Effects of ACSM protocol-compliant exercise on sarcopenia: a systematic review and meta-analysis of randomized controlled trials Authors: Chen Wang, Xiao Liu, Yu Liang, Qi Song, Hai Liu, Zhaochuan Xu, Xifang Liu Published: 2026 ABSTRACT AbstractBackgroundSarcopenia is a progressive, age-related skeletal muscle disorder characterized by a gradual loss of muscle mass, strength, and physical function, which is highly prevalent among older adults. This study aimed to investigate the association between compliance with American College of Sports Medicine (ACSM) exercise guidelines and intervention outcomes in patients with sarcopenia.MethodsFour electronic databases were systematically searched from inception to May 2025. Exercise interventions were evaluated for compliance with ACSM recommendations and categorized into high or low ACSM protocol compliance groups. A random-effects model was used to calculate standardized mean differences (SMD) and 95% confidence intervals (CI).ResultsTwenty-nine randomized controlled trials, comprising 1,524 patients with sarcopenia, were included. Exercise with high ACSM protocol compliance significantly improved the appendicular skeletal muscle mass index (SMD = 0.49; 95% CI 0.23, 0.76). Significant positive effects were observed for maximal voluntary contraction and the five-times sit-to-stand test performance; however, these findings remain inconclusive due to high subgroup heterogeneity and limited evidence. Grip strength demonstrated a larger effect size in the high ACSM protocol compliance group (SMD = 0.64; 95% CI 0.34, 0.94), but this was statistically significant only at the 70% threshold (p = 0.03), losing significance at the 75% (p = 0.19) and 80% (p = 0.24) thresholds, indicating substantial instability in the results. Subgroup differences for gait speed (p = 0.15) and the timed up-and-go test (p = 0.08) were not statistically significant. Exercise had no significant effect on body mass index or body fat percentage, and high ACSM protocol compliance was not associated with increased adverse events or dropout rates.ConclusionsExercise interventions with high ACSM protocol compliance significantly improved muscle mass in patients with sarcopenia. Although improvements were observed in grip strength, the five-times sit-to-stand test performance, and maximal voluntary contraction, inconsistent sensitivity analyses and high heterogeneity preclude definitive conclusions. For other outcomes, interventions with high ACSM compliance demonstrated no superior efficacy compared with other exercise protocols.Trial registrationThis systematic review was registered with PROSPERO (CRD420251104971).Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12877-026-07872-2.Keywords: Sarcopenia, Exercise, Muscle mass, Physical function, Meta-analysis Background Sarcopenia is a progressive, age-related skeletal muscle disorder characterized by diminished muscle mass, reduced muscle strength, and/or impaired physical function [1], a concept first introduced by Rosenberg in 1989 [2]. It now represents a significant public health challenge among older adults, with its worldwide prevalence estimated at 10% to 27% in individuals aged 60 and above [3]. The clinical setting exerts a substantial influence on the prevalence rates, with estimates of 9%–11% for community-dwelling older adults, compared to 23%–24% for hospitalized patients and 31%–51% for nursing home residents, respectively [4]. Sarcopenia is associated with multiple adverse clinical events, including falls, fractures [5], impaired mobility, loss of independence [6], and a heightened risk of mortality [7, 8], thereby imposing a considerable medical and economic burden on both individuals and society. Currently, there is no approved pharmacotherapy specifically indicated for the management of sarcopenia [9], underscoring the growing importance of non-pharmacological interventions. Exercise training is considered the most impactful intervention, with well-documented benefits in slowing muscle atrophy, augmenting muscle strength, and improving physical performance [10, 11]. Extensive research consistently demonstrates that exercise training enhances muscle mass, bolsters muscular strength, and improves physical function in patients with sarcopenia [12]. However, findings regarding the most effective intervention protocol remain inconsistent [13, 14]. A consistent finding in recent meta-analyses is that the efficacy of exercise interventions for sarcopenia differs markedly depending on the specific modality employed [15, 16]. Resistance training is widely considered the primary method for augmenting muscle mass and strength. For sarcopenic individuals, high-intensity resistance exercise, often defined as 80% of a one-repetition maximum (1RM), is specifically recommended as the optimal strategy for maximizing muscular strength [17]. Aerobic training primarily improves cardiorespiratory function and endurance, thereby enhancing performance in daily activities, while balance training is crucial for reducing fall risk and improving gait stability. Both resistance training alone and multimodal approaches (combining resistance with aerobic or balance training) have demonstrated effectiveness in enhancing quality of life and functional capacity [18]. A consensus exists that multimodal therapeutic strategies, encompassing physical training and nutritional support, represent the most effective comprehensive strategy for sarcopenia management [19]. Whole-body vibration training is also considered an efficacious prophylactic and therapeutic modality [20]. Although exercise training as a non-pharmacological therapy offers substantial advantages for sarcopenic individuals, the mechanisms by which different training parameters (e.g., intensity, frequency) influence outcomes remain unclear, highlighting the need for robust clinical studies to identify optimal exercise prescriptions. The American College of Sports Medicine (ACSM), a leading global authority in exercise science, publishes evidence-based guidelines integrating physiological, clinical, and public health evidence, which have been adopted as international standards for exercise prescription. The ACSM has established broad physical activity recommendations for healthy adults, detailing specific parameters for cardiorespiratory, resistance, and flexibility training [21]. Sarcopenia involves a significant decline in muscle mass, strength, and physical function, often accompanied by multiple comorbidities and reduced physiological reserves. Therefore, directly applying exercise parameters designed for healthy individuals to this frail