Effects of concurrent training on fat mass and its distribution in individuals with overweight or obesity: a systematic review, meta-analysis, and exploratory regression analysis

Abstract

Background

Obesity is a major global health challenge. While aerobic and resistance exercises are commonly used for obesity management, the effects of concurrent training (CT, combining both modalities) on fat mass and its distribution in individuals with overweight or obesity remain unclear. This study examined the effects of CT’s on visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT), and tried to identify optimal exercise prescriptions.

Methods

Six databases were systematically searched through September 2025. Randomized controlled trials comparing CT interventions with sedentary or usual-care controls in individuals with overweight or obesity were included. Meta-analyses calculated standardized mean differences (SMD) for VAT and SAT reduction. The percentage ratio of visceral/subcutaneous fat loss (%ΔV/%ΔS) was calculated, and predictors were explored through regression analysis.

Results

Eighteen studies were included. CT significantly reduced VAT (SMD = -0.36, 95% CI: -0.59 to -0.13, P < 0.01, I² = 57%) and SAT (SMD = -0.26, 95% CI: -0.40 to -0.12, P < 0.01, I² = 1%), with greater percentage reduction in VAT than SAT (%ΔV/%ΔS = 1.96 ± 1.18). Optimal effects were observed with ≥ 3 sessions per week, ≥ 60 min per session, > 180 min per week, > 15 weeks, and vigorous-intensity aerobic exercise combined with moderate-to-low-intensity resistance training. Intervention frequency and baseline VAT/SAT ratio significantly predicted fat distribution changes.

Conclusions

Current evidence suggests that CT can effectively reduce fat mass and may preferentially decrease VAT compared to a sedentary lifestyle in individuals with overweight or obesity. Our findings support CT as a promising and scalable strategy for population-level obesity management, particularly for high-risk groups such as postmenopausal women and older adults. Furthermore, the identified exercise prescription parameters provide practical guidance for populations aiming to reduce adipose tissue.

Trial registration

CRD42024595992.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-26869-5.

Introduction

The prevalence of obesity has been increasing due to excessive energy intake and sedentary lifestyles, becoming a major global public health challenge [1]. According to the World Health Organization (WHO), the number of obese adults worldwide has more than doubled since 1990, and obesity among adolescents has increased fourfold [2]. Obesity is closely associated with the development of a range of diseases, including cardiovascular disease, type 2 diabetes, chronic obstructive pulmonary disease, musculoskeletal disorders, and certain cancers [3–5], which reduce quality of life and burden healthcare systems.

In obesity, adipose tissue undergoes structural remodeling and functional impairment. The percentage of abdominal fat is considered an independent risk factor for cardiovascular disease in individuals with overweight or obesity [6]. Under normal physiological conditions, abdominal fat is mainly stored as subcutaneous adipose tissue (SAT), accounting for 80%–90%. SAT is located under the skin and is not connected to internal organs [7]. It is generally considered a relatively healthy form of fat storage, as it avoids ectopic fat deposition [8] and helps buffer excess energy without causing metabolic disorders [6]. In contrast, visceral adipose tissue (VAT) is more metabolically active and poses greater health risks by secreting various bioactive substances and inducing lipotoxicity in non-adipose tissues [9, 10]. Therefore, targeting VAT reduction through exercise intervention represents a primary therapeutic goal for alleviating metabolic disorders [11].

In recent years, many studies have found that dietary [12, 13], pharmacologic [14], surgical [15], and exercise interventions [16] are effective in reducing adipose tissue. However, these interventions differ considerably in their impact on treating obesity and related diseases. Several meta-analyses have shown that exercise is more effective in reducing VAT than dietary restriction and medication [17–19]. Regular exercise improves the function of VAT through mechanisms such as increased insulin sensitivity [20], enhanced fat metabolism efficiency [21], suppression of chronic cellular inflammation [22], and modulation of obesity-related genes [23]. Additionally, exercise can remodel the SAT in adults with overweight or obesity, inducing specific adaptive changes in adipocyte structure and function, which in turn improves the effective storage capacity of SAT and reduces ectopic deposition of fat in other tissues or organs [24].

Recently, studies have begun to focus on the effects of concurrent aerobic and resistance training (concurrent training, CT) on individuals with different body compositions, but the results have been inconsistent. Research on professional athletes has found that CT has an inhibitory effect on cardiovascular aerobic endurance and muscle strength and mass, especially when one of the two types of training is excessively strengthened [25–27]. However, for non-professional populations, the cardiorespiratory and muscle effects of CT do not appear to inhibit each other, due to the lower total training load [28–30]. Nevertheless, the impact of CT on fat mass and its distribution still shows great heterogeneity in individuals with overweight and obesity. Some studies have concluded that CT reduces VAT and SAT in individuals with overweight or obesity [31–33]. However, others have found no significant effect of CT on reducing VAT [34–36]. This controversy may stem from the complexity of CT variables (type, intensity, frequency, duration, and training volume) and the large number of factors influencing individual VAT and SAT (age, gender, genetics, and ethnicity). Additionally, most recent studies have focused only on the effects of exercise on VAT or whole-body fat, neglecting the role of SAT changes in improving metabolic health and the specific ways exercise affects fat distribution. Therefore, understanding the effects of exercise on fat mass and its distribution—particularly the disparities in the magnitude and speed of response of VAT and SAT to exercise interventions—holds substantial theoretical and practical significance for investigating the mechanisms underlying the response of regional adipose metabolism to exercise and optimizing exercise prescriptions of fat reduction and weight management.

Based on this, we conducted a comprehensive systematic review and meta-analysis of randomized controlled trials (RCTs) to investigate CT effects on VAT and SAT in individuals with overweight or obesity, and performed subgroup analyses to identify optimal exercise protocols for these individuals. Next, we analyzed the percentage ratio of visceral/subcutaneous fat loss (%ΔV/%ΔS) after the intervention to reveal CT characteristics in modulating fat distribution. Finally, the key CT factors affecting %ΔV/%ΔS were explored by linear regression analysis.

Materials and methods

Meta-analysis

The Cochrane Handbook for Systematic Reviews of Interventions [37] and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines [38] were followed to conduct this systematic review and meta-analysis. The study protocol was registered in the International Prospective Register of Systematic Reviews (CRD42024595992).

Search strategy

For this study, we systematically searched the PubMed, Web of Science, Scopus, Cochrane Library, Embase, and EBSCO databases (28 July 2024, updated on 14 September 2025). The search terms employed included “concurrent training”, “overweight”, “obese”, “visceral adipose tissue” and “subcutaneous adipose tissue”. The complete search strategy for each database is available in Supplement Table 1. In addition, we searched the reference lists of selected articles and reviews to identify any relevant studies that electronic searches might have missed. Title and abstract screening, as well as full-text review, were conducted independently by two authors (XL and GL) using a standardized protocol. If there was any disagreement, the corresponding author (SW) was consulted to resolve the discrepancies and reach a consensus.

Table 1.

