Systematic review and meta-analysis of the effects of exercise in older adults with sarcopenia

Abstract

Aging is associated with progressive physiological changes that impair physical functioning and quality of life (QoL), as exemplified by sarcopenia, a condition characterized by the loss of skeletal muscle mass and strength that increases the risk of falls, dependency, and mortality. Although physical exercise is widely recognized as an effective intervention, the literature remains heterogeneous with respect to the most appropriate modalities for this population. This study aimed to synthesize evidence from randomized controlled trials investigating the effects of physical exercise on skeletal muscle mass, muscle strength, physical performance, QoL, and mortality in older adults with sarcopenia. A systematic search was conducted in PubMed, Embase, CENTRAL, Web of Science, CINAHL, PEDro, SPORTDiscus, and LILACS databases, with no restrictions on language or publication date, up to May 26, 2025. The methodological quality of included studies was assessed using the PEDro scale. Meta-analyses were performed using standardized mean difference or mean difference for post-intervention data, and the certainty of evidence was graded according to the GRADE approach. Sensitivity analyses evaluated the impact of studies with a high risk of bias, and subgroup analyses compared different exercise modalities and diagnostic criteria for sarcopenia. Following systematic screening, 72 studies were included, of which 36 were classified as having a low risk of bias. Physical exercise was superior to control conditions in improving skeletal muscle mass (small effect size), muscle strength, and physical performance (moderate effect size), with no significant effects on QoL. Resistance exercise was more effective than aerobic exercise for increasing skeletal muscle mass (large effect size). The combination of physical exercise and nutritional supplementation yielded greater benefits than either intervention alone for muscle mass and strength and, in sensitivity analyses, also for physical performance. A meta-analysis for mortality was not performed, as only one RCT reported this outcome, representing very limited evidence that precludes any meaningful pooled estimate or definitive conclusions regarding the effects of exercise on mortality in this population. Subgroup analyses stratified by diagnostic criteria demonstrated overall consistent effects across the different sarcopenia definitions. The certainty of evidence ranged from very low to moderate. Physical exercise should be considered a first-line intervention for older adults with sarcopenia, given its potential to improve skeletal muscle mass, muscle strength, and physical performance, supported by moderate-certainty evidence. The effects of exercise combined with nutritional supplementation and the comparative effectiveness of different exercise modalities remain uncertain, as the certainty of evidence for these analyses was low to very low, and future studies are likely to modify these conclusions.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-55850-w.

Introduction

Population aging is one of the defining phenomena of contemporary society, closely linked to declining fertility and mortality rates, resulting in a gradual increase in the proportion of older adults worldwide^1,2. Among the chronic conditions associated with this demographic shift, sarcopenia stands out as one of the most clinically significant. It is classified as a chronic degenerative disease and is recognized in the International Classification of Diseases (ICD-10: M62)^3,4. According to the European Working Group on Sarcopenia in Older People (EWGSOP), sarcopenia is defined as a progressive loss of skeletal muscle mass and strength occurring with advancing age, further aggravated by low physical performance^5,6.

The global prevalence of sarcopenia varies widely according to diagnostic criteria and population characteristics, ranging from 10 to 27% among older adults across different regions of the world, with even higher rates observed in hospital settings and long-term care institutions. The occurrence of severe sarcopenia, although less frequent, has also been reported, affecting approximately 2% to 9% of the older population ⁴. Beyond its substantial magnitude and its tendency to increase with advancing age, sarcopenia has a significant impact on public health: in addition to impairing quality of life (QoL)^7,8, it is associated with several adverse outcomes, including higher risk of falls, fractures, functional disabilities, hospitalizations, and mortality^9–11.

Evidence-based clinical guidelines strongly support the use of exercise as a first-line approach for the management of sarcopenia^12,13. Although resistance training is widely recognized as the primary stimulus for gains in muscle mass and strength in this population^5,6,11,14, important nuances remain unresolved in the literature. These include the relative effectiveness of multicomponent versus isolated training modalities, the role of different training intensities, and the optimal combination of exercise with nutritional supplementation¹⁵. Furthermore, the effects of exercise on broader outcomes such as quality of life and overall functional capacity, which are often among the most clinically meaningful endpoints from the patient’s perspective, have been inconsistently reported across studies^10,16.