population theoretically poses a potential risk of excessive intensity or insufficient tolerance. However, in recent years, the applicability of this evidence-based framework to frail populations has been supported by empirical evidence. A meta-analysis by Pan et al. clearly demonstrated that exercise with high compliance with the ACSM guidelines was significantly superior to those with low compliance in improving frailty phenotypes and grip strength, confirming its efficacy in frail older adults [22]. Given the high overlap between frailty and sarcopenia in terms of pathophysiological mechanisms and clinical manifestations, this review adopts the ACSM framework as an evaluation benchmark. It aims to examine the effects of varying degrees of ACSM protocol compliance on body composition, muscle strength, and physical function in patients with sarcopenia, and to further assess the actual tolerability of the ACSM guidelines in this population. Methods Search strategy To ensure comprehensive identification of eligible studies, four electronic databases—PubMed, Embase, the Cochrane Library, and Web of Science—were systematically searched from inception through May 2025. The search strategy was developed based on the PICOS framework, using controlled vocabulary terms (e.g., “Exercise,” “Resistance Training,” “Plyometric Exercise,” “Traditional Chinese Exercise,” and “Sarcopenia”) combined with relevant free-text keywords. To maximize comprehensiveness, we additionally screened the reference lists of eligible studies and relevant systematic reviews. When required, the original study authors were contacted to obtain supplementary data. Supplementary Table S2 details the full search strategy. Study selection criteria Included studies met the following criteria: (a) published randomized controlled trials (RCTs); (b) individuals clinically diagnosed with sarcopenia; (c) interventions involving any exercise modality, such as resistance, aerobic, or flexibility training; and (d) control groups that received non-exercise comparators or alternative interventions, such as conventional physiotherapy or health education. Exclusion criteria were as follows: (a) non-original research, e.g., conference abstracts, reviews, animal studies, guidelines, or editorials; (b) control groups involving any form of exercise intervention; (c) concurrent use of pharmacological agents during exercise interventions; (d) studies with unavailable or non-extractable data; (e) absence of a control group or use of a non-randomized design; or (f) multiple articles stemming from a single trial. Titles and abstracts underwent independent screening by two authors (CW and XL); full texts were retrieved if a study was deemed potentially eligible by either reviewer. Following the initial screening, the full texts underwent independent eligibility assessment by CW and XL. Discrepancies were resolved through discussion with a third reviewer (QS) until a consensus was reached. No restrictions were imposed on participants’ age, sex, publication year, or language. Data synthesis and analysis Data extraction was independently performed by two authors (CW and YL). A standardized Excel form was used for the systematic collection of various study parameters, including: bibliographic information (title, authors, year), methodological features (design, intervention protocols), subject demographics (age, sex distribution), training variables (frequency, intensity, duration, repetitions, sets), risk of bias assessment, and all relevant outcome data. Additionally, we systematically extracted and summarized adverse events and participant dropout rates reported across all included studies. Following data extraction, we systematically appraised the exercise dosage and protocol compliance within the included studies. Each study’s exercise intervention was independently assessed by two authors (CW and HL) for compliance with the ACSM recommendations for aerobic capacity, muscle performance, and neuromotor function in healthy adults [21], which are considered applicable to patients with sarcopenia (Table 1). This evaluation focused on key elements of exercise dosage: frequency, intensity, duration, repetitions, and sets. Each parameter was rated on a scale of 0–2: 2 (full), 1 (partial/uncertain), and 0 (non-compliance). Any discrepancies were resolved through discussion with a third reviewer until agreement was achieved. Notably, this ACSM protocol compliance score reflects the extent to which the prescribed exercise complied with the guidelines in its design, and does not measure participant attendance or individual adherence to the sessions. Based on the total score, interventions were categorized as having high ACSM protocol compliance (≥ 70%) or low ACSM protocol compliance (< 70%). Table 1The ACSM recommendations for cardiorespiratory fitness, muscular strength, and flexibility in healthy adultsExercise doseCardiorespiratoryexerciseResistance exerciseFlexibility exerciseFrequency3-5 days per week2-3 days per week>2-3 day/week, dailyIntensity/workload40%-59%VO2R or HRR, RPE of 12-13 on a 6-20 scaleStart with 40%-50% 1RM, more capable with 60%-70% 1RMStretch until you feel your muscles being pulled tight or a slight discomfortDurationContinuous or cumulative 30 minStarting with one set of 8-12 repetitions, increase to two sets after about 2 weeks. Perform no more than 8-10 exercises per sessionStatic stretching held for 10-30 s, repeated 2-4 timesVO2R oxygen uptake reserve, HRR heart rate reserve, RPE ratings of perceived exertion, 1RM one-repetition maximum The ACSM recommendations for cardiorespiratory fitness, muscular strength, and flexibility in healthy adults VO2R oxygen uptake reserve, HRR heart rate reserve, RPE ratings of perceived exertion, 1RM one-repetition maximum Statistical analysis The meta-analysis was conducted using Review Manager 5.4 and STATA 17.0. All studies were stratified into two subgroups: “high ACSM protocol compliance” and “low ACSM protocol compliance”. Within-group heterogeneity was quantified using the Higgins I² statistic, classified as follows: low (I²≤ 25%), moderate (25% < I²≤ 50%), high (50% < I²≤ 75%), and very high (I²>75%). Given the expected heterogeneity in intervention frequency, dosage, duration, session length, and exercise type, a random-effects model was pre-specified to account for between-study variation. Effect sizes were reported as standardized mean differences (SMD) with corresponding 95% confidence intervals (95% CI). Forest plots were generated to visualize pooled effect sizes and their distributions. Publication bias was assessed by visually evaluating the symmetry