36-Week personalized resistance training improves muscle function and circulating myokines in older women with possible sarcopenic obesity: a randomized clinical trial

Abstract

Background

This study aims to explore the effects of personalized resistance training on muscle function and serum myokines in possible sarcopenic obese older women.

Methods

25 possible SO older women were randomly divided into exercise group (n = 13) and control group (n = 12). According to the FITT-VP principle, personalized resistance training programs were developed based on their individual health differences. The training was implemented in three stages with gradually increasing intensity from low to moderate. Each session lasted 40 min, 3 times a week, for a total of 36 weeks. Body composition, muscle function, and serum myokines (irisin, IGF-1, and myostatin) were tested before and after the intervention in both groups.

Results

The intervention yielded significant between-group improvements in body fat percentage (P < 0.05, Hedges’ g = 1.42), ASMI (P < 0.01, Hedges’ g = 1.79), HGS (P < 0.01, Hedges’ g = 1.91), 6-metre walk speed (P < 0.01, Hedges’ g = 1.36), five times sit-to-stand performance (P < 0.05, Hedges’ g = 1.08), and TUG (P < 0.01, Hedges’ g = 1.20). After 36 weeks, the exercise group also showed higher IGF-1 (P < 0.05, Hedges’ g = 1.46) and lower MSTN (P < 0.01, Hedges’ g = 1.02) levels compared to controls.

Conclusion

The 36-week personalized resistance training program results in elevated serum IGF-1 concentrations and reduced myostatin levels, alongside independent improvements in muscle strength and muscle function among older women with probable sarcopenic obesity.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12877-025-06355-0.

Introduction

In 2022, the European Society for Clinical Nutrition and Metabolism and European Association for the Study of Obesity defined sarcopenic obesity (SO) as the coexistence of excess adiposity and reduced muscle mass/function [1]. Driven by aging, sedentarism, poor nutrition, chronic inflammation, and metabolic dysregulation, SO synergistically elevates risks of frailty, cardiometabolic diseases, and mortality by 1.5-3 fold compared to isolated sarcopenia or obesity [2–7]. Globally affecting 5–10% of adults, it demonstrates sex-specific patterns—women exhibit 30% higher incidence linked to postmenopausal hormonal changes [2, 8]. Current guidelines recommend dual-energy X-ray absorptiometry and bioelectrical impedance analysis for early detection, alongside resistance training, protein optimization, and anti-inflammatory strategies [8].

Resistance exercise emerges as a cornerstone therapeutic strategy, demonstrating multi-modal benefits through satellite cell activation, enhanced myoprotein synthesis, and suppression of proteolytic pathways – collectively promoting skeletal muscle hypertrophy [9]. Clinical trials further substantiate its efficacy in ameliorating body composition parameters, restoring metabolic homeostasis, and modulating age-associated inflammatory cascades in SO patients [10, 11]. While conventional high-intensity resistance training optimizes mechanical tension for muscle protein anabolism, its cardiovascular and orthopedic risks render it suboptimal for older people. This necessitates the development of safer loading paradigms that balance efficacy with geriatric physiological tolerability [12]. Contemporary exercise science underscores the limitations of generic training recommendations, particularly given interindividual variability in baseline fitness, metabolic adaptability, and age-related physiological constraints. Precision exercise prescription – incorporating FITT-VP principles (Frequency, Intensity, Time, Type, Volume, Progression) – has consequently gained prominence as a paradigm shift toward personalized therapeutic exercise [13, 14]. This study examines underinvestigated exercise responses in older women with possible sarcopenic obesity through personalized home-based resistance training, aiming to isolate mechanistic effects and establish targeted prevention protocols.

Materials and methods

This study employed a randomized clinical trial design and strictly adhered to the Consolidated Standards of Reporting Trials (CONSORT) guidelines throughout the research process [15].

Participants and selection criteria

Community-dwelling older women presenting with possible sarcopenic obesity were recruited through stratified random sampling from Haidian District, Beijing, China. Following rigorous screening against inclusion/exclusion criteria, 30 eligible participants underwent block randomization (1:1 ratio) to either an exercise group (n = 15) or a control group (n = 15). After being thoroughly informed about the objectives of the study, all participants provided their informed consent.

