28 29 33 37 41 51} self-reported food diaries,^{27 46 50 51} weekly weigh-ins^{34 40 45 51} and in some cases, the provision of preprepared or formula diets to ensure precise caloric control.³⁴,³⁶⁴⁰ Overall, adherence to both intervention components was generally higher in studies with shorter durations, supervised protocols or structured monitoring procedures.

Study quality and risk of bias

As shown in figure 2, all studies had low risk of selection bias, as they provided information on the method of random sequence generation. Allocation concealment was reported in only one study, with a low risk of bias for this domain.⁴¹ All studies were at high risk of performance bias, due to the inability to blind investigators/participants to exercise participation. One study had low detection bias, meaning that there was a possibility of bias in the assessment of the outcomes.³⁷ In all other studies, the risk of bias in blinding of outcome assessment was unclear. Most longer duration intervention studies had a high risk of attrition bias, with high dropout rates (>25%), not stating how they handled missing data from these dropouts and lacking intention-to-treat analysis.^{30 33 39 47 49} All studies had a low risk of reporting and other bias. Since many of the included studies did not report concealment of allocations or blinding of assessment of outcomes, there is a lack of clarity regarding their potential bias. For the GRADE assessment, the overall certainty of the evidence for each outcome shown in online supplemental figure 1 ranged from low to high. The main reasons for downgrading were inconsistency, because of heterogeneity, and imprecision, because of the small number of trials evaluating resistance exercise during weight loss. Fat mass, maximum oxygen consumption (VO_2max) and triglycerides were rated as high certainty. Body weight, fat-free mass, glucose, insulin, HDL cholesterol and systolic blood pressure were rated as moderate certainty. However, total cholesterol, LDL cholesterol, diastolic blood pressure and muscle strength were rated as low certainty.

Figure 2. Study quality assessment using Cochrane Risk of Bias tool. Green (+) = low risk of bias; yellow (?) = unclear risk of bias; red (-) = high risk of bias.

Effect of resistance exercise intervention on body weight/composition

25 studies reported change in body weight, with weight losses ranging from −2.5 kg to −20.9 kg in diet plus resistance exercise groups and −0.7 kg to −20.4 kg in diet-only groups. As shown in figure 3, the meta-analysis revealed no significant differences in change in body weight between diet plus resistance exercise and diet only groups (MD: −0.32 kg, 95% CI −1.00kg to 0.35 kg; p=0.35, I²=60%). Similar findings were found when analysing both short and long duration interventions. For short duration interventions (≤5 months), the MD was −0.07 kg, (95% CI −0.86 kg to 0.73 kg; p=0.87, I²=55%) and for long duration interventions (≥6 months), the MD was −0.87 kg, (95% CI −2.09 kg to 0.35 kg; p=0.16, I²=63%) (figure 3).

Figure 3. (A) Effect of dietary weight loss plus resistance exercise versus diet-only weight loss on body weight in people living with overweight or obesity. (B) Effect of dietary weight loss plus resistance exercise versus diet-only weight loss on body weight in people living with overweight or obesity in studies of duration ≤5 months. (C) Effect of dietary weight loss plus resistance exercise versus diet-only weight loss on body weight in people living with overweight or obesity in studies of duration ≥6 months. RT, resistance training.

Figure 4, with 22 studies reported, shows a significantly greater reduction in fat mass for diet plus resistance exercise groups, compared with diet-only groups (SMD: −0.36, 95% CI −0.49 to −0.23; p<0.00001, I²=9%). Similar improvements were demonstrated in both short and long duration interventions. Short duration interventions (≤5 months) showed an SMD of −0.33, (95% CI −0.50 to −0.17; p<0.0001, I²=0%) and long duration interventions (≥6 months) showed an SMD of −0.38, (95% CI −0.62 to −0.14; p=0.002, I²=40%) (figure 4).