Characteristics of the studies included in this meta-analysis

Author and published year	Country	Sample size	Gender (Male/Female)	Age (Baseline)	Weight (kg) (Baseline)	BMI (kg/m²) (Baseline)	Concomitant conditions	Concurrent training intervention information
								Exercise type	Intensity	Duration per session	Frequency (times/week)	Weeks	Supervised or unsupervised
Alberga 2015 [57]	Canada	CT: 55 CON: 57	NA	CT: 15.5 ± 1.3 CON: 15.6 ± 1.3	CT: 97.5 ± 16.3 CON: 97.9 ± 18.8	CT: 34.6 ± 4.2 CON: 34.1 ± 4.9	NA	AE: treadmill, cycle ergometer or elliptical machine RT: resistance machines (whole-body exercises)	AE: 70–85% HRmax RT: 100% 1RM	NA	4	22	Supervised
Barone 2009 [62]	USA	CT: 51 CON: 53	M: 51 F: 53	64.6 ± 5.7	CT: 83 ± 15 CON: 85 ± 19	CT: 29.4 ± 3.8 CON: 29.7 ± 5.0	Untreated prehypertension or mild hypertension	AE: aerobic exercise RT: lat pulldown, leg extension, leg curl, bench press, leg press, shoulder press, and seated mid-rowing	AE: 60–90% HRmax RT: 50% 1RM	NA	3	26	Supervised
Bouchonville 2014 [58]	Mexico	CT: 22 CON: 23	NA	CT: 70 ± 4 CON: 69 ± 4	CT: 99.2 ± 17.4 CON: 101.0 ± 16.3	CT: 36.9 ± 5.4 CON: 37.3 ± 4.7	NA	AE: treadmill, stationary cycling, and stair climbing RT: weight-lifting machines (nine upper and lower exercises)	AE: 65–85% HRpeak RT: 65–85% 1RM	90	3	48	Supervised
Cuff 2003 [64]	Canada	CT: 10 CON: 9	M: 0 F: 19	CT: 63.4 ± 2.2 CON: 60.0 ± 2.9	CT: 89.5 ± 3.9 CON: 95.6 ± 6.5	CT: 33.3 ± 1.5 CON: 36.7 ± 2.0	T2DM	AE: treadmills, stationary bicycles, recumbent steppers, elliptical trainers, and rowing machines RT: leg press, leg curl, hip extension, chest press, and lat pulldown	AE: 60–75% HRR RT: NA	75	3	16	Supervised
Davis 2011 [61]	USA	CT: 14 CON: 12	M: 0 F: 26	CT: 15.7 ± 1.1 CON: 15.8 ± 1.0	CT: 80.2 ± 10.7 CON: 94.4 ± 13.4	CT: 32.4 ± 3.2 CON: 36.4 ± 5.2	NA	AE: treadmill, elliptical machines, and aerobic classes RT: whole-body training (including upper and lower muscle groups; six exercises)	AE: 70–85% HRmax RT: NA	60–90	2	16	Supervised
Dobrosielski 2012 [60]	USA	CT: 51 CON: 63	NA	CT: 57 ± 6 CON: 56 ± 6	CT: 97.4 ± 2.1 CON: 99.2 ± 14.1	CT: 33.0 ± 0.6 CON: 33.6 ± 0.5	T2DM and Hypertension	AE: treadmill, stationary cycle, or stairstepper RT: multistation machine (seven exercises)	AE: 60–90% HRmax RT: 50% 1RM	NA	3	26	Supervised
Donges 2013 [59]	Australia	CT: 13 CON: 8	M: 21 F: 0	CT: 46.2 ± 1.4 CON: 49.5 ± 2.6	CT: 96.4 ± 1.7 CON: 92.2 ± 6.9	CT: 30.2 ± 0.7 CON: 29.6 ± 2.1	NA	AE: cycle ergometry, elliptical cross training RT: chest and shoulder press, seated rows, lat pulldown, leg press, leg curls, lunges, machine squats, and deadlifts	AE: 75–80% HRmax RT: 75–80% 1RM	NA	3	12	Supervised
Dupuit 2022 [47]	France	CT: 8 CON: 9	M: 0 F: 17	CT: 58.8 ± 5.3 CON: 60.9 ± 4.8	CT: 77.8 ± 12.4 CON: 80.3 ± 11.2	CT: 30.3 ± 3.5 CON: 31.5 ± 3.4	NA	AE: cycling RT: 10 exercises	AE: 84–94% HRmax RT: 8–12 RM	45	3	12	Supervised
Fortuin-de Smidt 2020 [51]	South Africa	CT: 20 CON: 15	M: 0 F: 35	CT: 22 ± 1.5 CON: 23 ± 3.1	CT: 84.1 ± 8.7 CON: 87.8 ± 10.9	CT: 34.1 ± 2.8 CON: 33.4 ± 2.7	NA	AE: dance, running, skipping and stepping RT: body weight, bands and free weights	AE: 75–80% HRpeak RT: 60–70% HRpeak	40–60	4	12	Supervised
Hens 2021 (3 month) [48]	Belgium	CT + D: 28 D: 26	M: 0 F: 54	CT: 37.64 ± 8.53 CON: 36.11 ± 8.94	NA	CT: 32.98 ± 3.60 CON: 32.27 ± 3.50	NA	AE: 3–4 different cardio devices RT: core stability training, four strength exercises isotonic strength training	AE: 90–95% HR achieved at respiratory compensation point RT: NA	50–65	3	12	Unsupervised
Hens 2021 (6 month) [48]	Belgium	CT + D: 27 D: 23	M: 0 F: 50	CT: 37.64 ± 8.53 CON: 36.11 ± 8.94	NA	CT: 32.98 ± 3.60 CON: 32.27 ± 3.50	NA	AE: 3–4 different cardio devices RT: core stability training, four strength exercises isotonic strength training	AE: 90–95% HR achieved at respiratory compensation point RT: NA	50–65	3	24	Unsupervised
Houghton 2017 (1) [53]	UK	CT: 12 CON: 12	NA	CT: 54 ± 12 CON: 51 ± 16	CT: 90 ± 18 CON: 94 ± 9	CT: 33 ± 7 CON: 33 ± 5	Nonalcoholic steatohepatitis	AE: cycling RT: hip and knee extension, horizontal row, chest press, vertical row, and knee extension	AE: Borg 16-18 RT: Borg 14–16	45–60	3	12	Supervised
Houghton 2017 (2) [54]	UK	CT: 14 CON: 13	NA	CT: 51 ± 10 CON: 56 ± 13	NA	CT: 33 ± 6 CON: 31 ± 3	NA	AE: cycling RT: hip and knee extension, horizontal row, chest press, vertical row, and knee extension	AE: Borg 16-18 RT: Borg 14–16	45–60	3	12	Supervised
Mendham 2020 [50]	South Africa	CT: 20 CON: 15	M: 0 F: 35	CT: 22 ± 1.5 CON: 23 ± 3.1	CT: 84.1 ± 8.7 CON: 87.8 ± 10.9	CT: 34.1 ± 2.8 CON: 33.4 ± 2.7	NA	AE: dancing, running, skipping and stepping RT: upper and lower-body exercises at body weight that progressed to the use of equipment	AE: 75–80% HRpeak RT: 60–70% HRpeak	40–60	4	12	Supervised
Monteiro 2015 [56]	Brazil	CT: 14 CON: 16	M: 17 F: 13	CT: 11.03 ± 1.34 CON: 11.04 ± 1.90	CT: 86.67 ± 12.98 CON: 77.60 ± 16.97	CT: 33.17 ± 4.70 CON: 30.95 ± 3.42	NA	AE: walking and running RT: leg press, low rowing, bench press, squat rack, lat pulldown, leg curl, arm curl, seated chest fly, triceps, leg extension, sit up, and supine trunk extension	AE: 65–85% VO2peak RT: 55–75% 1RM	60	3	20	Supervised
Nono Nankam 2020 [49]	South Africa	CT: 20 CON: 15	M: 0 F: 35	27.5 ± 10.6	CT: 84.1 ± 2.2 CON: 87.8 ± 2.5	CT: 34.1 ± 0.6 CON: 33.4 ± 0.7	NA	AE: dance, running, skipping and stepping RT: squats, lunges, bicep curls, push-ups and shoulder press	AE: 75–80% HRpeak RT: 60–70% HRpeak	40–60	4	12	Supervised
Park 2003 [63]	Korea	CT: 10 CON: 10	M: 0 F: 20	CT: 43.4 ± 1.04 CON: 43.1 ± 1.67	CT: 67.5 ± 5.10 CON: 65.2 ± 1.87	CT: 25.8 ± 1.43 CON: 25.5 ± 0.86	NA	AE: side by side, step touch, lunge side, v-step, grapevine, pivot turn, cha cha cha, mambo rock, diamond step, single hamstring walking, heel touch, sit-up, push up, fast walking, turn round, heel side, knee-up, scissors double, hop and jump, jumping jack, side kick, full turn, double kick RT: bench press, side raise, triceps push away, barbell curl, leg curl, leg extension, leg press, leg raise, abdominal crunch, lat pulldown	AE: 60–70% HRmax RT: 60–70% 1RM	60	6	24	NA
Park 2015 [55]	Korea	CT: 10 CON: 10	M: 0 F: 20	CT: 57.20 ± 2.57 CON: 57.20 ± 1.69	CT: 64.9 ± 3.83 CON: 64.7 ± 3.26	CT: 26.02 ± 1.55 CON: 26.80 ± 1.09	NA	AE: running RT: Arm curl, leg curl, leg extension, squat, chest press, low back, shoulder press, lat pulldown, back extension, and sit-up	AE: 40–75% HRR RT: 60–70% 1RM	90	3	12	NA
Rashti 2019 (HIIT) [52]	Iran	CT: 15 CON: 10	M: 0 F: 25	CT: 57.11 ± 4.13 CON: 54.10 ± 5.08	CT: 70.32 ± 6.49 CON: 72.45 ± 12.36	CT: 28.31 ± 2.20 CON: 29.92 ± 4.15	NA	AE: treadmill RT: circuit exercises (including upper and lower muscle groups; nine exercises)	AE: 65–95% HRmax RT: 60–85% 1RM	50–65	3	10	Supervised
Rashti 2019 (MICT) [52]	Iran	CT: 14 CON: 10	M: 0 F: 24	CT: 54.10 ± 5.08 CON: 54.10 ± 5.08	CT: 74.45 ± 6.17 CON: 72.45 ± 12.36	CT: 30.62 ± 3.08 CON: 29.92 ± 4.15	NA	AE: treadmill RT: circuit exercises (including upper and lower muscle groups; nine exercises)	AE: 60–75% HRmax RT: 40–70% 1RM	50–65	3	10	Supervised