Eligible participants were required to meet the following composite criteria: (1) female sex (60–70 years, ≥ 1 year postmenopausal), (2) Body mass index (BMI) ≥ 24 kg/m² and possible sarcopenic obesity defined by AWGS 2019 criteria [16] – specifically diminished muscle strength (dominant-hand grip < 18 kg) and impaired physical function (5-chair stand > 12 s) with preserved appendicular skeletal muscle mass index (ASMI) (BIA-ASMI ≥ 5.7 kg/m²), (3) Non-exercisers (< 3 sessions/week, < 20 min/session) for ≥ 6 months pre-study, (4) absence of disabling conditions or major comorbidities (cardiovascular diseases, diabetes, neurodegenerative disorders), (5) physician-cleared exercise readiness via Physical Activity Readiness Questionnaire PAR-Q assessment [17].

Participants were excluded if presenting with active arthritic conditions, recent musculoskeletal injuries (≤ 3 months), anticipated protocol non-adherence (< 80% compliance), concurrent enrollment in structured exercise programs, or participation in conflicting clinical trials during the study period.

The study was approved by the Ethics Committee of Sports Science Experiments at Beijing Sport University (Approval number: 2021093 H) and registered in the Chinese Clinical Trial Registry (Registration Number: ChiCTR2300069599, Registration Date: 21/03/2023).

Individualized exercise prescription

This study implemented FITT-VP-guided resistance training protocols. Certified exercise specialists conducted pre-intervention assessments of physical function and body composition, integrating demographic, clinical, and anthropometric data to develop personalized regimens. Participants completed structured skill-acquisition phases under supervision to ensure movement proficiency before formal training.

The individualized exercise prescription incorporated six FITT-VP components: (1) Mode: Progressive elastic band resistance training combined with bodyweight exercises (standing calf raises, posterior leg swings, seated leg raises, elastic band chest press, elastic band bicep curls, elastic band leg presses, and elastic band knee lifts); (2) Intensity: To ensure precise implementation of personalized exercise prescriptions, our intervention employed a dual-intensity monitoring system integrating mechanical loading parameters and physiological feedback. Resistance training protocols were standardized with progressive overload achieved through color-coded elastic bands (blue: 18 lbs, pink: 23 lbs, green: 28 lbs), requiring participants to perform 10–12 repetitions per exercise across 2–3 sets with 2–3 min interset recovery. The biomechanical loading was dynamically adjusted through band color selection based on real-time cardiovascular monitoring using validated wearable technology (OPPO Smart Wristband), maintaining exercise intensity within target heart rate zones (THRz = 60–80% HRmax, calculated as 220 - age). This multimodal monitoring paradigm ensures: (a) dose-controlled mechanical stimulus through repetition-set-band resistance triadic regulation; (b) physiological fidelity via continuous THRz compliance verification; (3) Progression: Participants progressed through three structured training phases using a color-coded elastic band system (blue: 18 lbs, pink: 23 lbs, green: 28 lbs) calibrated to maintain target intensity thresholds: Weeks 1–2 (Neuromuscular adaptation): Movement pattern familiarization at 60% HRmax; Weeks 3–7 (Progressive overload): Systematic resistance escalation until sustained 60–80% HRmax; Week 8–36 (Autoregulated consolidation): Self-regulated intensity selection based on acquired exercise competence, and it is necessary to ensure that the exercise intensity is maintained within 60–80% HRmax; (4) Time: 40-min sessions (5-min warm-up, 30-min target training, 5-min cool-down); (5) Frequency: 3 supervised sessions/week; (6) Volume: Adaptation phase (2 sets ×10–12 reps, 2–3 min recovery) progressing to intensification/consolidation phases (3 sets ×10–12 reps), sustained over 36 weeks (Fig. 1).

Fig. 1.

Experimental protocol

The individualized exercise prescription incorporated six FITT-VP components (see Fig. 1):

1. Mode: Progressive elastic band resistance training combined with bodyweight exercises (standing calf raises, posterior leg swings, seated leg raises, elastic band chest press, elastic band bicep curls, elastic band leg presses, and elastic band knee lifts).