of the funnel plot, supplemented by quantitative validation using Begg’s rank correlation test and Egger’s linear regression test. A two-tailed p < 0.05 served as the predefined criterion for statistical significance. Furthermore, to assess the robustness of our findings, we conducted sensitivity analyses by setting the threshold for defining “high ACSM protocol compliance” at 75% and 80%, respectively. Four authors (CW, XL, HL, and ZCX) independently evaluated the methodological quality of all included RCTs using the Cochrane Risk of Bias tool (RoB 2) [23]. Based on the Cochrane Handbook [24], this tool encompasses seven elements: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other sources of bias. Each domain was assigned a rating of “low risk,” “some concerns,” or “high risk”. The overall risk of bias was judged according to the following criteria: it was classified as “low” if all domains were rated “low risk”; as “some concerns” if any domain indicated “some concerns” without any “high risk” ratings; and as “high” if any domain was deemed “high risk”. Results Literature search A total of 10,807 records were initially identified from four electronic databases (PubMed: n = 3,206; Embase: n = 1,803; Web of Science: n = 2,061; Cochrane Library: n = 3,737), with five further records obtained from other sources. After removing 4,176 duplicates, 6,631 records remained for title and abstract screening. Of these, 194 articles underwent full-text eligibility assessment. Ultimately, 29 studies were included in the review [25–53]. The complete search and selection process is presented in the PRISMA flow diagram (Fig. 1). PRISMA study flow diagram Study characteristics The search for this meta-analysis was updated through May 2025, identifying 29 eligible studies conducted between 2012 and 2024. No new qualifying studies were identified between January and May 2025. These studies involved a total of 1,524 participants (765 in intervention groups and 759 in control groups), with sample sizes ranging from 19 to 169 participants per trial and average ages ranging from 64 to 89 years (overall mean age: 74.2 years). After excluding the study that did not report sex distribution [29], sex was reported for 1,484 participants, of whom 1,023 (69.0%) were female and 461 (31.0%) were male. This sex distribution indicates that the study population was predominantly female, potentially reflecting the higher prevalence of sarcopenia in older women (Table 2). Table 2Basic characteristics of the included studiesAuthor, yearCountrySample size (n)Age (years)Mean (SD)BMI (Kg/m²)Mean (SD)IG (F:M)CG (F:M)IGCGIGCGBagheri et al. (2020) [38]Iran10 (0:10)10 (0:10)63.8 (3.6)65 (3.9)22.8 (1.0)22.9 (1.3)Bellomo et al. (2013) [26]Italy10 (0:10)10 (0:10)70.9 (5.2)28.2 (5.9)Cebrià I Iranzo et al. (2018) [30]Spain11 (9:2)17 (12:5)82.6 (9.1)81.2 (5.4)31.4 (7.5)31.2 (5.9)Chen et al. (2018) [31]China17 (17:0)16 (16:0)66.7 (5.3)68.3 (2.8)NRNRChen et al. (2023) [33]China30 (30:0)30 (30:0)65.7 (2.5)65.2 (2.6)NRNRFlor-Rufino et al. (2023) [47]Spain20 (20:0)18 (18:0)79.9 (7.2)79.6 (7.7)26.2 (4.2)27.7 (2.9)Guo et al. (2024) [53]Japan33 (19:14)30 (14:16)66.9 (4.4)65.4 (4.0)23.2 (2.1)21.9 (2.9)Jung et al. (2019) [35]Korea13 (13:0)13 (13:0)75.0 (3.9)74.9 (5.2)21.8 (1.5)21.9 (1.5)Kim et al. (2012) [25]Japan39 (39:0)39 (39:0)79.0 (2.9)78.7 (2.8)18.9 (2.0)18.8 (1.7)Kim et al. (2013) [27]Japan32 (32:0)32 (32:0)79.6 (4.2)80.2 (5.6)NRNRLi et al. (2021) [41]China62 (48:14)59 (47:12)73.7 (5.7)72.9 (6.3)23.3 (2.8)22.6 (2.9)Li et al. (2022) [43]China35 (25:10)35 (24:11)80.6 (8.9)77.9 (10.4)22.0 (0.5)22.6 (0.6)Liao et al. (2024) [48]China35 (19:16)46 (22:24)72.0 (5.0)73.2 (5.0)21.5 (3.5)21.9 (2.6)Liu et al. (2024) [49]China41 (29:12)45 (27:18)74.2 (4.7)75.6 (6.4)20.4 (2.4)20.7 (1. 9)Mafi et al. (2019) [36]Iran14 (0:14)16 (0:16)69.0 (2.4)68.0 (3.0)23.2 (1.9)23.5 (1.9)Makizako et al. (2020) [40]Japan36 (26:10)36 (25:11)74.1 (6.6)75.8 (7.3)20.9 (2.7)20.6 (2.1)Maruya et al. (2016) [28]Japan34 (19:15)18 (10:8)69.2 (5.6)68.5 (6.2)20.3 (1.7)19.7 (2.0)Mori and Tokuda (2022) [45]Japan23 (19:4)24 (20:4)77.6 (5.2)77.8 (4.5)20.3 (2.9)20.1 (3.1)Rezaei et al. (2024) [50]Iran10 (0:10)9 (0:9)72.5(4.2)76.5 (3.5)23.5 (0.9)23.2 (2.8)Sen et al. (2021) [42]Turkey46 (40:6)44 (36:8)73.0 (4.8)72.7 (5.0)28.7 (4.8)28.7 (4.6)Seo et al. (2021) [44]Korea12 (12:0)10 (10:0)70.3 (5.4)72.9 (4.8)22.9 (2.0)22.4 (1.5)Tsekoura et al. (2018) [34]Greece18 (16:2)18 (16:2)74.6 (6.0)72.9 (8.3)22.0 (2.2)23.0 (2.3)Wei et al. (2016) [29]China20 (NR)20 (NR)75.0 (6.0)76.0 (6.0)23.4 (2.5)23.8 (2.7)Wei et al. (2017) [32]China20 (13:7)20 (15:5)76.0 (6.0)75.0 (6.0)23.8 (2.7)23.4 (2.5)Wei et al. (2022) [52]China30 (16:14)30 (14:16)66.7 (4.1)65.4 (4.0)22.8 (2.2)21.9 (2.9)Yuenyongchaiwat and Akekawatchai (2022) [46]Thailand30 (20:10)30 (24:6)69.2 (6.7)71.9 (5.2)NRNRZhang et al. (2024) [51]China20 (10:10)20 (10:10)84.5 (4.0)85.5 (4.1)20.7 (2.5)20.3 (2.3)Zhu et al. (2019) [37]China40 (29:11)37 (29:8)74.5 (7.1)72.2 (6.6)18.8 (1.9)18.9 (2.2)Zhu et al. (2019) [39]China24 (0:24)27 (0:27)88.8 (3.7)87.5 (3.0)22.4 (3.1)23.5 (3.9)SD standard deviation, NR Not Reported, IG Intervention group, CG Control group, F:M Female: Male Basic characteristics of the included studies SD standard deviation, NR Not Reported, IG Intervention group, CG Control group, F:M Female: Male Intervention durations across the included studies ranged from 8 weeks to 6 months, with 2 to 5 exercise sessions weekly. Interventions encompassed resistance training, kettlebell exercises, suspension training, Ditang Quan, vibration training, circuit training, Tai Chi, Yi Jin Jing, and multicomponent exercise (Table 3). When categorized by type, 18 studies included aerobic training, 27 resistance training, and 4 flexibility training. Table 3Characteristics of the study intervention and outcomeAuthor, yearInterventionsLength ofinterventionControlOutcomeBagheri et al. (2020) [38]Resistance and Endurance training8 weeksCONBMI/BFPBellomo et al. (2013) [26]Resistance training12 weeksCONMVCCebrià I Iranzo et al. (2018) [30]Peripheral muscles training12 weeksCONMVC/Grip strength/Gait speedChen et al. (2018) [31]Kettlebell training8 weeksCONGrip strength/ASMIChen et al. (2023) [33]Tai Chi and Resistance training8 weeksCONGrip strength/ASMI/ FTSST/Gait speedFlor-Rufino et al. (2023) [47]High-intensity resistance training6 monthsCONMVC/Grip strength/ASMI/BMI/Gait speedGuo et al. (2024) [53]Tai chi and Strength training24 weeksCONGrip strengthJung et al. (2019) [35]Circuit training12 weeksCONASMI/BMI/BFPKim et al. (2012) [25]Strengthening exercise, balance and gait training3 monthsCONKES/BMI/Gait speedKim et