Figure 4. (A) Effect of dietary weight loss plus resistance exercise versus diet-only weight loss on fat mass in people living with overweight or obesity. (B) Effect of dietary weight loss plus resistance exercise versus diet-only weight loss on fat mass in people living with overweight or obesity in studies of duration ≤5 months. (C) Effect of dietary weight loss plus resistance exercise versus diet-only weight loss on fat mass in people living with overweight or obesity in studies of duration ≥6 months. RT, resistance training.

Figure 5, with 18 studies reported, shows a significantly lower reduction in fat-free mass for diet plus resistance exercise groups, compared with diet-only groups (SMD: 0.40, 95% CI 0.18 to 0.61; p=0.0003, I²=59%). This effect was significant for short (≤5 months) (SMD: 0.52, 95% CI 0.25 to 0.78; p=0.0001, I²=43%), but not long (≥6 months) (SMD: 0.20, 95% CI −0.09 to 0.48; p=0.17, I²=57%) duration interventions (figure 5).

Figure 5. (A) Effect of dietary weight loss plus resistance exercise versus diet-only weight loss on fat-free mass in people living with overweight or obesity. (B) Effect of dietary weight loss plus resistance exercise versus diet-only weight loss on fat-free mass in people living with overweight or obesity in studies of duration ≤5 months. (C) Effect of dietary weight loss plus resistance exercise versus diet-only weight loss on fat-free mass in people living with overweight or obesity in studies of duration ≥6 months. RT, resistance training.

Effects of resistance exercise intervention on markers of cardiometabolic health and physical function

Cardiometabolic health and physical function outcomes studied included muscle strength, cardiorespiratory fitness, blood glucose, insulin levels/sensitivity, lipid profile, blood pressure and glycaemic control. As shown in figure 6, eight studies reported that muscle strength was greater in diet plus resistance exercise compared with diet-only groups (SMD=2.36 95% CI 1.38 to 3.34; p=0.00001, I²=86%).

Figure 6. Effect of dietary weight loss plus resistance exercise versus diet-only weight loss on muscular strength in people living with overweight or obesity. RT, resistance training.

With eight studies included, no between-group differences were observed in cardiorespiratory fitness (VO_2max or peak oxygen consumption, VO_2peak), with a MD of 0.46 mL/kg/min, (95% CI −0.05 mL/kg/min to 0.96 mL/kg/min, I²=0%) (online supplemental figure 2). Blood glucose, with eight studies included, showed a MD of −0.01 mmol/L, (95% CI −0.05 mmol/L to 0.04 mmol/L, I²=0%) (online supplemental figure 3), and insulin levels showed a MD of −0.28 mU/L, (95% CI −1.18 mU/L to 0.62 mU/L, I²=0%) (online supplemental figure4). Insulin sensitivity, with six studies included, found an SMD of −0.18, (95% CI −0.44 to 0.09, I²=0%) (online supplemental figure 5) .

There were no differences in blood lipids in the nine studies which compared these between the diet plus resistance exercise and diet-only weight loss groups: total cholesterol: (MD=−0.01 mmol/L, 95% CI −0.20 mmol/L to 0.19 mmol/L, I²=84%); HDL cholesterol: (MD=−0.01 mmol/L, 95% CI −0.04 mmol/L to 0.03 mmol/L, I²=37%); LDL cholesterol: (MD=0.10 mmol/L, 95% CI −0.05mmol/L to 0.24 mmol/L, I²=69%); and triglycerides: (MD=0.00 mmol/L, 95% CI −0.00 mmol/L to 0.01 mmol/L, I²=0%) (online supplemental figures 6 and 7), (online supplemental figures 8 and 9).

Similarly, there were no between-group differences in blood pressure in the six studies reporting this. Systolic blood pressure had a MD of 0.05 mm Hg (95% CI −0.94 mm Hg to 1.04 mm Hg, I²=0%); and diastolic blood pressure demonstrated a MD of −0.68 mm Hg (95% CI −1.64 mm Hg to 0.28 mm Hg, I²=7%) (online supplemental figures 9 and 10).