2. Intensity: To ensure precise implementation of personalized exercise prescriptions, our intervention employed a dual-intensity monitoring system integrating mechanical loading parameters and physiological feedback. Resistance training protocols were standardized with progressive overload achieved through color-coded elastic bands (blue: 18 lbs, pink: 23 lbs, green: 28 lbs), requiring participants to perform 10–12 repetitions per exercise across 2–3 sets with 2–3 min interset recovery. The biomechanical loading was dynamically adjusted through band color selection based on real-time cardiovascular monitoring using validated wearable technology (OPPO Smart Wristband), maintaining exercise intensity within target heart rate zones (THRz = 60–80% HRmax, calculated as 220 - age). To clarify the mechanism of resistance adjustment based on heart rate feedback: Prior to the exercise, participants were provided with a personalized elastic band selection plan that correlated with the target heart rate zones. For instance, participants with relatively better physical conditions might start with the blue elastic band (18 lbs). If their heart rate consistently remained below the lower limit of the THRz (60% HRmax) after a 5–10 min adaptation period during the exercise, they were instructed to first check the proper wearing of the wristband to ensure accurate heart rate monitoring. If the wristband was worn correctly and the heart rate was still low, they were advised to increase the movement speed or amplitude slightly. If, after this adjustment, the heart rate still did not reach the lower limit of the THRz within another 5–10 min, they were recommended to switch to the pink elastic band (23 lbs) to increase the exercise intensity. Conversely, if participants felt excessive fatigue, pain, or other discomforts after switching to a higher - resistance band, they were instructed to stop the exercise immediately and contact the research team. This multimodal monitoring paradigm ensures: (a) dose - controlled mechanical stimulus through repetition - set - band resistance triadic regulation; (b) physiological fidelity via continuous THRz compliance verification.

3. Progression: Participants progressed through three structured training phases using a color - coded elastic band system (blue: 18 lbs, pink: 23 lbs, green: 28 lbs) calibrated to maintain target intensity thresholds: Weeks 1 - 2 (Neuromuscular adaptation): Movement pattern familiarization at 60% HRmax; Weeks 3 - 7 (Progressive overload): Systematic resistance escalation until sustained 60 - 80% HRmax; Week 8 - 36 (Autoregulated consolidation): Self - regulated intensity selection based on acquired exercise competence, and it is necessary to ensure that the exercise intensity is maintained within 60 - 80% HRmax.

4. Time: 40 - min sessions (5 - min warm - up, 30 - min target training, 5 - min cool - down).

5. Frequency: 3 supervised sessions/week.

6. Volume: Adaptation phase (2 sets × 10–12 reps, 2–3 min recovery) progressing to intensification/consolidation phases (3 sets × 10–12 reps), sustained over 36 weeks.

Outcome measures

Dietary status

All subjects are advised to adhere to their current dietary intake and habits throughout the study period. A three-day estimated food record and a 72-hour recall were employed to track and calculate the energy intake in both groups before and after the intervention. Food energy was analyzed using the “BoHe” application, a professional team that offers diet analysis and photo identification of food calories.

Body composition

BMI was calculated using the standard equation: BMI = body weight (kg)/height² (m²). Body composition profiling, including muscle mass (appendicular lean mass and whole body skeletal muscle mass) and obesity-related indicators (body fat percentage), was performed through multi-frequency BIA using a medically validated impedance plethysmograph (InBody 720; Biospace Co., Seoul, Korea). Appendicular skeletal muscle mass index (ASMI) was computed with the following formula: ASMI (kg/m²) = appendicular lean mass (kg)/height² (m²) [16].

Measurement requirements: Empty bladder after 2 h of fasting or eating, no metal objects on the body, rest quietly for more than 5 min, and adopt a sitting position to test the subjects. The testers were professionally trained and the test was performed by one person to reduce errors [18].

Muscle function

Prior to the formal tests, all participants underwent comprehensive training to ensure they fully understood and mastered the basic procedures of the tests. The training included a detailed explanation of the test objectives, procedures, and specific requirements for each step. Participants were also provided with live demonstrations and opportunities for hands-on practice until they demonstrated proficiency.

Handgrip strength (HGS)

An electronic spring-type dynamometer (Jianmin electronic grip meter) was used to evaluate handgrip strength. In accordance with the HGS measurement protocol outlined by AWGS 2019 [16], the participant was instructed to assume a standing position, with the shoulder aligned with the torso, the elbow fully extended, and the wrist in a neutral position (0 degrees). The maximum reading from two trials was recorded, utilizing the dominant hand during a maximum-effort isometric contraction.

Six-meter walking speed

As AWGS 2019 recommended [16], each participant was asked to walk 6 m at their daily gait speed from a moving start without deceleration. This walk would be timed by a researcher using a manual stopwatch, and the gait speed was calculated using the formula: Gait speed = Walking distance (meters)/Walking time (seconds). Each participant completed two tests, and the mean result from these two assessments was documented as their final gait speed.

Five times sit-to-stand

Participants were instructed to perform five sit-to-stand transitions as quickly as possible within a specified time frame. They were seated on a hard chair without armrests or a backrest, facing forward with their hands placed on the opposite shoulders. To assess the time taken for these five transitions, participants folded their arms across their chest and executed the chair stands in rapid succession [19]. Each participant underwent two trials, and the average result of the two tests was taken as their time of five times sit-to-stand.