AE aerobic exercise, BMI body mass index, CT concurrent training group, CON control group, D diet group, F female, HR heart rate, HRmax maximum heart rate, HRpeak peak  reserve, M male, RT resistance training, RM repetition maximum, Reps repetitions, SAT subcutaneous adipose tissue, T2DM type 2 diabetes mellitus, VO₂peak maximal oxygen uptake peak, VAT visceral adipose tissue. heart rate, HRR heart rate

Eligibility criteria

The inclusion criteria were based on the PICOS (Participants, Interventions, Comparators, Outcomes, Study design) approach [39]:

• (P) Individuals who are overweight or obese, with a body mass index (BMI) ≥ 25 kg/m².

• (I) Concurrent training. It is widely accepted that exercise interventions shorter than 8 weeks do not yield substantial physiological adaptations. Consequently, this study restricted its inclusion to research with an intervention period of at least 8 weeks.

• (C) To control for lifestyle, environmental, and behavioral confounding factors, control group participants were required to maintain their usual daily routines without engaging in any structured exercise program. The control groups represent sedentary or usual-care controls rather than active intervention comparisons. Therefore, the observed effects reflect the benefit of CT compared to maintaining habitual lifestyle behaviors, not the superiority of CT over other exercise modalities.

• (O) The primary outcomes included VAT and SAT.

• (S) RCTs.

Exclusion criteria

(1) Duplicate publications; (2) Literature reviews; (3) Conference abstracts; (4) Animal studies; (5) Reports on the acute effects of a single exercise session; (6) Non-English publications.

Data extraction

Data extraction was conducted by teams of two reviewers (XL and GL), working independently and in duplicate. Disagreements were resolved through consultation with the corresponding author (SW). The extracted data included: (a) study characteristics (first author, year of publication, country, sample size); (b) participant characteristics (sex, age, weight, BMI, disease type); (c) intervention characteristics (type, intensity, frequency, session duration, and duration of intervention); and (d) outcome measures (VAT and SAT). If means and standard deviations were not reported in the text or tables, baseline and post-intervention data were requested from the corresponding authors. If data were only available in graphical form and not provided upon request, they were extracted using WebPlotDigitizer (version 4.8) [40].

Methodological quality assessment

The methodological quality of the included studies was assessed using the Cochrane risk of bias criteria, which included 7 items: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete result data (loss bias), selective reporting (report bias), and other biases [41]. Two researchers (XL and GL) independently conducted methodological quality assessments. If there was disagreement between the two authors, a third researcher (SW) was consulted to reach a consensus.

Evidence certainty

This study used the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) system to assess the overall quality of evidence [42]. Randomized controlled trials (RCTs) are initially rated as high-quality evidence by default. The quality level of evidence for outcome indicators is determined based on several factors: risk of bias, inconsistency, indirectness, imprecision, and publication bias.

Statistical analysis

Statistical analysis of the included data was performed using RevMan 5.4. The outcome indicators included in this study were continuous variables. Standardized Mean Difference (SMD) and its 95% Confidence Interval (CI) were used as indicators of effect sizes, as some studies used different measures and units, and all analyses used a random effects model to generate forest plots. Interpretation of effect sizes according to the Cochrane guidelines is shown below: 0.20–0.49 as small, 0.50–0.79 as medium, and ≥ 0.80 as large. Between-study heterogeneity was assessed using the I² statistic, with values of 25%, 50%, and 75% indicating low, medium, and high heterogeneity [43]. Publication bias was evaluated using funnel plots and Egger’s test. Sensitivity analyses were conducted by sequentially excluding each study to examine its impact on the overall effect size. Statistical significance was defined as P < 0.05.

In subgroup analyses, we attempted to use participants’ BMI (≥ 25 to < 30 kg/m², ≥ 30 kg/m²), age (< 18 years, 18–44 years, 45–59 years, 60–74 years), geographic affiliation (Asia, Africa, Europe, Oceania, North America, and South America), frequency of intervention (< 3 sessions per week, ≥ 3 sessions per week), exercise duration per session (< 60 min per session, ≥ 60 min per session), only aerobic exercise duration per session (< 30 min per session, ≥ 30 min per session), only resistance training duration per session (< 30 min per session, ≥ 30 min per session), exercise duration per week (≤ 180 min per week, > 180 min per week), duration of intervention (< 12 weeks, 12–15 weeks, > 15 weeks), intensity of concurrent aerobic and resistance training [44, 45] (Moderate + Medium-low, Vigorous + Medium-low, Vigorous + High, High + Medium-low, High + High), and women’s menopausal status (premenopausal, postmenopausal).

Linear regression analysis

Data extraction

Changes in visceral and subcutaneous adipose tissue were recorded, and the percentage ratio of VAT and SAT loss (%VAT loss and %SAT loss), as well as the ratio between them (%ΔV/%ΔS), were calculated for each study intervention. When %ΔV/%ΔS > 1, it indicates a higher percentage loss in VAT; when %ΔV/%ΔS = 1, it indicates equal percentage loss in VAT and SAT; and when %ΔV/%ΔS < 1, it indicates a higher percentage loss in SAT.

Exploratory regression analysis

To explore potential moderators of the %ΔV/%ΔS ratio, we conducted an exploratory linear regression analysis using the R, following the established methodology of Chaston et al. [46]. This analysis was hypothesis-generating in nature, aimed at identifying factors that may influence the differential loss of visceral versus subcutaneous fat following CT interventions.