In terms of physical function outcomes, we examined measures such as walking tests, chair stands, balance and flexibility. Although no meta-analyses were performed due to limited data reporting, these functional outcomes are important as they demonstrate the practical benefits of maintaining muscle mass/strength during weight loss. Walking performance yielded varied results. Two studies found no effects of resistance exercise during weight loss on 6 min walk test distance.^{50 51} However, one study found that physical function, measured by shuttle walk test, was greater in the resistance exercise group (Δ 165.0 ± 183.30 m, 53.7%, p=0.06) compared with the diet only group (Δ −14.3 ± 38.7 m, −9.7%).⁴¹ For other functional measures, one study reported no impact of resistance exercise during weight loss on physical function, assessed by the 5-chair stand and 400 m walk.²⁸ In one study, balance, as measured by one leg stand test, was greater after the weight loss intervention in the resistance exercise, relative to the diet only group.⁵¹ Additionally, one study reported improvements in 1 min sit-up test and reach flexibility test in the resistance exercise group compared with a diet only group (p<0.05).⁵⁰

Discussion

Our systematic review and meta-analysis examined the impact of resistance exercise on body weight/composition, muscle strength and cardiometabolic markers in people living with overweight or obesity taking part in dietary weight loss interventions. The addition of resistance exercise had no effect on changes in body mass but resulted in a greater loss in fat mass, preservation of fat-free mass and greater muscle strength. The certainty of evidence varied across outcomes, with high certainty for loss of fat mass, moderate certainty for fat-free mass preservation, but low certainty for muscle strength improvements—due to high heterogeneity and small sample sizes. Our subgroup analyses indicate that the effects on fat mass were similar regardless of intervention duration, but effects on fat-free mass were only seen in shorter duration interventions. There was no evidence of effects on cardiometabolic markers, although there was less certainty here due to the limited data that contributed to the comparisons.

Dietary weight loss results in loss of both fat-free mass and fat mass, and some studies have indicated that the addition of resistance exercise may have a role in preserving (or possibly increasing) fat-free mass.^{42 52} The current systematic review presents the first evidence synthesis on this subject, providing a level of evidence not shown before. To our knowledge, there is only one previous systematic review and meta-analysis which demonstrated that resistance exercise and caloric restriction was effective for decreasing fat mass and improving fat-free mass in people living with overweight or obesity.¹⁶ However, this review included children and adolescents, along with adults and older adults. On top of this, some of the included studies in the meta-analysis involved combined resistance and aerobic exercise and did not examine muscle strength/function or cardiometabolic health measures. The current systematic review and meta-analysis is therefore the first to specifically examine resistance exercise during dietary weight loss in adults living with overweight or obesity across a broad range of outcomes. Our analysis showed that resistance exercise protects against the loss of fat-free mass during weight loss, although this effect was not seen when the duration of intervention was greater than 5 months, which may reflect difficulties in maintaining adherence to the exercise in the long term. Indeed, data from longer duration studies (≥6 months) revealed significant dropout rates ranging from 25% to 49%,^{30 33 39 47 49} with generally poor adherence levels among completers—although this was poorly reported. For example, one study stated that during a 6-month period, only 33% of the resistance exercise plus diet group attended ≥90% of training sessions.³³

Fat-free mass preservation was only present in shorter duration interventions (<5 months). While declining adherence over time may contribute to this finding, other factors are also likely involved. For example, longer interventions may have featured less intensive supervision as they progressed, potentially reducing exercise quality and intensity, although reporting on this was generally poor. Additionally, physiological adaptations occur over time, with the adaptations plateauing after several months of the intervention with progressive overload required to stimulate further adaptations. One study reported a clear progression protocol including regular one-repetition maximum reassessments to adjust training loads, maintaining 75–85% intensity during the intervention.³⁷ Other studies, however, provided limited details on progression and only stated initial training intensity. Furthermore, dietary factors such as protein intake and overall caloric deficit, which may influence findings, are easier to control in shorter interventions. Future research is required to investigate the longer-term role of resistance exercise during weight loss.