Timed Up and Go test (TUG)

The TUG assessed the time it took for participants to rise from a chair, walk at a normal pace to a marked line on the floor located 3 m away, turn around, and then walk back to the chair and sit down [20]. Each participant repeated the test twice, and the average of the two trials was taken as the final result.

Blood sampling and analysis

Following a standardized 12-hour overnight fast pre- and post-intervention, antecubital venous blood samples (5 mL) were collected under fasting conditions. Whole blood was immediately processed through refrigerated centrifugation (4 °C, 3500 ×g, 15 min; Thermo Scientific ST 16R). Resultant serum aliquots were cryopreserved at −80 °C in polypropylene cryovials until batch analysis. Serum concentrations of metabolic myokines (irisin, IGF-1) and the atrophy-associated factor myostatin (MSTN) were quantified using commercially available ELISA kits (Jianglai Biotechnology, Shanghai, China) according to manufacturer’s protocols, with optical density determinations performed on a BioTek SYNERGY H1 microplate reader (Agilent Technologies).

Statistical analysis

Statistical analysis of the data was conducted using SPSS 26.0, and all data are presented as mean ± standard deviation. Normality of data was verified using the Shapiro-Wilk test. Where normality assumptions were violated, distributional characteristics were further assessed through skewness and kurtosis analyses. For normally distributed data, repeated-measures ANOVA was employed to examine within-group and between-group differences across time points. Mauchly’s test of sphericity first determined the covariance structure validity, with Greenhouse-Geisser correction applied when sphericity assumptions were violated to control Type I error inflation. Significant main effects or interactions were subsequently analyzed using Bonferroni-adjusted post hoc comparisons. Effect size calculations employed Cohen’s d (within-group pre-post changes) and Hedges’ g (between-group change score differences). The significance level was set at P < 0.05.

Results

Baseline characteristics

Two participants from the exercise group and three from the control group were unable to complete the experiment. Consequently, the final data analysis included 12 members from the control group and 13 members from the exercise group. As indicated in Table 1, there were no statistically significant differences in age, height, weight, or BMI between the two groups (P > 0.05).

Table 1.

Baseline characteristics of participants

	n	Age (years)	Height (cm)	Weight (kg)	BMI(kg/m2)
Control group	12	67.17 ± 2.89	158.18 ± 2.42	62.21 ± 8.11	25.71 ± 2.31
Exercise group	13	65.69 ± 3.68	156.10 ± 4.10	63.58 ± 6.83	26.16 ± 2.48

Macronutrient and energy intake

As shown in Table 2, there were no significant differences in daily intake of carbohydrate (P = 0.986), fat (P = 0.855), protein (P = 0.902), and total energy intake (P = 0.370) between the two groups.

Table 2.

Macronutrient and energy intake before and after intervention

	Group	Mean (SD)		△	P-value
		Pre-intervention	Post-intervention
Carbohydrate (g/d)	Control	169.84 ± 22.66	171.09 ± 36.18	1.25	Group = 0.197 Time = 0.866 Group×time = 0.986
	Exercise	156.80 ± 31.70	157.83 ± 26.95	1.02
Fat (g/d)	Control	47.65 ± 7.96	47.89 ± 12.66	0.24	Group = 0.043 Time = 0.799 Group×time = 0.855
	Exercise	41.63 ± 9.92	43.09 ± 10.40	1.45
Protein (g/d)	Control	47.41 ± 4.90	46.95 ± 5.30	−0.46	Group = 0.930 Time = 0.551 Group×time = 0.902
	Exercise	47.70 ± 5.95	47.00 ± 5.44	−0.70
Total energy intake (kcal/d)	Control	1300.86 ± 179.69	1266.58 ± 188.94	−32.06	Group = 0.301 Time = 0.915 Group×time = 0.370
	Exercise	1233.15 ± 89.80	1260.49 ± 97.63	25.28

Body composition

As shown in Table 3; Fig. 2, no significant group × time interaction was observed considering BMI (P = 0.926) and total muscle mass (P = 0.215). In the analysis of body fat percentage, the exercise group had lower values than the control group after 36-week intervention (P < 0.05, Hedges’ g = 1.42). Within the exercise group, body fat percentage decreased significantly by 4.10 ± 3.52% from pre- to post-intervention (P < 0.01, Cohen’s d = 0.98). In the analysis of ASMI, the exercise group had higher values than the control group after 36-week intervention (P < 0.01, Hedges’ g = 1.79). Within the exercise group analysis, there was an increase between pre-intervention and post-intervention of 0.50 ± 0.35 kg/m² (P < 0.01, Cohen’s d = 1.15).