al. (2013) [27]Strengthening, balance and gait training3 monthsCONKES/Grip strength/ Gait speed/TUGLi et al. (2021) [41]Resistance training and Outdoor activity12 weeksCONGrip strength/ASMI/BFPLi et al. (2022) [43]Ditangquan exercise24 weeksconventionalexercisesTUGLiao et al. (2024) [48]Resistance training16 weeksCONGrip strength/ASMI/BMI/Gait speedLiu et al. (2024) [49]Progressive resistance training and Aerobic walking exercise12 weeksCONKES/ASMI/BFP/Gait speed/TUGMafi et al. (2019) [36]Resistance training8 weeksCONASMI/BMI/TUGMakizako et al. (2020) [40]Multicomponent exercise12 weeksCONGrip strength/ FTSST/Gait speed/TUGMaruya et al. (2016) [28]Walking and lower limb resistance training6 monthsCONKES/Grip strength/ASMI/BMI/BFP/Gait speedMori and Tokuda (2022) [45]Resistance training24 weeksCONKES/Grip strength/ASMI/Gait speedRezaei et al. (2024) [50]TRX suspension training8 weeksCONGrip strength/ASMI/BMI/BFP/ FTSST/TUGSen et al. (2021) [42]Multicomponent exercise3 monthsCONTUGSeo et al. (2021) [44]Resistance training16 weeksCONMVC/Grip strength/BMI/BFP/Gait speedTsekoura et al. (2018) [34]Comprehensive progressive group exercise and walking3 monthsCONGrip strength/ASMI/BMI/FTSST/Gaitspeed/TUGWei et al. (2016) [29]Whole-body vibration12 weeksCONFTSST/Gait speed/TUGWei et al. (2017) [32]Whole-body vibration training12 weeksCONMVCWei et al. (2022) [52]Yi Jin Jing and Resistance training24 weeksCONGrip strengthYuenyongchaiwat and Akekawatchai (2022) [46]Walking and TheraBand elastic resistance exercise12 weeksCONGrip strength/ASMI/Gait speedZhang et al. (2024) [51]Conventional training and Whole-body vibration8 weeksConventional trainingMVC/FTSST/Gait speed/TUGZhu et al. (2019) [37]Resistance exercises and Aerobic exercises12 weeksCONKES/Grip strength/ASMI/ FTSST/Gait speedZhu et al. (2019) [39]Tai Chi8 weeksCONMVC/Grip strengthCON Control group with routine care, MVC Maximal voluntary contraction, BMI Body mass index, BFP Body fat percentage, ASMI Appendicular skeletal muscle mass index, FTSST Five-times sit-to-stand Test, TUG Timed-up-and-go test, KES Knee extension strength Characteristics of the study intervention and outcome CON Control group with routine care, MVC Maximal voluntary contraction, BMI Body mass index, BFP Body fat percentage, ASMI Appendicular skeletal muscle mass index, FTSST Five-times sit-to-stand Test, TUG Timed-up-and-go test, KES Knee extension strength Risk of bias Across all studies, a low risk of bias was consistently found for random sequence generation and selective reporting. For allocation concealment, 11 studies demonstrated low risk due to adequate methods, in contrast to 18 studies classified as having unclear risk owing to incomplete reporting. Performance bias was predominantly high (28 studies at high risk, 1 at low risk), a direct consequence of the challenges inherent in blinding participants and personnel within exercise interventions. For blinding of outcome assessment, 20 studies were classified as low risk, whereas 9 were categorized as unclear risk due to insufficient reporting. Regarding incomplete outcome data, most studies (n = 23) were classified as low risk, whereas 3 studies with < 13 dropouts were judged as having unclear risk, and another 3 studies with ≥ 36 dropouts were rated as high risk. Finally, other bias sources included 17 low-risk, 7 unclear-risk, and 3 high-risk studies, attributable to small sample sizes and inadequate control of nutrition (Figs. 2 and 3). Bias risk summary Risk of bias graph To determine whether high ACSM protocol compliance was associated with more rigorous study methodology, we analyzed the relationship between the risk of bias and ACSM protocol compliance scores. The risk of bias was quantified by summing the scores for each domain (0 = low, 1 = uncertain, 2 = high). A Pearson correlation analysis showed no significant correlation between the total risk of bias score and the compliance score (r = − 0.1137, p = 0.557). This indicates that the level of ACSM protocol compliance was independent of the methodological quality of the included studies. ACSM protocol compliance The compliance of the prescribed exercise protocols with ACSM recommendations was systematically evaluated across the 29 included studies. Fourteen studies demonstrated high ACSM protocol compliance (rates ≥ 70%), while fifteen studies demonstrated low ACSM protocol compliance (rates < 70%). The main reasons for insufficient compliance were deviations in dosage parameters (e.g., intensity, frequency, and duration) from ACSM recommendations and incomplete reporting of exercise prescription details in some studies, which limited accurate evaluation (Table 4). Table 4Exercise interventions evaluated according to the ACSM recommendationsAuthor, YearCardiorespiratory exerciseResistance exerciseFlexibility exerciseACSM Protocol ConsistencyFreqIntensity/workloadDurationFreqIntensity/work loadRepetitionsSetsFreqIntensity/workloadDuration3-5 days per week40–59% VO2R/HRR, 64%-76%HRmax; RPE of 12-13 on a 6-20 scaleContinuous or cumulative 30 min2-3 days per weekStart with 40%-50% 1RM, more capable with 60%-70% 1RM8-12≥1 group≥2-3 day/week, dailyStretch until you feel your muscles being pulled tight or a slight discomfortStatic stretching held for 10-30 s, repeated 2-4 timesPoints (Percent)Bagheri et al. (2020) [38]3++55%-70% HRmax+15-30min+3++40%-75%1RM+8-16+2-3++10/14(71%)Bellomo et al. (2013) [26]2++60%-80% 1RM+10-12++3++7/8(88%)Cebrià I Iranzo et al. (2018) [30]3++40%-60 %1RM++12++NR+7/8(88%)Chen et al. (2018)[31]2++40%-70 %1RM++8-12++3++8/8(100%)Chen et al. (2023) [33]3++Tai Chi 2 groups++NR+3++40%-80%1RM+8-12++2-3++12/14(86%)Flor-Rufino et al. (2023) [47]2++≥70 % 1RM-10-15+3++5/8(63%)Guo et al. (2024) [53]3++Tai Chi++30min++3++40%-85%1RM+5-20+2-4++12/14(86%)Jung et al. (2019) [35]3++60%-80%HRR-25-75min+3/6(50%)Kim et al. (2012) [25]2-Balance and gait training-20min-2++Borg RPE: 12-14++8++NR+7/14(50%)Kim et al. (2013) [27]2-Gait and balance training-20min-2++Borg RPE: 12-14++8-10++1-2++8/14(57%)Li et al. (2021) [41]3++Outdoor walk+60min++3++NR+8++NR+11/14(79%)Li et al. (2022) [43]3++Di Tang Quan++60min++6/6(100%)Liao et al. (2024) [48]5+NR+NR+NR+5/8(63%)Liu et al. (2024) [49]NR+continuous walking-10min-NR+RPE 12-14++NR+NR+6/14(43%)Mafi et al. (2019) [36]3++60%-80%1RM+8-12++3++7/8(88%)Makizako et al. (2020) [40]NR+anterior-posterior or lateral stepping-6min-NR+NR+10++NR+6/14(43%)Maruya et al. (2016) [28]NR+walking-NR+NR+NR+6-20+3++7/14(50%)Mori and Tokuda (2022) [45]2++50%-70%1RM++12-20+2-3++7/8(88%)Rezaei et al. (2024) [50]3++60-65% 1RM++8-15+3++7/8(88%)Sen et al. (2021) [42]3++Balance exercises-10min-3++Borg RPE:10-12-3-10+1-2++3++NR+5min++12/20(60%)Seo et al. (2021) [44]3++NR+6-15+3-5++3++NR+5min++11/14 (79%)Tsekoura et al. (2018) [34]3++Borg RPE: 10–12+15-20min-3++Borg RPE: 10–12++8-12++1-2++3++NR+5-10min++16/20(80%)Wei et al. (2016) [29]3++40Hz-4-NR+3/8(38%)Wei et al. (2017) [32]3++40 Hz-NR+NR+4/8(50%)Wei et al. (2022) [52]NR+Yi Jin Jing++30min++NR+40%-85%1RM+5-20+2-4++10/14(71%)Yuenyongchaiwat and Akekawatchai. (2022) [46]5++walking ≥7500 steps-NR+2++NR+NR+NR+8/14(57%)Zhang et al. (2024) [51]5++9-14 Hz-NR+5++5/8(63%)Zhu et al. (2019) [37]2-NR+20min-2++40%1RM++6-8+6++8/14(57%)Zhu et al. (2019) [39]5++Tai Chi 2 groups++20min- i4/6(67%)ACSM American College of Sports Medicine, VO2R oxygen uptake reserve, HRR heart rate reserve, HRmax maximum heart rate, RPE ratings of perceived exertion, 1RM one-repetition maximum, NR not reported. 