Although three recent systematic reviews with network meta-analyses are available^16–18, conflicts in results, methodological limitations, and heterogeneity among the exercise modalities analyzed remain a concern. Two of these reviews^16,18 did not assess the overall certainty of evidence and failed to include a large randomized controlled trial involving 1,519 older adults diagnosed with sarcopenia¹⁹, which could substantially influence the pooled findings. Moreover, the most recent review, conducted by Shen et al.¹⁷, included only studies published in English and, despite its comprehensiveness, reported high heterogeneity among the interventions analyzed.

Given that the findings on the effects of physical exercise in individuals diagnosed with sarcopenia remain diverse and, at times, conflicting with respect to the most effective modality, the present study aimed to conduct a systematic review and meta-analysis to evaluate the effects of different types of physical exercise on relevant outcomes in older adults with sarcopenia, including skeletal muscle mass, muscle strength, physical performance, QoL, and mortality.

Methods

This systematic review and meta-analysis was registered in PROSPERO (CRD420251062085). The methodological development followed the recommendations of the Cochrane Collaboration²⁰ and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist for study reporting²¹.

Inclusion and exclusion criteria

Studies meeting the following criteria were included: randomized controlled trials (RCTs) investigating the effects of physical exercise compared with any type of nutritional intervention, placebo, usual care, or different training modalities in older adults (aged ≥ 60 years) diagnosed with sarcopenia. With respect to the diagnosis of sarcopenia, studies adopting any recognized international consensus criteria or operational definitions based on objective measures of muscle mass, muscle strength, and/or physical performance were considered eligible. Studies relying solely on self-reported sarcopenia were not included.

Studies were excluded for the following reasons: study designs other than RCTs; absence of a physical exercise intervention or exclusive assessment of the effects of nutritional supplementation; failure to assess muscle strength, skeletal muscle mass, physical performance, QoL, or mortality in older adults with sarcopenia; study populations not consisting of older adults with sarcopenia; and inclusion of participants with neurological and/or neurodegenerative diseases.

Databases and search strategy

The electronic search was conducted in the following databases: PubMed, Embase, CENTRAL, Web of Science, CINAHL, PEDro, SPORTDiscus, and LILACS, without applying any filters restricting publication date or language. The search strategy was structured according to the recommendations of the Cochrane Handbook²⁰, encompassing three sets of search terms: population (older adults aged 60 years or older with sarcopenia), intervention (physical exercise), and study design (RCTs). Clinical trial registries (ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform, apps.who.int/trialsearch/) were also consulted to identify potentially unpublished studies. Furthermore, the reference lists of all included studies were manually screened to identify additional records not captured in the database searches. The search strategy, adapted for each database, is provided in Additional File A1. The last search was performed on May 26, 2025.

Study selection

The initial search was performed by one reviewer, who extracted titles and abstracts and removed duplicate records. Subsequently, study selection was conducted independently and in a blinded manner by two reviewers based on the screening of titles and abstracts. Potentially eligible articles were then read in full to determine inclusion in the systematic review and meta-analysis according to the predefined criteria. Disagreements that could not be resolved through discussion were referred to a third reviewer for a final decision. A snowball strategy was additionally applied throughout the review process to identify studies that may have been missed during the database searches.

Data extraction

The following data were extracted from each eligible study: (a) author, year, and country of publication; (b) severity and diagnostic criteria of sarcopenia; (c) intervention performed; (d) population characteristics (mean age, standard deviation, and sex distribution); (e) study duration and setting; (f) outcome assessment instruments (muscle strength, skeletal muscle mass, physical performance, QoL, and mortality); (g) adverse events; and (h) between-group results. A supplementary table was also constructed to describe the intervention protocols and weekly frequency for each included study. Data extraction was independently and blindly performed by both reviewers using a standardized form. When data were unclear or unavailable, the corresponding authors were contacted via email to request the missing information.

Methodological quality assessment

The methodological quality of each RCT was assessed using the PEDro scale (Physiotherapy Evidence Database)²², based on scores available in the database (www.pedro.org.au/search). When a study was not indexed in PEDro, two independent reviewers performed the assessment blindly, with disagreements resolved by a third reviewer. The PEDro scale comprises 11 items, with the first item not contributing to the final score, yielding a total score ranging from 0 to 10 points. Studies with a final score ≥ 6 were classified as having a low risk of bias, while those scoring < 6 were considered to have a high risk of bias. This threshold has been shown to have moderate agreement with the Cochrane Risk of Bias tool, whereas a threshold of 5 points yields only low agreement²³.