To ensure a linear relationship in the data, %ΔV/%ΔS was natural log transformed (log, i.e., ln). All data analyses were unweighted for sample size (n), i.e., data points from each study were treated as independent observations. Outliers were identified and excluded using box-and-whisker plots to ensure robustness of the data. The Shapiro-Wilk test was used to assess the normal distribution of ln %ΔV/%ΔS. Using the regsubsets method, predictor variables were included in the model to find the optimal model and assess the quality of model fit based on the Bayesian Information Criterion (BIC), Adjusted R² and Mallow’s Cp. Finally, the optimal model was refitted to assess the association of the predictor variables with ln %ΔV/%ΔS after the CT intervention, and the variance inflation factor (VIF) was calculated to test for multicollinearity problems.

Results

Study selection

As shown in Fig. 1, a total of 4,054 articles were initially retrieved from the databases, along with 2 additional articles identified through other sources. After removing duplicates, 2,247 articles remained. Following title and abstract screening, 2,205 articles were excluded for not meeting the inclusion criteria. The full texts of the remaining 42 studies were reviewed, leading to the exclusion of 24 studies for the following reasons: (1) no data (n = 3); (2) non-English (n = 2); (3) no control group (n = 8); and (4) no outcome indicators (n = 11). Finally, 18 articles [47–64], totaling 20 studies, met the inclusion criteria for the systematic review and meta-analysis. Among these, 1 article included two exercise intervention groups (a moderate-intensity group and a high-intensity interval group) [52], and another article reported two endpoint values (at 3 and 6 months) [48].

Fig. 1.

PRISMA flowchart of literature search for eligible studies

Study characteristics

The main characteristics of the participants and interventions are shown in Table 1. The included studies involved a total of 777 participants across 19 experimental groups and 18 control groups. These studies were from Asia (n = 3) [52, 55, 63], Africa (n = 3) [49–51], Europe (n = 4) [47, 48, 53, 54], North America (n = 6) [57, 58, 60–62, 64], South America (n = 1) [56], and Oceania (n = 1) [59]. Among the 18 studies, 1 study [59] involved only men, 10 studies [47–52, 55, 61, 63, 64] involved only women, 5 studies [53, 54, 57, 58, 60] did not report participants’ gender, and 2 studies [56, 62] involved both men and women. The age of the participants ranged from 11 to 70 years. Specifically, 3 studies [56, 57, 61] involved participants aged < 18 years, 5 studies [48–51, 63] involved participants aged 18–44 years, 7 studies [47, 52–55, 59, 60] included participants aged 45–59 years, and 3 studies [58, 62, 64] involved participants aged 60–74 years. All participants were classified as overweight or obese based on BMI. Specifically, participants in 4 exercise groups [52, 55, 62, 63] were overweight (≥ 25 to < 30 kg/m²), while those in 16 exercise groups [47–51, 53, 54, 56–61, 64] were obese (≥ 30 kg/m²). In 17 studies [47, 49–64], the intervention in the exercise group was concurrent training, while the control group maintained their daily lifestyle; 1 study [48] was concurrent training plus a low-calorie diet intervention, with the control group following the same dietary restrictions. The intervention duration ranged from 10 to 48 weeks. The duration of intervention per session ranged from 40 to 90 min. The frequency of intervention per week ranged from 2 to 6 sessions. The aerobic exercise intensity ranged from moderate to high. The resistance training intensity ranged from low to high. 15 studies were supervised [47, 49–54, 56–62, 64], 1 study [48] was unsupervised, and 2 studies [55, 63] did not report supervision details.

Quality assessment

The results of the quality assessment for the 18 included studies are shown in Fig. 2. Among them, 38.8% (7/18) explicitly reported randomized sequence generation, while 16.7% (3/18) reported allocation concealment. Since all interventions were exercise-based, a fully double-blind design was not feasible, and all studies were therefore assessed as having a high risk of performance bias. Outcome measures were generally objective, resulting in an unclear or low risk of detection bias. Regarding attrition bias, most studies had low dropout rates that were adequately described or analyzed, and were thus judged as having an unclear or low risk. All studies were fully reported, with no evidence of other sources of bias.

Fig. 2.

Results of Cochrane risk of bias tool

Meta-analysis results

Effects of CT on VAT and SAT in individuals with overweight or obesity

CT significantly reduced VAT in individuals with overweight or obesity, with a small effect size (SMD = -0.36, 95% CI: -0.59 to -0.13, P < 0.01) and medium heterogeneity (I² = 57%). CT also significantly reduced SAT, with a small effect size (SMD = -0.26, 95% CI: -0.40 to -0.12, P < 0.01) and low heterogeneity (I² = 1%) (Fig. 3 and Supplementary Fig. 1).

Fig. 3.

Results of moderator analysis. AE: aerobic exercise; CI: confidence interval; RT: resistance training SAT: subcutaneous adipose tissue; SMD: standardized mean difference; VAT: visceral adipose tissue

Subgroup analysis

Subgroup analyses showed (Fig. 3) that CT significantly reduced VAT (overweight: SMD = -1.57, 95% CI: -2.68 to -0.46, P < 0.01, I² = 85%) (obesity: SMD = -0.16, 95% CI: -0.31 to -0.00, P = 0.05, I² = 0%) and SAT (overweight: SMD = -0.63, 95% CI: -0.94 to -0.32, P < 0.01, I² = 0%) (obesity: SMD = -0.17, 95% CI: -0.33 to -0.02, P = 0.03, I² = 0%) in individuals with BMI ≥ 25 to < 30 kg/m² (overweight) and ≥ 30 kg/m² (obesity) compared with the control group.

CT significantly reduced VAT in middle-aged (45–59 years: SMD = -0.35, 95% CI: -0.65 to -0.06, P = 0.02, I² = 21%) and older adults (60–74 years: SMD = -0.65, 95% CI: -0.98 to -0.31, P < 0.01, I² = 8%) with overweight or obesity. However, no significant effect was observed in individuals aged < 45 years (< 18 years: SMD = -0.14, 95% CI: -0.44 to 0.17, P = 0.38, I² = 0%) (18–44 years: SMD = -0.44, 95% CI: -1.08 to 0.20, P = 0.18, I² = 80%). In addition, CT significantly reduced SAT in older adults with overweight or obesity (60–74 years: SMD = -0.70, 95% CI: -1.01 to -0.38, P < 0.01, I² = 0%), but had no significant effect on SAT in individuals with overweight or obesity < 60 years of age (< 18 years: SMD = -0.15, 95% CI: -0.67 to 0.37, P = 0.57, I² = 53%) (18–44 years: SMD = -0.12, 95% CI: -0.38 to 0.14, P = 0.36, I² = 0%) (45–59 years: SMD = -0.18, 95% CI: -0.42 to 0.06, P = 0.14, I² = 0%).

CT significantly reduced VAT in individuals with overweight or obesity in Asia (SMD = -1.51, 95% CI: -2.92 to -0.10, P = 0.04, I² = 87%) and North America (SMD = -0.35, 95% CI: -0.64 to -0.06, P = 0.02, I² = 48%), but had no significant effect on VAT in individuals with overweight or obesity in Africa (SMD = -0.22, 95% CI: -0.61 to 0.17, P = 0.27, I² = 0%), Europe (SMD = -0.08, 95% CI: -0.38 to 0.22, P = 0.59, I² = 0%), Oceania (SMD = -0.26, 95% CI: -1.14 to 0.63, P = 0.56), and South America (SMD = -0.05, 95% CI: -0.77 to 0.67, P = 0.89). In addition, CT significantly reduced SAT in individuals with overweight or obesity in North America (SMD = -0.44, 95% CI: -0.69 to -0.18, P < 0.01, I² = 35%), but had no significant effect on SAT in individuals with overweight or obesity in Asia (SMD = -0.37, 95% CI: -0.80 to 0.05, P = 0.09, I² = 0%), Africa (SMD = -0.24, 95% CI: -0.62 to 0.15, P = 0.23, I² = 0%), Europe (SMD = -0.01, 95% CI: -0.31 to 0.29, P = 0.94, I² = 0%), Oceania (SMD = -0.22, 95% CI: -1.11 to 0.66, P = 0.62), and South America (SMD = 0.43, 95% CI: -0.30 to 1.15, P = 0.25).