2 points for meeting recommendations (++), 1 point for uncertain compliance (+), and 0 points for not meeting recommendations (-) Exercise interventions evaluated according to the ACSM recommendations ACSM American College of Sports Medicine, VO2R oxygen uptake reserve, HRR heart rate reserve, HRmax maximum heart rate, RPE ratings of perceived exertion, 1RM one-repetition maximum, NR not reported. 2 points for meeting recommendations (++), 1 point for uncertain compliance (+), and 0 points for not meeting recommendations (-) Effects of ACSM protocol compliance on body composition For appendicular skeletal muscle mass index (ASMI), a meta-analysis of 14 studies (n = 689) revealed that exercise interventions significantly improved outcomes, with a pooled SMD of 0.27 (95% CI: 0.10, 0.45; p = 0.003) and low heterogeneity (I² = 24%). In subgroup analyses, the high ACSM protocol compliance subgroup (7 studies) showed a significant improvement (SMD = 0.49, 95% CI: 0.23, 0.76; p < 0.001), whereas the low ACSM protocol compliance subgroup (7 studies) yielded non-significant results (SMD = 0.10, 95% CI: − 0.10, 0.30; p = 0.32). Subgroup analyses (p = 0.02) further confirmed that exercise programs with high ACSM protocol compliance are essential for superior ASMI enhancement. For body mass index (BMI), a meta-analysis of 10 studies (n = 385) demonstrated no significant effect of exercise interventions, with a pooled SMD of − 0.06 (95% CI: − 0.26, 0.15; p = 0.59) and no heterogeneity (I² = 0%). Subgroup analyses confirmed that neither the high ACSM protocol compliance subgroup (5 studies; SMD = 0.08, 95% CI: − 0.27, 0.43; p = 0.64) nor the low ACSM protocol compliance subgroup (5 studies; SMD = − 0.12, 95% CI: − 0.37, 0.12; p = 0.33) demonstrated significant effects. The difference between subgroups was not statistically significant (p = 0.35). For body fat percentage (BFP), a meta-analysis of 7 studies (n = 283) demonstrated no significant effect of exercise interventions, with a pooled SMD of − 0.18 (95% CI: − 0.54, 0.19; p = 0.34) and high heterogeneity (I² = 51%). Subgroup analyses confirmed similar findings, with neither the high ACSM protocol compliance subgroup (4 studies; SMD = − 0.15, 95% CI: − 0.64, 0.34; p = 0.54) nor the low ACSM protocol compliance subgroup (3 studies; SMD = − 0.23, 95% CI: − 0.91, 0.45; p = 0.51) demonstrating significant effects. The difference between subgroups was not statistically significant (p = 0.86) (Fig. 4). Forest plot of the effects of exercise on body composition in patients with sarcopenia. A Appendicular skeletal muscle mass index; B Body mass index; C body fat percentage Effects of ACSM protocol compliance on muscle strength For grip strength, a meta-analysis of 16 studies (n = 855) revealed that exercise interventions significantly improved grip strength, with a pooled SMD of 0.42 (95% CI: 0.21, 0.62; p < 0.001) and high heterogeneity (I² = 53%). In subgroup analyses, the high ACSM protocol compliance subgroup (8 studies) demonstrated significant improvements with an SMD of 0.64 (95% CI: 0.34, 0.94; p < 0.001), whereas the low ACSM protocol compliance subgroup (8 studies) demonstrated smaller but significant improvements (SMD = 0.23, 95% CI: 0.02, 0.43; p = 0.03). Subgroup analyses (p = 0.03) further confirmed that exercise programs with high ACSM protocol compliance resulted in greater grip strength improvements. For maximal voluntary contraction (MVC), a meta-analysis of 7 studies (n = 234) showed a significant improvement, with a pooled SMD of 1.19 (95% CI: 0.58, 1.81; p < 0.001) and very high heterogeneity (I² = 77%). Given the small number of included studies and high heterogeneity, these results should be interpreted with caution. In subgroup analyses, the high ACSM protocol compliance subgroup (3 studies) yielded an SMD of 2.24 (95% CI: 1.17, 3.30; p < 0.001), whereas the low ACSM protocol compliance subgroup (4 studies) yielded an SMD of 0.60 (95% CI: 0.28, 0.91; p < 0.001). Subgroup difference (p = 0.004) suggests that exercise programs with high compliance with the ACSM protocol may markedly improve MVC; however, because the high ACSM protocol compliance subgroup included only 3 studies, the uncertainty surrounding this finding remains high (Fig. 5). Forest plot of the effects of exercise on muscle strength in patients with sarcopenia. A Grip strength; B Maximal voluntary contraction Effects of ACSM protocol compliance on physical function For gait speed, a meta-analysis of 15 studies (n = 811) demonstrated that exercise interventions improved performance, with a pooled SMD of 0.55 (95% CI: 0.29, 0.82; p < 0.001) and high heterogeneity (I² = 70%). Subgroup analyses revealed a larger effect size for the high ACSM protocol compliance group (4 studies; SMD = 1.05, 95% CI: 0.24, 1.86, p = 0.01) compared to the low ACSM protocol compliance group (11 studies; SMD = 0.43, 95% CI: 0.17, 0.69, p = 0.001). However, the difference between subgroups was not statistically significant (p = 0.15), and since the high ACSM protocol compliance subgroup included only 4 studies, the stability of the result is limited and should be interpreted with caution. For the Timed Up and Go test (TUG), a meta-analysis of 10 studies (n = 536) demonstrated that exercise interventions improved performance, with a pooled SMD of − 0.67 (95% CI: − 0.92, − 0.42; p < 0.001) and moderate heterogeneity (I² = 48%). In subgroup analyses, the high ACSM protocol compliance subgroup (4 studies) demonstrated improvement (SMD = − 0.97, 95% CI: − 1.41, − 0.54; p < 0.001), whereas the low ACSM protocol compliance subgroup (6 studies) demonstrated a moderate effect (SMD = − 0.52, 95% CI: − 0.78, − 0.26; p < 0.001). Although the effect size was larger in the high ACSM protocol compliance group, the difference between subgroups was not