Table 3.

Changes in body composition and muscle function profiles before and after intervention

Variable	Group	Mean (SD)		P-value	Within-Group		Between-Group
		Pre-intervention	Post-intervention		Mean Change (95% CI)	Effect Size (Cohen’d)	Effect Size (Hedges’ g)
Body fat percentage (%)	Control	38.48±3.49	38.33±3.02	Group = 0.386 Time = 0.001 Group×time = 0.002	-0.16 (-1.2 to 0.88)	0.04	1.42g
	Exercise	39.18±4.18	35.08±4.44**#		-4.10 (-6.23 to -1.97)	0.98c
BMI (kg/m2)	Control	25.71±2.31	25.02±2.10	Group = 0.626 Time = 0.003 Group×time = 0.926	-0.69 (-1.32 to -0.05)	0.24a	0.04
	Exercise	26.16±2.48	25.51±2.85		-0.65 (-1.25 to -0.05)	0.26a
Total muscle mass (kg)	Control	20.61±1.90	20.85±2.51	Group = 0.475 Time = 0.062 Group×time = 0.215	0.24 (-1.04 to 1.52)	0.11	0.51f
	Exercise	20.73±2.02	21.88±2.18		1.14 (0.24 to 2.05)	0.57b
ASMI (kg/m2)	Control	6.14±0.53	6.04±0.61	Group = 0.084 Time = 0.006 Group×time < 0.001	-0.09 (-0.29 to 0.11)	0.19	1.79g
	Exercise	6.20±0.43	6.69±0.50**##		0.50 (0.28 to 0.71)	1.15c
HGS  (kg)	Control	18.99±2.07	16.83±2.11**	Group = 0.098 Time = 0.787 Group×time < 0.001	-2.16 (-3.25 to -1.07)	0.61b	1.91g
	Exercise	18.73±3.07	20.65±3.54**##		1.92 (0.43 to 3.42)	0.63b
6-metre walking speed  (m/s)	Control	0.97±0.18	0.88±0.07	Group = 0.247 Time = 0.309 Group×time = 0.003	-0.05 (-0.13 to 0.02)	0.44a	1.36g
	Exercise	0.90±0.08	1.00±0.12**##		0.10 (0.04 to 0.16)	0.95c
five times sit-to-stand (s)	Control	12.97±1.35	13.18±1.85	Group = 0.329 Time = 0.031 Group×time = 0.012	0.21 (-1.04 to 1.47)	0.11	1.08g
	Exercise	13.67±2.14	11.14±2.51**#		-2.52 (-4.3 to -0.75)	0.86c
TUG (s)	Control	9.91±1.76	10.47±1.24	Group = 0.126 Time = 0.196 Group×time = 0.007	0.56 (-0.44 to 1.56)	0.36a	1.20g
	Exercise	10.11±1.66	8.65±1.39**##		-1.46 (-2.54 to -0.38)	0.82c

* P < 0.05, ** P < 0.01, post-intervention, compared with pre-intervention

^#P< 0.05,^##P< 0.01, comparisons between the exercise and control groups at corresponding time points

Cohen’s d of 0.2, 0.5, 0.8 represents small, medium, and large effect size, a: small effect size, b: medium effect size, c: large effect size

Hedges’g of 0.2, 0.5, 0.8 represents small, medium, and large effect size, e: small effect size, f: medium effect size, g: large effect size

Fig. 2.

Changes in body fat percentage (BF) (a), BMI (b), total muscle mass (c) and ASMI (d) before and after 36-week intervention. **P < 0.01, post-intervention, compared with pre-intervention;^#P < 0.05,^##P < 0.01, comparisons between the exercise and control groups at corresponding time points

Muscle function

As shown in Table 3; Fig. 3, the exercise group demonstrated significantly greater improvement in HGS compared to controls post-intervention (P < 0.01, Hedges’ g = 1.91). Within the exercise group analysis, HGS increased significantly by 1.92 ± 2.47 kg from pre- to post-intervention (P < 0.01, Cohen’s d = 0.63).

Fig. 3.

Changes in handgrip strength (HGS) (a), 6-metre walking speed (b), five times sit-to-stand (c) and TUG (d) before and after 36-week intervention. ** P < 0.01, post-intervention, compared with pre-intervention; ^# P < 0.05, ^## P < 0.01, comparisons between the exercise and control groups at corresponding time points

The results showed significant improvement between the study groups with respect to 6-metre walking speed after performing the intervention (P < 0.01, Hedges’ g = 1.36). Within the exercise group analysis, walking speed increased significantly by 0.10 ± 0.10 m/s from pre- to post-intervention (P < 0.01, Cohen’s d = 0.95).