CT conducted ≥ 3 sessions per week significantly reduced VAT (SMD = -0.37, 95% CI: -0.60 to -0.13, P < 0.01, I² = 59%) and SAT (SMD = -0.25, 95% CI: -0.39 to -0.11, P < 0.01, I² = 0%) in individuals with overweight or obesity. However, CT conducted < 3 sessions per week had no significant effect on VAT (SMD = -0.32, 95% CI: -1.09 to 0.46, P = 0.42) and SAT (SMD = -0.69, 95% CI: -1.49 to 0.10, P = 0.09) in individuals with overweight or obesity.

CT conducted ≥ 60 min per session significantly reduced VAT (SMD = -1.02, 95% CI: -1.85 to -0.19, P = 0.02, I² = 81%) and SAT (SMD = -0.48, 95% CI: -0.89 to -0.08, P = 0.02, I² = 35%) in individuals with overweight or obesity. However, CT conducted < 60 min per session had no significant effect on VAT (SMD = -0.16, 95% CI: -0.38 to 0.06, P = 0.16, I² = 0%) and SAT (SMD = -0.10, 95% CI: -0.32 to 0.12, P = 0.37, I² = 0%) in individuals with overweight or obesity.

Only aerobic exercise conducted ≥ 30 min per session significantly reduced VAT (SMD = -0.38, 95% CI: -0.69 to -0.06, P = 0.02, I² = 71%) and SAT (SMD = -0.25, 95% CI: -0.44 to -0.06, P = 0.01, I² = 29%) in individuals with overweight or obesity. However, only aerobic exercise conducted < 30 min per session had no significant effect on VAT (SMD = -0.37, 95% CI: -0.80 to 0.05, P = 0.08, I² = 0%) and SAT (SMD = -0.20, 95% CI: -0.62 to 0.22, P = 0.34, I² = 0%) in individuals with overweight or obesity.

Only resistance training conducted ≥ 30 min per session significantly reduced VAT (SMD = -1.10, 95% CI: -2.10 to -0.10, P = 0.03, I² = 85%) but had no significant effect on SAT (SMD = -0.42, 95% CI: -0.86 to 0.03, P = 0.07, I² = 40%) in individuals with overweight or obesity. However, only resistance training conducted < 30 min per session had no significant effect on either VAT (SMD = 0.05, 95% CI: -0.28 to 0.39, P = 0.76, I² = 0%) or SAT (SMD = 0.03, 95% CI: -0.31 to 0.36, P = 0.87, I² = 0%) in individuals with overweight or obesity.

CT conducted > 180 min per week significantly reduced VAT (SMD = -0.48, 95% CI: -0.90 to -0.07, P = 0.02, I² = 71%) and SAT (SMD = -0.24, 95% CI: -0.45 to -0.03, P = 0.02, I² = 0%) in individuals with overweight or obesity. However, CT conducted ≤ 180 min per week had no significant effect on VAT (SMD = -0.29, 95% CI: -0.62 to 0.04, P = 0.08, I² = 0%) and SAT (SMD = -0.16, 95% CI: -0.49 to 0.17, P = 0.35, I² = 0%) in individuals with overweight or obesity.

CT conducted ≥ 12 weeks significantly reduced VAT (12–15 weeks: SMD = -0.30, 95% CI: -0.58 to -0.02, P = 0.04, I² = 21%) (> 15 weeks: SMD = -0.44, 95% CI: -0.83 to -0.04, P = 0.03, I² = 76%) in individuals with overweight or obesity, whereas < 12 weeks had no significant effect on VAT (< 12weeks: SMD = -0.30, 95% CI: -0.87 to 0.28, P = 0.31, I² = 0%). In addition, CT conducted > 15 weeks significantly reduced SAT (> 15 weeks: SMD = -0.33, 95% CI: -0.60 to -0.06, P = 0.02, I² = 51%) in individuals with overweight or obesity, whereas ≤ 15 weeks had no significant effect on SAT (< 12 weeks: SMD = -0.12, 95% CI: -0.69 to 0.45, P = 0.67, I² = 0%) (12–15 weeks: SMD = -0.19, 95% CI: -0.43 to 0.05, P = 0.12, I² = 0%).

All included studies involved concurrent training interventions. We combined aerobic exercise intensity (Moderate, Vigorous, High) with resistance training intensity (Medium-low, High), creating five subgroups: Moderate + Medium-low, Vigorous + Medium-low, Vigorous + High, High + Medium-low, High + High. The Vigorous + Medium-low intensity combination significantly reduced VAT (SMD = -0.30, 95% CI: -0.59 to -0.01, P = 0.04, I² = 40%) in individuals with overweight or obesity, whereas the other concurrent training intensities had no significant effect on VAT (Moderate + Medium-low: SMD = -2.00, 95% CI: -4.13 to 0.14, P = 0.07, I² = 90%) (Vigorous + High: SMD = -0.21, 95% CI: -0.49 to 0.07, P = 0.15, I² = 0%) (High + Medium-low: SMD = -0.44, 95% CI: -1.25 to 0.37, P = 0.29) (High + High: SMD = -0.35, 95% CI: -0.91 to 0.20, P = 0.21, I² = 0%). In addition, five combinations of concurrent training intensities had no significant effect on SAT (Moderate + Medium-low: SMD = -0.45, 95% CI: -0.95 to 0.06, P = 0.08, I² = 0%) (Vigorous + Medium-low: SMD = -0.24, 95% CI: -0.54 to 0.06, P = 0.12, I² = 44%) (Vigorous + High: SMD = -0.27, 95% CI: -0.55 to 0.02, P = 0.06, I² = 0%) (High + Medium-low: SMD = -0.19, 95% CI: -0.99 to 0.61, P = 0.65) (High + High: SMD = -0.26, 95% CI: -0.81 to 0.29, P = 0.35, I² = 0%) in individuals with overweight or obesity.

Among studies involving only female participants, 7 included premenopausal women (aged 15–43 years) and 5 included postmenopausal women. Subgroup analyses showed that CT had a significant effect on VAT (SMD = -0.67, 95% CI: -1.14 to -0.19, P < 0.01, I² = 27%) but no significant effect on SAT (SMD = -0.33, 95% CI: -0.72 to 0.06, P = 0.10, I² = 0%) in postmenopausal women. However, CT had no significant effect on VAT (SMD = -0.40, 95% CI: -0.95 to 0.16, P = 0.16, I² =77%) and SAT (SMD = -0.18, 95% CI: -0.43 to 0.07, P = 0.16, I² = 0%) in premenopausal women.

Publication bias and sensitivity analysis

Visual interpretation of the funnel plot showed an asymmetric distribution for VAT, which was confirmed by Egger’s test (P = 0.022). The funnel plot for SAT showed a symmetric distribution, confirmed by Egger’s test (P = 0.725) (Fig. 4). Furthermore, sensitivity analysis (Fig. 5) indicated that excluding any single study did not significantly alter the overall effect size, demonstrating result stability.

Fig. 4.

Funnel plot of effects of CT on VAT and SAT

Fig. 5.

Sensitivity analysis of effects of CT on VAT and SAT

Evidence certainty

The GRADE assessment is summarized in Table 2. Evidence certainty for VAT and SAT reduction was LOW (⊕⊕○○) due to very serious risk of bias from lack of blinding and poor allocation concealment. However, %ΔV/%ΔS was rated VERY LOW (⊕○○○), as it was further downgraded for serious indirectness because the values were calculated rather than directly reported.

Table 2.