Alongside changes in fat-free mass, our review demonstrates that resistance exercise during dietary weight loss results in greater muscle strength, although this finding is based on low certainty evidence. The link between changes in fat-free mass and muscle strength is complex and not always directly proportional. Although meta-analyses of resistance exercise in healthy adults clearly demonstrate that resistance exercise significantly increases muscle mass⁵³ as well as reducing total body fat,⁵⁴ there is evidence that resistance training exercise partially increases strength independently of changes in mass. This occurs via neural adaptations, such as increased recruitment of motor units, increased firing frequency and improved coordination movements between muscles.^{55 56} Our findings indicate that resistance exercise during dietary weight loss enhances strength (SMD 2.36) while also preserving fat-free mass to a lesser extent (SMD 0.40), suggesting that changes in strength are occurring both dependently and independently of changes in mass. Extending these findings, our review suggests that including resistance exercise in weight loss interventions may be beneficial in maximising fat loss, and importantly minimising the potential negative consequences of weight loss, such as loss of lean muscle mass.

Despite fat-free mass and muscle strength being associated with a variety of positive health outcomes,¹³ and resistance exercise resulting in improvement in cardiometabolic health markers,¹⁴ we did not find any impact of the addition of resistance exercise to a weight loss intervention on markers of cardiometabolic health. The certainty of evidence for these outcomes ranged from low to high, with total cholesterol and LDL cholesterol being particularly uncertain due to high heterogeneity. Instead of simply reflecting the importance of weight loss itself in changes in these outcomes, this may indicate several possibilities. The level of caloric restriction in weight loss interventions may mask the cardiometabolic benefits of resistance exercise that are seen in non-calorie-restricted settings. It is also possible that the prescribed resistance exercise training volume, intensity or progression in these interventions was not sufficient to produce further cardiometabolic benefits over and above those from weight loss alone. Additionally, the timing of outcome assessments may not have captured the whole adaptive response, especially if metabolic benefits from resistance training follow time frames that are distinct from the metabolic benefits of dietary restriction. It is crucial to emphasise that there is strong evidence supporting resistance training as an effective single intervention for improving cardiometabolic outcomes frequently with few or no changes in body weight.^{13 57} Alternatively, there were fewer studies and greater uncertainty in our analysis of these outcomes, and we would suggest that further work, including larger RCTs with carefully designed protocols and appropriate assessments, is needed on this topic.

The WHO’s physical activity guidelines recommend performing muscle strengthening exercises involving all major muscle groups twice a week⁵⁸ and we have shown that resistance exercises can help with the preservation of fat-free mass during weight loss. Unfortunately, it is also known that participation in resistance exercise is generally very low (17–30%)^{59 60} and so to achieve any of the benefits of resistance exercise, strategies to increase its uptake are needed.⁶¹ This can be particularly challenging as the majority of the studies included in this review employed supervised resistance exercise at specialist facilities. Resistance exercise training traditionally involves specialised equipment, which may not be accessible to many people due to barriers generally associated with any physical activity such as work, time, vacations, weather, boredom, tiredness, injury/illness and family commitments⁶²,⁶⁴ which can further limit participation. However, it is important to highlight that just making resistance exercise training more accessible without maintaining adequate intensity and progression may limit effectiveness, as shown in some large trials using simpler, non-supervised approaches.⁶⁵ Despite this challenge, there are interventions that balance accessibility with effectiveness, which include progressive bodyweight exercise training,^{66 67} programmes using resistance bands with systematic progression protocols⁶⁸ and home-based programmes that use household objects while emphasising appropriate technique and progressive overload.⁶⁹ We would argue, therefore, that we need to develop pragmatic resistance training exercises that are simple, easy to use and effective in preserving fat-free mass during weight loss. These might include home-based resistance training programmes requiring minimal equipment which could be widely implemented among people with overweight or obesity who are trying to lose weight.