In the analysis of five times sit-to-stand, the exercise group demonstrated significantly shorter completion times than controls post-intervention (P < 0.05, Hedges’ g = 1.08). Within the exercise group analysis, completion time decreased significantly by 2.52 ± 2.93 s from pre- to post-intervention (P < 0.01, Cohen’s d = 0.86).

In the analysis of TUG, the exercise group showed greater improvement than the control group post-intervention (P < 0.01, Hedges’ g = 1.20). Within the exercise group analysis, TUG test time decreased significantly by 1.46 ± 1.79 s from pre- to post-intervention (P < 0.01, Cohen’s d = 0.82).

Serum myokines

As it is shown in Table 4; Fig. 4, following the intervention, no significant group × time interaction was observed for irisin (P = 0.056). In the analysis of IGF-1, the exercise group showed significantly higher post-intervention levels than the control group (P < 0.05, Hedges’ g = 1.46). Within the exercise group analysis, IGF-1 concentrations increased significantly by 7.73 ± 8.19 ng/mL from pre- to post-intervention (P < 0.05, Cohen’s d = 0.94). In the analysis of MSTN, the exercise group showed significantly lower post-intervention levels than the control group (P < 0.01, Hedges’ g = 1.02). Within the exercise group analysis, MSTN concentrations decreased significantly by 0.49 ± 0.45 ng/mL from pre- to post-intervention (P < 0.01, Cohen’s d = 1.09).

Table 4.

Changes in serum myokines profiles before and after intervention

Variable	Group	Mean (SD)		P-value	Within-Group		Between-Group
		Pre-intervention	Post-intervention		Mean Change (95% CI)	Effect Size (Cohen’d)	Effect Size (Hedges’ g)
Irisin (ng/mL)	Control	153.74 ± 17.24	152.61 ± 13.95	Group = 0.171 Time = 0.094 Group×time = 0.056	−1.14 (−11.47 to 9.2)	0.07	0.81g
	Exercise	153.15 ± 19.88	169.21 ± 18.92		16.06 (0.88 to 31.24)	0.64b
IGF-1 (ng/mL)	Control	62.71 ± 5.32	56.02 ± 9.47*	Group = 0.367 Time = 0.796 Group×time = 0.001	−6.69 (−13.94 to 0.55)	0.59b	1.46g
	Exercise	57.56 ± 7.12	65.29 ± 7.34*#		7.73 (2.78 to 12.68)	0.94c
MSTN (ng/mL)	Control	2.43 ± 0.82	2.44 ± 0.53	Group = 0.012 Time = 0.024 Group×time = 0.017	0.01 (−0.33 to 0.35)	0.04	1.02g
	Exercise	2.17 ± 0.32	1.68 ± 0.35**##		−0.49 (−0.76 to −0.22)	1.09c

* P < 0.05, ** P < 0.01, post-intervention, compared with pre-intervention

^#P< 0.05,^##P < 0.01, comparisons between the exercise and control groups at corresponding time points

Cohen’s d of 0.2, 0.5, 0.8 represents small, medium, and large effect size, a: small effect size, b: medium effect size, c: large effect size

Hedges’g of 0.2, 0.5, 0.8 represents small, medium, and large effect size, e: small effect size, f: medium effect size, g: large effect size

Fig. 4.

Changes in irisin (a), IGF-1 (b) and MSTN (c) before and after 36-week intervention. * P < 0.05, ** P < 0.01, post-intervention, compared with pre-intervention; ^# P < 0.05, ^## P < 0.01, comparisons between the exercise and control groups at corresponding time points

Discussion

Our 36-week FITT-VP-based individualized resistance training intervention demonstrated clinically meaningful benefits for women with probable sarcopenic obesity. Age-related skeletal muscle deterioration involves adipose-mediated pathophysiological cascades, where ectopic lipid infiltration disrupts muscular homeostatic architecture and contractile capacity, ultimately elevating fall susceptibility and frailty phenotypes [21]. This metabolic derangement is exacerbated by concurrent reductions in lean mass, physical inactivity, and chronic low-grade inflammation - key drivers of visceral adipogenesis, insulin resistance, and obesogenic progression [22]. Under the premise that dietary patterns and total caloric intake remain unchanged, our 36-week personalized resistance training intervention significantly improved body composition in older women with probable sarcopenic obesity. Both within-group (pre-post) and between-group (vs. control) analyses demonstrated substantial reductions in body fat percentage and increases in ASMI—key determinants of favorable body composition in this population. These findings corroborate evidence that resistance training ≥ 4 weeks effectively mitigates adiposity-related parameters, including total fat mass [23], while extending prior knowledge by highlighting the necessity of optimizing exercise modality, intensity, and volume to maximize muscle hypertrophy [24]. Notably, a dose-response relationship was observed: thrice-weekly high-intensity training elicited superior gains in muscle mass compared to low-frequency protocols in older adults (> 65 years) [25]. However, meta-analytical evidence underscores delayed musculoskeletal adaptations in aging populations, suggesting prolonged interventions are required irrespective of training strategy [26].