Statistical analysis

Effect measures were expressed as the standardized mean difference (SMD) or mean difference (MD) between groups at the post-intervention time point. Heterogeneity was assessed using the Cochrane Q test, with statistical significance defined as p ≤ 0.10, and quantified using the I² statistic. Fixed-effect models were applied when I² was < 50%, and random-effects models were used otherwise²⁰. The effects of interventions were considered statistically significant when p < 0.05, and effect sizes were classified as small (< 0.50), moderate (0.50–0.79), or large (≥ 0.80)²⁰. For physical performance outcomes in which lower values indicate better performance, the scale was inverted prior to analysis so that positive effect sizes consistently represent improvement across all outcomes. Publication bias was assessed using funnel plots when ≥ 10 RCTs were available for the same meta-analysis. All analyses were performed using Review Manager (RevMan), version 5.4 (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen).

Certainty of evidence assessment

The certainty of evidence was rated according to the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach^24,25. Two independent reviewers performed the assessment blindly, and any disagreements were resolved by consensus.

Sensitivity and subgroup analyses

To verify whether studies with a high risk of bias were influencing the primary analyses, sensitivity analyses were performed retaining only studies classified as having high methodological quality (PEDro score ≥ 6). In addition, given the considerable heterogeneity in exercise modalities, combinations with supplementation, and comparators included in the primary analyses, subgroup analyses were performed to independently examine the effects of different exercise modalities and to assess consistency across the diagnostic criteria used to classify sarcopenia, allowing more clinically specific conclusions to be drawn beyond the pooled estimates.

Results

Qualitative synthesis

Included studies

A total of 26,302 records were identified through the initial database search, in addition to 171 clinical trial records. After screening titles and abstracts, 15,681 records were excluded for not meeting the inclusion criteria, and 481 reports were retrieved for full-text assessment. Of these, 19 reports could not be retrieved, as they consisted primarily of conference abstracts and clinical trial records without an associated full text (Additional File A2). Of the 462 studies assessed for eligibility, 386 were excluded (Additional File A3). Seventy-six reports met the inclusion criteria, and one additional report was identified through reference list screening. The present systematic review therefore comprised 77 reports^26–102 and 72 unique studies (Flor-Rufino et al.^41,42, Kemmler et al.^65–68, and Wei et al.^94,95 each represent a single study reported across multiple publications). The PRISMA flow diagram illustrating the processes of identification, screening, and inclusion is presented in Fig. 1.

Fig. 1.

PRISMA Flow Diagram.

Additional File A4 provides a summary of the included studies. Among the 77 reports, 27 were conducted in China^{26–29,31–33,36,37,39,40,48,50,53,54,57,58,69,80–82,84,88,90,92,94,95}; 8 in Brazil^{30,52,72,73,75,89,96,98}; 9 in South Korea^{34,35,49,61,64,71,74,91,93}; 5 in Iran^{38,62,63,70,85}; 4 in Spain^41,42,47,83; 1 in Tunisia⁴³; 4 in Italy^44,78,86,100; 1 in Chile⁴⁵; 1 in Saudi Arabia⁵¹; 1 in Thailand⁵⁵; 6 in Germany^{56,65–68,76}; 5 in Japan^{59,79,97,101,102}; 1 in Turkey⁶⁰; 1 in Singapore⁷⁷; 1 in Greece⁸⁷; 1 in the United States⁹⁹; and one multicenter study conducted across Europe⁴⁶. The earliest included study was published in 2012¹⁰², and the most recent in 2025²⁶.

Characteristics of participants

A total of 6,204 older adults with sarcopenia were analyzed, with mean ages ranging from 61.1 years⁶² to 89.5 years ⁸¹. With respect to the diagnostic criteria adopted, 26 studies used the Asian Working Group for Sarcopenia (AWGS) criteria^{26–29,31–34,36,37,39,40,48,50,51,54,55,58,59,69,71,77,79–82}; 11 studies applied the European Working Group on Sarcopenia in Older People (EWGSOP) criteria^{41,42,57,62,64,72,73,84,86,92,99}; and 1 study combined both AWGS and EWGSOP definitions⁵³.