GRADE certainty of evidence

Outcome	Study design	No. of studies	Risk of bias	Inconsistency	Indirectness	Imprecision	Publication bias	Quality of evidence	Effect estimate (95% CI)	Importance
VAT reduction	RCTs	20	Very seriousa	Not serious	Not serious	Not serious	Undetected	⊕⊕○○  LOW	SMD − 0.36 (-0.49 to -0.23)	CRITICAL
SAT reduction			Very seriousa	Not serious	Not serious	Not serious	Undetected	⊕⊕○○ LOW	SMD − 0.19 (-0.29 to -0.09)	CRITICAL
%ΔV/%ΔS		17	Very seriousa	Not serious	Seriousb	Not serious	Undetected	⊕○○○ VERY LOW	β = 0.43 (frequency); β = 1.11 (baseline VAT/SAT)	IMPORTANT

^aVery serious risk of bias: All studies had unavoidable performance bias due to inability to blind exercise interventions, and most inadequately reported allocation concealment

^bSerious indirectness: The values for %ΔV/%ΔS were not directly reported in the original studies but were calculated based on the mean changes of VAT and SAT provided in the research reports, which may introduce potential indirectness risk

Exploratory regression analysis

We calculated %VAT loss, %SAT loss, and %ΔV/%ΔS for each study. The results showed (Table 3) that %VAT loss averaged 10.24 ± 9.94, %SAT loss averaged 5.42 ± 5.44, and %ΔV/%ΔS averaged 1.96 ± 1.18. This indicates that CT effectively reduced both VAT and SAT, with greater reductions in VAT.

Table 3.

Variables included in the multivariate linear regression analysis

Author	BMI	Age	Frequency	Weeks	VAT/SAT	%VAT loss	%SAT loss	%ΔV/%ΔS
Alberga 2015 [57]	34.6	15.5	4	22	0.16	5.10	3.82	1.33
Barone 2009 [62]	29.4	64.6	3	26	0.52	18.37	9.12	2.01
Bouchonville 2014 [58]	36.9	70	3	48	0.69	5.15	2.55	2.02
Cuff 2003 [64]	33.3	63.4	3	16	0.54	10.47	4.69	2.23
Davis 2011 [61]	32.4	15.7	2	16	0.13	5.74	9.73	0.59
Dobrosielski 2012 [60]	33	57	3	26	0.38	5.41	4.58	1.18
Donges 2013 [59]	30.2	46.2	3	12	0.58	12.07	4.48	2.70
Dupuit 2022 [47]	30.3	58.8	3	12	0.97	12.50	3.03	4.13
Fortuin-de Smidt 2020 [51]	34.1	22	4	12	0.17	1.50	0.76	1.98
Hens 2021 (3 month) [48]	32.98	37.64	3	12	0.23	13.58	9.83	1.38
Hens 2021 (6 month) [48]	32.98	37.64	3	24	0.23	13.88	8.50	1.63
Houghton 2017(1) [53]	33	54	3	12	0.47	11.52	22.25	0.52
Houghton 2017(2) [54]	33	51	3	12	0.50	9.74	6.20	1.57
Mendham 2020 [50]	34.1	22	4	12	0.17	1.50	0.76	1.98
Monteiro 2015 [56]	33.17	11.03	3	20	1.41	1.46	-4.47	-0.33
Nono Nankam 2020 [49]	34.1	27.5	4	12	0.26	3.65	1.13	3.22
Park 2003 [63]	25.8	43.4	6	24	0.31	46.13	9.57	4.82
Park 2015 [55]	26.02	57.2	3	12	0.47	16.97	6.68	2.54
Rashti 2019 (HIIT) [52]	28.31	57.11	3	10	0.11	8.24	3.99	2.06
Rashti 2019 (MICT) [52]	30.62	54.1	3	10	0.12	1.85	1.18	1.57

%VAT percentage loss in visceral adipose tissue, %SAT percentage loss in subcutaneous adipose tissue, BMI body mass index

To explore potential factors associated with %ΔV/%ΔS in individuals with overweight or obesity, we initially considered BMI, age, intervention frequency, intervention duration, and baseline VAT/SAT as candidate variables for the regression model (Table 3). However, variables with substantial missing data were excluded from the final analysis.

A total of 20 studies were included in the multiple linear regression analysis. After logarithmic transformation of %ΔV/%ΔS, one study [56] showed an extreme value and was excluded, leaving 19 studies. Box-and-whisker plots identified two additional outliers [53, 61] using the interquartile range (IQR) method (Fig. 6), which were excluded. The remaining 17 data points conformed to a normal distribution by the Shapiro-Wilk test (W = 0.95; P = 0.39).

Fig. 6.

Box-and-whisker plots of ln %ΔV/%ΔS

Using the optimal subset regression method based on BIC, Adjusted R², and Mallow’s Cp (Fig. 7), intervention frequency and baseline VAT/SAT were selected for the final model. The regression results showed that both intervention frequency (β = 0.43, P < 0.01) and baseline VAT/SAT (β = 1.11, P < 0.05) were significantly associated with ln %ΔV/%ΔS. Variance inflation factors (VIF) for both variables were 1.04, indicating no multicollinearity and stable regression coefficients.

Fig. 7.

Optimal Model Scoring System (BIC、Adjusted R²、Mallow’s Cp)

Discussion

Effect of CT on fat mass and its distribution in individuals with overweight or obesity

This study aimed to investigate the effects of concurrent aerobic and resistance training (concurrent training, CT) on fat mass and its distribution in individuals with overweight or obesity. The results suggest that CT may provide potential benefits for reducing fat mass and regulating fat distribution in this population. Next, we identified potential exercise prescription parameters through subgroup analysis of training variables. Finally, exploratory regression analysis revealed that intervention frequency and baseline VAT/SAT were significantly associated with %ΔV/%ΔS. It should be noted that these regression findings represent predictive associations rather than causal effects, and further prospective studies are needed to validate these results.

CT was first proposed by American physiologist Hickson to describe the simultaneous performance of resistance training and aerobic exercise within the same training duration. His research indicated that an “interference effect” emerges during the latter stages (8–10 weeks) of CT, where aerobic exercise leads to a decline in muscle strength [65]. Over the subsequent four decades, most studies have indicated that CT can moderately enhance both muscle strength and cardiorespiratory fitness, though its effectiveness may vary depending on population characteristics and intervention protocols [28, 66, 67]. Some research suggests that well-trained athletes may experience an “interference effect” during CT [68, 69]. However, in individuals with overweight or obesity, CT is generally recognized as a more efficient integrated intervention [70, 71]. By increasing muscle mass to elevate basal metabolic rate and combining aerobic exercise to promote fat oxidation and energy expenditure, CT plays a crucial role in long-term weight management and relapse prevention in individuals with overweight or obesity.

Regular concurrent aerobic and resistance training represents a potential strategy for improving adipose tissue characteristics in individuals with overweight or obesity [56, 72]. Exercise can promote angiogenesis, drive mitochondrial remodeling, and attenuate chronic inflammation-effects that are thought to have a strong association with reduced adipose mass and improved metabolic health [73, 74]. Bouamra et al. [70] found that CT was more effective than resistance training or high-intensity interval exercise alone in improving body weight, BMI, and body fat in sedentary obese adolescents with the same training volume. The potential underlying mechanism is that aerobic exercise increases the secretion of catecholamine and growth hormone by activating the sympathetic-adrenergic system [75, 76], whereas resistance training can elevate resting energy expenditure by increasing muscle mass. All these physiological responses and adaptations further promote lipolysis [77]. The same phenomenon has also been validated in the elderly population [78]. However, the effects of CT on adipose tissue are diverse and even controversial. Silva et al. [79] found that neither resistance training combined with moderate-intensity aerobic exercise nor resistance training combined with high-intensity interval training had a significant effect on reducing body weight and fat mass. Furthermore, a study by Monteiro et al. [56] found that CT did not reduce SAT in obese adolescents, but rather slightly increased it.