This longitudinal investigation implemented a graded resistance training protocol progressing from moderate-low to moderate-high intensity, incorporating repetition variation during thrice-weekly 30-minute sessions to optimize adaptive responses while minimizing accommodation effects [27]. During the initial 7-week neuromuscular adaptation phase, low-moderate intensity training prioritized neural drive enhancement through motor unit activation, synchronization, and firing rate modulation over structural hypertrophy [28]. Type II fiber hypertrophy requires satellite cell activation through high-threshold motor unit recruitment and sustained mTORC1-mediated mechanical tension [29], necessitating extended training from weeks 8–36 versus initial neural adaptation (weeks 1–7). This extended temporal framework enhanced musculoskeletal plasticity via load-specific adaptability, reduced non-response likelihood, and potentiated anabolic signaling through mTORC1 pathway upregulation, collectively optimizing protein synthesis and lean mass accrual in older people [30].

According to the 2019 Asian Sarcopenia Working Group, muscular strength is a better predictor of adverse outcomes than muscle mass. Early detection and remediation of decreased muscular strength can improve older individuals’ ability to walk independently [16]. Given its practicality in community settings, this study uses handgrip strength as an indicator of muscular strength improvements. Based on participants’ specific conditions and workout demands, the resistance training exercises primarily target the upper and lower limbs. This study found that handgrip strength significantly increased in both within-group and between-group comparisons. Although lower limb muscular strength was not quantitatively assessed, observed improvements in functional capacity demonstrated significant associations with strength augmentation. While our intervention did not result in statistically significant increases in whole-body skeletal muscle mass among older women with probable SO, it did lead to notable enhancements in muscular performance. Specifically, these improvements were evidenced by an increase in walking speed, as well as a significant reduction in the time required to complete the five times sit-to-stand test and the TUG test. These experimental findings align with prior research evidence. For instance, Liao et al. [31] found that a 12-week resistance training program with elastic bands for women with SO would significantly improve the women’s grip strength, gait velocity, five times sit-to-stand time, and TUG time. After three months and nine months of follow-up, there was still a significant increase compared to the control group. Additionally, a randomized controlled trial documented the efficacy of 15-week progressive resistance training in improving gait velocity (4-meter walk test) and lower-extremity strength parameters (five-times sit-to-stand test scores) in SO patients aged 60–90 years [32]. Targeted loading of axial and appendicular musculoskeletal systems is posited to be a critical determinant of functional gains. However, geriatric exercise prescription necessitates a balance between progressive overload paradigms and personalized risk mitigation protocols, particularly when addressing coexisting osteopenia, cardiometabolic comorbidities, and diminished physiological reserve. A graduated dose-escalation strategy was implemented to optimize training adaptability while minimizing injury risks in this vulnerable population. In the present study, exercise dosage strictly adhered to the FITT-VP framework (Frequency, Intensity, Time, Type, Volume, and Progression), initiating with low-intensity resistance exercises followed by systematic progression calibrated to participants’ age, medical history, health status, and baseline physical activity levels - all clinically relevant covariates in older women. Training intensity was progressively titrated until achieving target heart rate zones that optimized risk-benefit ratios through physiological adaptation monitoring. It should be noted that during data analysis, we observed a marked reduction in 6-meter walking speed from 1.4 m/s at baseline to 1 m/s post-intervention in one control group participant. Retrospective investigation revealed that the participant experienced significant life stressors during the post-intervention assessment period, which likely compromised test performance. This finding highlights the importance of incorporating psychological status evaluation in future functional assessments of older adults to ensure accurate outcome interpretation.