statistically significant (p = 0.08), and given the small number of studies in the high ACSM protocol compliance subgroup, it is not yet possible to determine whether there is a clear dose-response relationship between the level of compliance with the ACSM protocol and the degree of improvement in the TUG. For the five-times sit-to-stand test (FTSST), a meta-analysis of 7 studies (n = 328) showed that exercise interventions improved performance, with a pooled SMD of − 0.52 (95% CI: − 0.90, − 0.15; p = 0.006) and high heterogeneity (I² = 62%). Subgroup analyses revealed that the high ACSM protocol compliance subgroup (3 studies) exhibited an SMD of − 1.03 (95% CI: − 1.44, − 0.62; p < 0.001), whereas the low ACSM protocol compliance subgroup (4 studies) showed an SMD of − 0.21 (95% CI: − 0.52, 0.11; p = 0.21), which was not statistically significant. Subgroup analyses (p = 0.002) suggest that exercise with high ACSM protocol compliance may improve FTSST performance. However, given that the high ACSM protocol compliance subgroup included only 3 studies, the stability of this subgroup finding is limited (Fig. 6). Forest plot of the effects of exercise on physical function in patients with sarcopenia. A Gait speed; B Timed up-and-go test; C Five-times sit-to-stand test Adverse events and dropout rates Regarding the safety and adherence of exercise interventions, the 29 included studies demonstrated overall good tolerability (Table 5). In terms of safety, the vast majority of studies (22/29) did not report any adverse events. Among the studies that explicitly reported adverse events, most (6 studies) reported none; only one study [47] reported a minor adverse event, specifically “muscle discomfort.” This indicates that exercise interventions designed with high ACSM protocol compliance did not increase the risk of injury. Table 5Adverse events and dropout ratesAuthor, yearACSM Protocol ConsistencyAdverse eventsDropout rates70%75%80%Bagheri et al. (2020) [38]HighLowLowNR0Bellomo et al. (2013) [26]HighHighHighNR9%Cebrià I Iranzo et al. (2018) [30]HighHighHighNone26%Chen et al. (2018) [31]HighHighHighNR0Chen et al. (2023) [33]HighHighHighNone26%Flor-Rufino et al. (2023) [47]LowLowLowMuscle discomfort13%Guo et al. (2024) [53]HighHighHighNR12%Jung et al. (2019) [35]LowLowLowNR0Kim et al. (2012) [25]LowLowLowNR8%Kim et al. (2013) [27]LowLowLowNR6%Li et al. (2021) [41]HighHighLowNR40%Li et al. (2022) [43]HighHighHighNR9%Liao et al. (2024) [48]LowLowLowNR30%Liu et al. (2024) [49]LowLowLowNR25%Mafi et al. (2019) [36]HighHighHighNR18%Makizako et al. (2020) [40]LowLowLowNR8%Maruya et al. (2016) [28]LowLowLowNR24%Mori and Tokuda (2022) [45]HighHighHighNR15%Rezaei et al. (2024) [50]HighHighHighNR17%Sen et al. (2021) [42]LowLowLowNR8%Seo et al. (2021) [44]HighHighLowNR14%Tsekoura et al. (2018) [34]HighHighHighNone0Wei et al. (2016) [29]LowLowLowNone15%Wei et al. (2017) [32]LowLowLowNR15%Wei et al. (2022) [52]HighLowLowNR9%Yuenyongchaiwat and Akekawatchai (2022) [46]LowLowLowNR7%Zhang et al. (2024) [51]LowLowLowNR17%Zhu et al. (2019) [37]LowLowLowNone28%Zhu et al. (2019) [39]LowLowLowNone20%High ACSM protocol compliance was considered as a compliance rate of ≥ 70%, ≥ 75%, or ≥ 80%. NR not reported Adverse events and dropout rates High ACSM protocol compliance was considered as a compliance rate of ≥ 70%, ≥ 75%, or ≥ 80%. NR not reported With respect to participant adherence, dropout rates varied considerably across studies, ranging from 0% to 40%, but no trend was observed suggesting that higher ACSM protocol compliance led to higher dropout rates. The high ACSM protocol compliance group included the study with the highest dropout rates [41] but also three studies with 0% dropout [31, 34, 38], with most studies in this group maintaining dropout rates between 9% and 26%. The low ACSM protocol compliance group had only one study with 0% dropout [35], while multiple studies showed dropout rates concentrated between 15% and 30%. Overall, exercise interventions with high ACSM protocol compliance did not result in higher rates of adverse events or dropouts. Subgroup analyses To explore potential sources of heterogeneity, we conducted a meta-regression analysis examining intervention duration, participant age, and sarcopenia diagnostic criteria. The results indicated that participant age was the primary source of grip strength heterogeneity, showing a significant negative correlation (coefficient = − 0.046, p = 0.009). This suggests that as participants age, the effectiveness of interventions in improving grip strength diminishes. Additionally, sarcopenia diagnostic criteria significantly influenced the outcomes for MVC (p = 0.020), walking speed (p = 0.045), and TUG (p = 0.029). Specifically, sarcopenia diagnostic criteria were significantly and negatively associated with the TUG test effect size (coefficient = − 0.119), suggesting systematic differences in effect sizes across studies using different diagnostic standards. However, none of the aforementioned covariates reached statistical significance (p > 0.05) for ASMI, BMI, BFP, or FTSST. This suggests that, beyond the analyzed variables, the high heterogeneity observed in BFP and FTSST may stem from other clinical or methodological differences not incorporated into the analysis. For instance, variations in specific exercise intensities, participants’ nutritional status, or intervention adherence across studies could all potentially contribute to heterogeneity, yet these factors are difficult to uniformly quantify based on the available data reports (Table S3). 3D Visualization To further elucidate the interplay between exercise parameters and clinical outcomes, we constructed a three-dimensional scatter plot visualizing the distribution of SMDs across varying intervention frequencies and durations (Note: this visualization does not include exercise intensity, as intensity parameters were reported inconsistently in the original studies and are difficult to standardize and quantify). The X-axis represents the SMD, the Y-axis denotes exercise frequency (sessions per week), and the Z-axis indicates the intervention duration (weeks). Color-coded clusters in the figure represent multiple outcome measures (Figure S1). The 3D visualization reveals a nonlinear relationship between exercise dosage parameters and the SMD. Analysis indicates that the vast majority of studies focused on a frequency range of 2–3 times per week and a duration of 8–12 weeks. This “moderate-dose” intervention model significantly improved both physiological indicators (grip strength, ASMI) and functional indicators (gait speed, TUG) of sarcopenia, with an SMD > 0.5, indicating moderate efficacy. However, it should be noted that this analysis did not include intensity data; “moderate dose” refers only to the combination of frequency and duration, not the intensity of the physiological load. Furthermore, while increasing frequency (> 3 times/week) or duration (> 24 weeks) maintained the effects, the SMD did not increase proportionally. This implies that exercise interventions with moderate frequency and duration, which often align with ACSM protocol recommendations, are sufficient to induce significant physiological adaptation and functional improvement in patients with sarcopenia. However, since intensity is a core component of the ACSM guidelines and is not reflected in this visualization, the above conclusions should be interpreted with caution in light of the intensity parameters reported in the specific studies. Sensitivity analysis and publication bias assessment To validate the robustness of the meta-analysis results, sensitivity analysis was performed by varying the classification thresholds for ACSM protocol compliance. We raised the threshold for defining “high ACSM protocol compliance” from the original 70% to 75% and 80%, respectively, and reassessed the impact of different compliance levels on each outcome measure (Figure S2, Figure S3). The results indicated that among all measures, only grip strength exhibited sensitivity to threshold changes. The p-value for between-group differences shifted from 0.03 (70% threshold) to 0.19 (75% threshold) and 0.24 (80% threshold), transitioning from statistically significant to non-significant. For the remaining measures—including ASMI, BMI, BFP, MVC, gait speed, TUG, and FTSST—between-group comparisons showed no significant changes. In summary, except for grip strength, the findings for all outcome measures were consistent across different thresholds for ACSM protocol compliance, indicating robust stability and enhancing the reliability of the conclusions. To assess potential publication bias, we evaluated outcomes reported in more than ten studies, namely ASMI (Fig. 7A), BMI (Fig. 7B), grip strength (Fig. 7C), gait speed (Fig. 7D), and TUG (Fig. 7E). The funnel plots for these results were broadly symmetrical on both sides. Begg’s test and Egger’s test yielded p-values greater than 0.05, indicating no significant publication bias for these outcomes (Table 6). Funnel plot of the influence of exercise in patients with sarcopenia. A Appendicular skeletal muscle mass index; B Body mass index; C Grip strength; D Gait speed; E Timed up-and-go test Table 6Publication bias for the studies included in the meta-analysisOutcomeEgger’s testBegg’s testt p z p ASMI2.100.0581.700.090BMI1.590.1501.880.060grip strength-0.290.7770.090.928gait speed2.090.0560.940.347TUG-2.020.077-1.700.089ASMI Appendicular skeletal muscle mass index, BMI Body mass index, TUG Timed-up-and-go test Publication bias for the studies included in the meta-analysis ASMI Appendicular skeletal muscle mass index, BMI Body mass index, TUG Timed-up-and-go test Discussion Based on a comprehensive literature search and screening, this systematic review and meta-analysis included 29 RCTs involving 1,524 patients with sarcopenia. This study is the first to quantitatively evaluate the compliance of exercise program designs with the ACSM recommendations, stratifying interventions into high (14 studies) and low (15 studies) ACSM protocol compliance groups to systematically compare the effects of different compliance levels. Analysis revealed that exercise interventions with high ACSM protocol compliance demonstrated significant advantages in improving muscle mass (ASMI, SMD = 0.49 vs. 0.10). However, the findings exhibit considerable inconsistency regarding muscle strength and physical function. First, although grip strength exhibited subgroup differences in the initial analysis, sensitivity analyses indicated that this finding lacked robustness. When the compliance threshold was adjusted to 75% (p = 0.19) or 80% (p = 0.24), the differences between the high and low ACSM protocol compliance groups lost statistical significance. Second, for MVC and FTSST, although effect sizes appeared larger in the high ACSM protocol compliance group, high subgroup heterogeneity and the limited number of included studies precluded definitive conclusions. Additionally, no significant differences were observed between subgroups for gait speed (p = 0.15) and TUG (p = 0.08), indicating that the high ACSM protocol compliance did not demonstrate superiority in enhancing gait speed or dynamic balance capacity. Research indicates that exercise significantly enhances muscle mass. Resistance training (RT) is considered the primary strategy for muscle hypertrophy [54, 55], exerting its effects through the activation of the mammalian target of rapamycin (mTOR) pathway in muscle fibers (particularly type II fast-twitch fibers) via mechanical loading. This process promotes muscle protein synthesis, inhibits degradation, and ultimately induces muscle fiber hypertrophy [56]. Multimodal training likewise exerts beneficial effects on muscle mass [57]. However, when the exercise dose falls below the ACSM-recommended threshold, it may be insufficient to adequately activate the molecular pathways for muscle growth, thereby limiting protein synthesis and weakening the effect against muscle loss. It should be noted that while this study observed improvements in muscle mass, neither BMI nor BFP showed significant changes. This seemingly contradictory result reflects the complexity of body composition changes and the limitations of assessment metrics. First, BMI, as a ratio of weight to height, cannot distinguish between changes in muscle and fat. Resistance training often involves muscle gain and fat loss, leading to body composition remodeling while weight remains relatively stable, making BMI less sensitive to internal changes. Second, a significant reduction in BFP typically requires sustained energy deficit. The studies included in this meta-analysis primarily focused on muscle strength and physical function. Without strict dietary control, exercise intervention alone may be insufficient to induce a significant decrease in BFP within the intervention period. Exercise interventions exert significant benefits on muscle strength in patients with sarcopenia, particularly in key indicators such as grip strength and MVC. Existing evidence consistently indicates that RT and multimodal training are the most effective approaches for improving grip strength [15, 18]. Furthermore, low-intensity blood flow restriction training exhibits muscle strengthening