To our knowledge, this is the first systematic review and meta-analysis assessing the effect of resistance exercise on body weight/composition, muscle strength and cardiometabolic markers in people with overweight or obesity taking part in weight loss interventions. All studies included in our review were RCTs, providing valuable evidence on the effects of resistance exercise during weight loss, despite their varying methodological quality. For the meta-analysis, resistance exercise was the only exercise performed by the participants, since studies that had only aerobic exercise or balance training were excluded in order to allow for a more focused analysis of the effects of resistance exercise during dietary weight loss interventions compared with dietary weight loss alone. Our research indicates that resistance training improves fat-free mass preservation (moderate certainty) and loss of fat mass (high certainty) during weight loss. Although muscle strength improvements were observed, this evidence is of low certainty.

The current review has a number of limitations. First, most included studies failed to report the concealment of allocation, blinding of outcome assessment and missing data. As a result, many studies were rated as having an unclear or high risk of bias, which may have influenced the heterogeneity of the analysis. Second, only a few studies reported cardiometabolic health outcomes, and available evidence was limited for each outcome. The quality of evidence for these outcomes was therefore either low or moderate, and their effect estimates may lack accuracy. The prediction intervals reported in the current meta-analysis indicate that adding resistance exercise to dietary weight loss interventions would result in beneficial effects, for example greater loss of fat mass, but due to low study numbers and high heterogeneity, prediction intervals cross zero for fat-free mass (95% PI −0.30 to 1.09) and muscle strength (95% PI −0.36 to 5.08) and, as such, more RCTs are required to provide greater clarity for these outcomes. Third, over half of the included studies (n=15) recruited only women, which may limit the generalisability of the findings. In order to make these findings more applicable and to gain a more comprehensive understanding of how resistance exercise during dietary weight loss may be more effective, future research should aim to include more representative samples. Fourth, most included studies employed supervised resistance exercise at specialist facilities, with very few examining unsupervised or home-based interventions. This prevented us from conducting subgroup analysis comparing the effectiveness of supervised versus unsupervised resistance training during weight loss. Given that supervision may influence adherence, exercise intensity and technique—all factors that could affect outcomes—this limitation highlights the need for more research on accessible, pragmatic approaches to resistance training that do not require extensive supervision. Fifth, only seven of the included studies were of 6-month duration, which has two important limitations: (1) the small sample size reduces statistical power for our subgroup analyses, increasing the uncertainty around the estimated effects, and (2) the inconsistent reporting of exercise adherence across these studies prevented us from conducting meta-regression analyses to assess whether declining adherence explains the decreased effects seen in longer interventions. Sixth, the presence of suspected publication bias contributed to downgrading the certainty of evidence too low for total cholesterol, LDL cholesterol and diastolic blood pressure in our GRADE assessments, with funnel plot asymmetry suggesting potential publication bias and limiting conclusions that can be drawn from this data. Seventh, functional outcomes, including walking, chair stand, balance and flexibility tests, should be considered exploratory as they were specified as additional outcomes in our PROSPERO protocol and data on these outcomes was limited. Finally, our review only included peer-reviewed papers and included English language publications and thus may have missed relevant studies published in the grey literature and other languages.

In conclusion, the current study indicates that the potential benefits of resistance exercise, including increasing loss of fat mass, reducing the loss of fat-free mass and improving muscle strength for people taking part in a dietary weight loss intervention. However, it is important to recognise the limitations that have been identified. Further research is needed to investigate the cardiometabolic effects of resistance exercise during weight loss. There is also a need to develop and evaluate more pragmatic resistance exercise interventions that can be widely implemented. Including resistance exercise in dietary weight loss interventions is important to improve outcomes and inform evidence-based practice.

Data availability statement

Data are available upon reasonable request.