The factors secreted by contracting skeletal muscle play a pivotal role in regulating muscle development, metabolic homeostasis, and functional integrity. Furthermore, skeletal muscle possesses the capability to modulate blood flow to various tissues and organs, thereby influencing their functional capacity through intricate paracrine and endocrine mechanisms [33]. Research indicates that the browning of white adipose tissue (WAT), a process that enhances fatty acid uptake, oxidation, and metabolism, is significantly reliant on the myokines secreted by skeletal muscle during physical exercise [34]. Irisin and IGF-1, myokines secreted by muscles activated by exercise, are essential in controlling muscle growth, inducing muscle hypertrophy, and promoting the browning of white fat. As a negative regulator of muscle growth, MSTN has the ability to inhibit both muscle growth and repair as well as the browning of white fat [33].

Our 36-week progressive resistance training intervention significantly elevated serum IGF-1 levels while suppressing MSTN in older women with possible SO. These myokine alterations exerted pleiotropic effects, enhancing skeletal muscle metabolism while systemically improving white adipose browning, hepatic lipid regulation, and bone remodeling through optimized energy partitioning [35]. Resistance exercise has been shown to augment protein synthesis through the IGF-1/PI3K/AKT signaling pathway, fostering the production of myogenic IGF-1 and thereby inducing muscle hypertrophy [36]. Additionally, studies reveal that resistance exercise may stimulate WAT browning via the AMPK/PGC-1α/FNDC5 pathway [37]. Notably, resistance exercise exhibits a more pronounced effect than aerobic exercise in downregulating MSTN expression and promoting gastrocnemius muscle hypertrophy. Specifically, resistance exercise promotes the expression and secretion of IGF-1, and inhibits MSTN through the extracellular signal-regulated kinase (ERK)-dependent pathway. When IGF-1 levels are elevated and MSTN levels are suppressed, skeletal muscle growth and fat loss are synergistically enhanced. IGF-1 and MSTN exert considerable influence on the regulation of skeletal muscle growth, muscle fiber type, and fat deposition by promoting the expression and phosphorylation of AKT and S6 kinase [38]. In our study, no significant group × time interaction was observed for irisin changes. This may be attributed to the absence of significant muscle mass alterations following the 36-week intervention—consistent with established evidence that elevated circulating irisin concentrations correlate positively with muscle mass accretion [39]. Furthermore, considerable intra- and inter-individual heterogeneity in FNDC5/irisin responses to diverse exercise modalities suggests that exercise typically does not upregulate skeletal muscle FNDC5 mRNA or circulating irisin in most individuals [40]. Collectively, these findings indicate that the health benefits conferred by exercise may occur independently of irisin-mediated pathways. Myokines play a crucial role in modifying exercise-based interventions for obesity, sarcopenia, and other disorders. However, the impact of exercise on myokines is intricately linked to various factors, including exercise type, intensity, timing of intervention, as well as the age, health status, training experience, and nutritional status of the participants. Current research has yet to fully elucidate these complex relationships. Therefore, future studies should adopt a more tailored approach to investigate the dose-response effects of diverse exercise modalities on the release of myokines.

Limitations

There are some limitations in this study. First, the study’s small sample size may restrict the statistical power and generalizability of the findings. While our study has placed particular emphasis on the selection of older female participants, thus augmenting the research data on possible sarcopenic obesity in women, it is recommended that future research should broaden the sample size and include diverse demographic subgroups (e.g., males, different ethnicities) to continue observing the effects of resistance exercise training. Second, although home-based resistance training enhances accessibility, the absence of direct supervision raises concerns about exercise fidelity (e.g., adherence to prescribed intensity and technique). Self-reported compliance and the lack of real-time monitoring may overestimate intervention adherence and efficacy. Lastly, we conducted surveys and recorded the dietary habits of the participants at both the baseline and post-intervention stages, but we did not conduct periodic follow-ups throughout the 36-week intervention period. Consequently, there may be some influencing factors (such as medication use and hormonal fluctuations) that could have impacted our results, particularly the changes in hematological indicators.

Conclusion

The 36-week personalized resistance training program results in elevated serum IGF-1 concentrations and reduced myostatin levels, alongside independent improvements in muscle strength and muscle function among older women with probable sarcopenic obesity. The tailored residence-based resistance training implemented in this investigation poses negligible risks yet confers substantial benefits to health, thereby constituting a recommended prophylactic strategy for individuals afflicted with sarcopenic obesity and thus requiring secondary prevention measures.

Authors’ contributions

Recruitment, intervention, conceptualization, methodology, software, validation, data curation, formal analysis, X.G., G.C.; intervention, data curation, methodology, Y.T., X.L., Y.Z., M.C., G.C.; writing, review and editing, X.G. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by Fundamental Research Funds for the Central Universities (No. 2024YDYY005) and the Herbalife Winter Sports Development Fund (No. KBL2021006).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.