From a clinical perspective, the preferential reduction of VAT through CT may hold implications for managing cardiometabolic risk, as VAT is strongly associated with insulin resistance, type 2 diabetes, cardiovascular disease, and metabolic syndrome [80–82]. Our finding that the percentage reduction in VAT was nearly twice that of SAT (%ΔV/%ΔS = 1.96 ± 1.18) suggests a potential capacity for CT interventions to target metabolically active fat depots. However, given the very low certainty of this evidence, these results should be interpreted as preliminary observations rather than definitive clinical proof. In the present study, VAT reduction was more pronounced than SAT reduction following the intervention, a finding that aligns with the preferential VAT loss reported by Chaston et al. [46]. However, VAT also exhibited greater heterogeneity in its response. This discrepancy likely stems from VAT’s unique physiological profile—including higher cellularity, vascular density, innervation, and heightened hormonal sensitivity [83, 84]. While these biological characteristics facilitate the preferential mobilization of VAT during exercise, they also contribute to the increased variability in its reduction across diverse study populations and metabolic states. Additionally, some studies observed greater SAT reduction after CT intervention [53, 61], indicating individual variability in fat distribution response to CT. To address this inconsistency, we conducted subgroup and regression analyses targeting factors that may underlie divergent results, aiming to identify potential moderators of CT intervention efficacy.

Subgroup analysis

BMI

BMI is a practical and widely used marker for identifying and classifying overweight or obesity [85]. Our subgroup analysis suggested that CT was associated with more pronounced reductions in VAT and SAT in individuals with overweight (BMI ≥ 25 to < 30 kg/m²) than in those with obesity (BMI ≥ 30 kg/m²). Adipose tissue is a plastic and metabolically active organ that can respond rapidly to changes in energy excess by increasing the size and number of adipocytes in the early stages of obesity [86]. However, as obesity progresses, adipose tissue function gradually enters a state of decompensation, as evidenced by a sustained elevation of pro-inflammatory adipokines (TNF-α, leptin, resistin, and WNT5A) and a decrease in the expression of anti-inflammatory adipokines (SRFP5 and lipocalin-2), which aggravates chronic inflammation, insulin resistance, and other metabolic abnormalities in the organism [87]. This could potentially diminish the impact of exercise interventions on fat loss and metabolic improvement in individuals with obesity. Therefore, initiating exercise interventions during the overweight stage, before the onset of severe obesity, is likely to yield more substantial reductions in adipose tissue.

Age

Regarding the reduction of adipose tissue by exercise, most of the studies have focused on middle-aged and older adults because fat distribution changes with age. VAT begins to increase around the age of 40 [88], with a more pronounced rise typically occurring around age 60. This age-related redistribution is attributed to behavioral and hormonal changes [89]. Our subgroup results indicated that CT was associated with significant reductions in VAT (age: ≥ 45 years) and SAT (age: ≥ 60 years) in middle-aged and older adults. This may be particularly relevant for public health, as aging is associated with progressive VAT accumulation and increased cardiometabolic risk [90]. Previous studies have identified caloric restriction and physical activity as the most widespread options for reducing adipose tissue [91]. Notably, among sedentary middle-aged and older adults who lost 10% of body weight, exercise was associated with a significantly greater reduction in VAT compared to caloric restriction alone [92]. This suggests that exercise may promote a healthier fat distribution in this population. Regular physical activity has been found to reduce leptin levels in overweight or obese middle-aged and older adults (> 45 years) [93], thereby restoring fat metabolism. In addition, aerobic exercise and resistance training have independent effects on fat reduction, so the combination of the two can have the added effect of reducing the adipose dysfunction associated with aging and obesity [94]. Our study came to a similar conclusion that CT seems to resist and even reverse the abnormalities in fat distribution due to aging. Moreover, based on our results, the older the age, the more pronounced the reduction in adipose tissue appeared to be, especially VAT.

Geographic affiliation (by Continent)

There are significant differences in fat distribution among individuals of different ethnicities, influenced by a combination of genetic, environmental, economic, and lifestyle factors. Populations in Asia, Europe, and North America show a predominant tendency toward visceral fat accumulation, while individuals of African descent tend to have a higher proportion of subcutaneous fat [95, 96]. Our subgroup analysis suggested that CT was associated with significant reductions in VAT in individuals with overweight or obesity in Asia and North America, as well as SAT in North American individuals. Notably, fat reduction was most pronounced in North America, which might be related to the higher baseline BMI and adiposity levels observed in participants from this region [97]. Our regression analysis also identified the baseline VAT/SAT as a potential key predictor of change in fat distribution, which may partially explain why CT was observed to significantly reduce VAT in the North American individuals. Furthermore, the effects of the same exercise intervention on lipid metabolism exhibit substantial inter-individual heterogeneity. Since our subgroup analysis was stratified by region rather than by specific ethnicity or country, these results should be interpreted with caution.

Frequency of intervention

To examine the influence of intervention frequency on adipose tissue in individuals with overweight or obesity, we conducted a corresponding subgroup analysis. The results suggested that a frequency of ≥ 3 sessions per week was associated with significant reductions in both VAT and SAT, whereas a frequency of < 3 sessions per week showed no significant association with fat reduction. The majority of studies also provide support for the observation that performing exercise at least 3 sessions per week can significantly reduce adipose tissue [35, 98, 99]. Moreover, performing CT 5 sessions per week has been found more beneficial for weight loss and fat reduction compared to aerobic or resistance training alone [100]. Relevant guidelines also recommend 3 to 5 exercise sessions per week for optimal outcomes in this population [101]. Furthermore, some studies suggest that exercising more frequently than three sessions per week helps establish a habitual exercise routine and promotes physiological adaptations, both of which are conducive to fat loss [102]. However, it should be noted that only one study in our analysis employed an intervention frequency of < 3 sessions per week, which may introduce bias. Consequently, caution is warranted when referring to our findings regarding the intervention frequency.

Exercise duration per session (min)

Regarding the duration of intervention per session, our subgroup analysis suggested that intervention conducted ≥ 60 min per session was associated with significant reductions in VAT and SAT, whereas intervention conducted < 60 min per session showed no significant association with fat reduction. The body produces energy during prolonged exercise mainly through aerobic metabolism. Although muscle relies on anaerobic metabolism for the first 1–2 min, aerobic metabolism is then dominant, in which muscle is able to fully mobilize fat for energy supply [103]. As the duration of exercise increases, the substrate utilization ratio shifts, characterized by an increased rate of fat oxidation and a decreased rate of carbohydrate oxidation [104]. A study involving women [105] reported that a regimen of 30 min of resistance training followed by 30 min of cycling significantly reduced fat mass. In our subgroup analyses, we also observed significant reductions in VAT and SAT with aerobic sessions alone ≥ 30 min, and VAT with resistance training sessions alone ≥ 30 min. Therefore, our results suggest that CT consisting of ≥ 30 min of aerobic exercise and ≥ 30 min of resistance training (i.e., a total duration of ≥ 60 min per session) may be more beneficial for reducing adipose tissue in individuals with overweight or obesity.

Exercise duration per week (min)

Next, we calculated total intervention duration per week for each study by the frequency of intervention per week and duration of intervention per session to provide new ideas for developing exercise prescriptions. WHO recommends that adults perform 150–300 min of moderate-intensity aerobic exercise or 75–150 min of vigorous-intensity aerobic exercise, as well as 2–3 sessions of resistance training per week [106]. Our subgroup results suggest that intervention conducted > 180 min per week was associated with significant reductions in VAT and SAT in individuals with overweight or obesity. In general, sustained physical activity of more than 150 min per week can result in a modest weight loss of 2–3 kg in individuals with overweight or obesity [16]. Jayedi et al. [107] also demonstrated that clinically significant reductions in adiposity can be achieved by performing > 150 min of exercise per week. In addition, Friedenreich et al. [108] found a dose-effect relationship between duration of intervention per session and adiposity, with increased duration of intervention per session being more beneficial for reducing fat and other obesity-related markers. Therefore, increasing frequency of intervention and prolonging duration of intervention per session may be a more effective approach to fat reduction in individuals with overweight or obesity. However, considering that excessive exercise volume may lead to a plateau effect and overtraining risks, controlling the weekly duration within a reasonable range (aligning with WHO recommendations) represents the optimal balance between efficacy and safety.