effects equivalent to those of traditional high-intensity RT [58]. Exercise with high ACSM protocol compliance may facilitate enhanced neural drive and motor unit synchronization, optimize fast-twitch fiber recruitment, induce neural adaptations, and provide the mechanical tension required for subsequent hypertrophy [59]. Conversely, exercise performed below the recommended dosage fails to provide sufficient overload stimulation, thereby limiting neural adaptations and preventing adequate activation and remodeling of fast-twitch fibers, which markedly attenuates improvements in muscle strength. Exercise interventions significantly enhance physical function in patients with sarcopenia, as indicated by increased gait speed and reduced completion times in the TUG and FTSST. Existing evidence indicates that high-velocity resistance training excels in enhancing gait speed, TUG, and FTSST, whereas traditional resistance training yields distinct advantages in the 30-second chair stand test, though the overall effects of these two modalities did not differ statistically [60]. Proprioceptive training has also been shown to reduce TUG time and confer notable benefits for balance, gait stability, and fall risk reduction [61]. Consistent with these observations, our results indicate that high ACSM protocol compliance was correlated with superior enhancements in physical function. Compared with existing meta-analyses, this review exhibits significant differences in the classification of intervention characteristics and outcome measures. Lu et al. [15] compared the effects of three exercise modalities on knee extensor strength and physical function in patients with sarcopenia, finding that resistance training and combined training improved knee extensor strength, gait speed, and TUG, while whole-body vibration training only improved TUG. Shen et al. [18] conducted a network meta-analysis comparing seven exercise types, concluding that combined resistance and balance training was most effective for improving gait speed and the TUG test, while combined nutritional supplementation enhanced grip strength. This review did not solely categorize interventions by exercise modality but instead classified them into high ACSM protocol compliance and low ACSM protocol compliance groups based on adherence to ACSM recommendations (frequency, intensity, and intervention duration), directly assessing the impact of protocol compliance on efficacy. Regarding outcome measures, Lu et al. primarily focused on knee extensor strength, gait speed, TUG, and FTSST test, while Shen et al. emphasized grip strength, gait speed, TUG, FTSST test, and quality of life. Our study not only covers these functional measures but also incorporates MVC, body composition indicators and safety outcomes (adverse events and dropout rates), providing more comprehensive evidence on efficacy and safety. While this review provides valuable insights, certain limitations warrant consideration. Firstly, twenty-eight of the 29 included studies were rated as having a high risk of performance bias. This was primarily due to the inherent difficulty of blinding participants and personnel in exercise interventions, rather than deficiencies in other methodological aspects such as randomization or outcome measurement. Such bias may potentially overestimate the overall effectiveness of exercise interventions. Although the included studies primarily relied on objective measures, which tends to mitigate measurement bias, the widespread lack of blinding may still compromise the validity and reliability of effect estimates. Secondly, MVC, FTSST, gait speed, and TUG exhibit significant heterogeneity and are supported by limited evidence, resulting in inconclusive findings. Finally, exercise intensity was not analyzed as an independent variable. Although 3D visualization presented the distribution characteristics of exercise frequency and duration, the intensity dimension was not incorporated into a stratified analysis. This limitation prevented us from fully quantifying and comparing the independent contributions of different exercise intensities to sarcopenia outcomes. The current scoring system assigns equal weight to frequency, intensity, and duration, potentially leading to higher composite scores for high-frequency but low-intensity interventions. Future studies should specifically examine the dose-response relationship between exercise intensity and efficacy through stratified analyses, regression models, or weighted scoring methods to more precisely guide clinical exercise prescription development. Conclusion This meta-analysis indicates that exercise interventions with high ACSM protocol compliance significantly improve muscle mass in patients with sarcopenia. Although improvements were observed in grip strength, FTSST performance, and MVC, inconsistent sensitivity analyses and high heterogeneity preclude definitive conclusions regarding these findings. For other outcomes, interventions with high ACSM protocol compliance demonstrated no superior efficacy compared with other exercise protocols. Supplementary Information Supplementary Material 1. Acknowledgements Not applicable. Abbreviations American College of Sports Medicine Standardized mean differences One-repetition maximum Randomized controlled trials Confidence intervals Appendicular skeletal muscle mass index Maximal voluntary contraction Timed Up and Go test Five-times sit-to-stand test Body mass index Body fat percentage Resistance training Authors’ contributions CW conceptualized and designed the study. Initial screening of titles and abstracts was independently performed by CW and XL, with QS resolving any ensuing discrepancies. Data extraction was independently conducted by CW and YL. CW and HL independently evaluated intervention compliance and applied ACSM-recommended dosages. The methodological quality of all included studies was independently assessed by CW, XL, HL, and ZCX. XL (corresponding author) provided overall supervision, coordinated the writing process, and performed critical revisions of the manuscript. All authors collaborated on the manuscript’s writing and granted final authorization. Funding This study is primarily supported by the Xi’an Municipal Bureau of Science and Technology Medical Research Project (No. 24YXYJ0049), with additional support from the Xi’an Municipal Health Commission project (No. SCL202403). Data availability All data generated or analyzed during this study are included in this article. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement All data generated or analyzed during this study are included in this article.