Duration of intervention

Duration of intervention is a key variable that influences weight loss outcomes. In this study, we categorized the included studies into 3 subgroups based on the intervention duration: < 12 weeks, 12–15 weeks, and > 15 weeks. Our subgroup analyses suggested that interventions lasting ≥ 12 weeks were associated with significant reductions in VAT, and there was a tendency toward more pronounced VAT reduction as the intervention duration increased. Interventions > 15 weeks showed a potential to reduce SAT. This result is consistent with our regression analysis, which suggested that VAT might be preferentially mobilized compared to SAT. The reason for the preferential and selective reduction of VAT after short-term interventions may be because it is metabolically more active and sensitive to lipolysis than SAT [89]. A systematic review demonstrated that VAT is preferentially reduced in moderate weight loss [46]. As the duration of the intervention increases, the body gradually adapts to stress hormones, and SAT undergoes gradual remodeling. A cross-sectional study [24] found that overweight or obese individuals who participated in exercise for more than 2 years had higher capillary density and fewer pro-inflammatory macrophages in their SAT, as well as greater angiogenic capacity and lipid storage compared with those who did not exercise. Therefore, fat loss after CT may be dominated by visceral fat depots gradually shifting to subcutaneous fat depots.

Intensity of intervention

Subgroup analyses of intervention intensity suggested that vigorous-intensity aerobic exercise combined with medium- to low-intensity resistance training was associated with significant reductions in VAT in individuals with overweight or obesity. The majority of studies have confirmed that vigorous-intensity aerobic exercise not only significantly reduces adipose tissue but also is more effective in reducing VAT [109, 110]. Previous studies have shown that the secretion of lipolytic hormones during aerobic exercise is positively correlated with the intensity of intervention [111]. Zhu et al. [112] found that 12 weeks of vigorous-intensity aerobic exercise promoted the release of epinephrine (EPI) and growth hormone (GH) at a higher rate than did moderate-intensity continuous exercise in obese women. Therefore, the reduction of adipose tissue induced by vigorous-intensity aerobic exercise can be partially attributed to the changes in hormone secretion. In addition, following vigorous-intensity exercise, the body exhibits a prolonged elevation in energy expenditure—a physiological phenomenon termed excess post-exercise oxygen consumption (EPOC). Notably, fat serves as the primary substrate during the EPOC period, thereby enhancing the efficiency of post-exercise fat oxidation [113]. This also explains to some extent that vigorous-intensity aerobic exercise is superior to moderate-intensity aerobic exercise for fat reduction.

On the other hand, maintaining muscle strength and mass is essential for reducing body fat. Studies [114] have shown that medium- and low-intensity resistance training results in higher energy expenditure during exercise than high-intensity resistance training with the same exercise duration. In contrast, when accounting for both exercise bouts and subsequent post-exercise recovery periods (expending up to 72 h), high-intensity resistance training leads to greater total energy expenditure—primarily attributable to elevated energy consumption during the recovery phased. However, the present study did not find that combining vigorous-intensity aerobic exercise with high-intensity resistance training yielded optimal outcomes for fat mass reduction. This may be due to the inclusion of only a limited number of studies examining this specific intensity combination in our meta-analysis, which could potentially introduce bias. Furthermore, long-term adherence to high-intensity resistance training protocols tends to be relatively low; more importantly, implementing such programs in overweight and obese populations with no prior exercise experience may elevate the risk of exercise-related injury. Consequently, the present study recommends the combination of vigorous-intensity aerobic exercise with moderate- and low-intensity resistance training as the optimal intensity regimen for reducing fat mass and regulating fat distribution. Nevertheless, it should be acknowledged that our detailed categorization into several intensity subgroups inevitably results in a relatively small number of trials within each specific category. Therefore, while these findings provide valuable insights for refining exercise prescriptions, they should be interpreted with caution due to the limited sample size per subgroup.

Women menopausal status

Changes in estrogens [Estrone (E1), Estradiol (E2), and Estriol (E3)] are a significant physiological phenomenon in postmenopausal women, and an increase in VAT is associated with a decrease in E2 levels [115]. Decreased concentrations of E2 in the circulation of postmenopausal women increase the ratio of androgens to estrogens (testosterone/estradiol) leading to redistribution of body fat [116]. In women, VAT typically increases from 5 to 8% of total body fat before menopause to 15–20% after menopause, often accompanied by a loss of muscle mass [117]. Thus, disturbances in visceral fat accumulation and distribution partly explain the higher prevalence of obesity and risk of cardiovascular disease in postmenopausal women. Given the substantial proportion of women’s lifespan spent in the postmenopausal state and the global aging of the female population, effective interventions targeting visceral adiposity in postmenopausal women have substantial implications for reducing the population-level burden of cardiometabolic diseases [118, 119]. Our subgroup analysis suggested that CT was associated with a significant reduction in VAT in postmenopausal women; however, no significant effect was observed on SAT, nor were significant changes found in premenopausal women. It was found that postmenopausal women who lost > 2 kg of fat following CT showed significantly increased levels of circulating sex hormone-binding globulin (SHBG), which can further regulate hormone production and bioavailability [120]. In addition, resistance exercise has been shown to stimulate androgen production, helping to prevent menopause-related muscle loss and even promote increases in muscle mass and strength [121]. A meta-analysis [122] showed that CT was effective in reducing VAT in postmenopausal women, which is consistent with the results of our subgroup analysis, and that it was more beneficial in promoting overall health in postmenopausal women compared to resistance training or aerobic exercise alone.

Limitation

This study has several limitations. First, the high risk of performance bias (due to the inability to blind participants and personnel in exercise interventions) and the inadequate reporting of allocation concealment both contributed to the downgrading of the certainty of evidence in our GRADE assessment. Second, the regression analysis may be subject to ecological bias due to the limited number of studies (n = 17) and the use of study-level data. Consequently, these exploratory findings should be interpreted with caution. Third, most included studies lacked strict dietary control, which may have confounded the fat loss outcomes and hindered the assessment of the independent effects of concurrent training.

Conclusions

Current evidence suggests that CT can effectively reduce fat mass and may preferentially decrease VAT compared to a sedentary lifestyle in individuals with overweight or obesity. Our findings support CT as a promising and scalable strategy for population-level obesity management, particularly for high-risk groups such as postmenopausal women and older adults. Furthermore, the identified exercise prescription parameters provide practical guidance for populations aiming to reduce adipose tissue.

Abbreviations

%ΔV/%ΔS

Percentage ratio of visceral/subcutaneous fat loss

AE

Aerobic exercise

BIC

Bayesian Information Criterion

BMI

Body mass index

CI

Confidence interval

CT

Concurrent training

EPOC

Excess post-exercise oxygen consumption

IQR

Interquartile range

RCTs

Randomized controlled trials

RT

Resistance training

SAT

Subcutaneous adipose tissue

SMD

Standardized mean difference

VAT

Visceral adipose tissue

VIF

Variance inflation factor

WHO

World Health Organization

Author contributions

XL and GL conducted the literature search, study selection, data extraction, and quality assessment. SW and JX contributed to the conception and design. XL took the lead in drafting and writing the manuscript, with guidance from SW. GL and JX conducted the meta-analysis and linear regression. All authors read and approved the final manuscript.

Funding

This study was supported by the Students Scientific Research Project of South China Normal University (25TKGA17).

Data availability

All data used in the analyses are available on request to the corresponding author.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.