Additionally, 18 studies diagnosed sarcopenia based on skeletal muscle mass assessment alone^{35,49,56,65–68,74,76,78,85,88,90,91,93–96}; 3 studies used skeletal muscle mass combined with muscle strength^30,38,89; 2 studies assessed muscle strength and physical performance^43,97; 5 studies considered skeletal muscle mass and physical performance^{44,46,63,75,83}; and 5 studies incorporated skeletal muscle mass, muscle strength, and physical performance^{45,52,70,101,103}. Furthermore, 1 study employed the Foundation for the National Institutes of Health (FNIH) criteria⁴⁷; 1 study applied clinical examination, physical performance, and the SARC-F questionnaire⁶⁰; 1 study used both the International Working Group on Sarcopenia (IWGS) and EWGSOP definitions⁶¹; 1 study relied solely on muscle strength⁹⁸; 1 study adopted the Centers for Disease Control and Prevention (CDC) criteria¹⁰⁰; and in 1 study, the diagnostic criterion was not reported ⁸⁷.

Fifty-five studies enrolled participants with sarcopenia exclusively^{26–34,36–42,44–48,50–55,58–62,64,69–75,77–83,86,87,91,94,95,100–102}; 12 included participants with sarcopenic obesity^{35,43,49,84,88,89,92,93,96–99}; 7 involved individuals with osteosarcopenia^{56,57,65–68,76}; 2 included participants with osteosarcopenic obesity^63,82; and 1 assessed participants with class I sarcopenia ⁸⁵. Study populations were drawn from community settings^{26,27,29–33,35,36,53,56–68,70,72–82,84–91,93–102}, hospital settings^{28,48,51,54,69,71,92}, and long-term care facilities^34,47,55,83.

Intervention

Additional File A5 summarizes the interventions. The total duration of the intervention period ranged from 3 weeks⁷¹ to 36 months⁴⁶. Weekly session frequency varied from once⁸⁰ to five times per week^36,64,81, with three studies not reporting this information^48,50,62. Session duration ranged from 25 min^62,74,91 to 120 min⁹⁸; 20 studies did not report session duration^{33,37,39,46–48,52,55,56,59,65–68,72,73,76,94–96}.

The number of comparison groups per study was two^{26,27,29,30,35,37–47,49–52,55–57,60,61,63–69,71,73–78,82,84,86,90–93,96–99}, three^{27,31–33,54,59,62,70,72,80,81,83,87,89}, or four^{34,36,48,53,58,79,85,88,94,95,97,100–102}, with the following interventions employed:

1. Multicomponent Exercise^{26,35,44,46,47,49,60,64,87,101,102};

2. Multicomponent Exercise plus Supplementation^101,102;

3. Yi Jin Jing Exercise²⁷;

4. Tai Chi Exercise^27,32,81;

5. Resistance Exercise^{28–31,33,34,36,38,39,41,42,45,48,52,54,57,59,61,63,65–73,74,75,78,79,82–86,88–90,96,98–100};

6. Aerobic Exercise^45,88;

7. Whole-Body Vibration^{29,81,94,95,102};

8. Exercise with Focal Mechanic-Acoustic Vibratory Therapy¹⁰²;

9. Resistance Exercise plus Other Interventions (Tai Chi^31,40; Yi Jin Jing^33,54; Supplementation^{36,48,56,59,72,76,79,85}; Aerobic exercise^{37,51,53,55,58,62,70,74,80,88,91–93,97}; Postural exercise⁴³; Supplementation plus aerobic exercise^53,58,80,97; Balance⁶⁹);

10. Supplementation (amino acids, calcium, protein, vitamin D, unspecified, iron, epicatechin)^{36,48,53,56,58,59,65–68,76,79,85,97,101,102};

11. Ditangquan Exercise⁵⁰;

12. Physiotherapy⁷¹;

13. Resistance Exercise plus Balance, Cognitive Training, and Supplementation⁷⁷;

14. Respiratory Muscle Training⁸³;

15. Postural Exercise⁸⁶;

16. Control Groups^{26,31,33–38,41–44,46,48,49,52–55,57,58,60–64,70,72,74,75,77–85,87–98,100–102}.

With respect to adverse events, the following were reported across the included studies: muscle pain^{26,27,37,63,98}, abdominal distension³⁶, falls⁴⁶, joint pain^{55,63,67,68,99}, and death⁴⁶. Seventeen studies reported no adverse events^{27,29,39,40,57,61,65,66,69,76,79–81,87,91,95,97}, while 49 studies did not report this information^{30–35,38,41–45,47–54,56,58–60,62,64,70–75,77,78,82–86,88–91,93,94,96,100–102}.

Assessment Instruments (skeletal muscle mass, muscle strength, physical performance, QoL, mortality)

The following methods were used to assess skeletal muscle mass: dual-energy X-ray absorptiometry (DXA)^{30,46–48,52,57,61,63,65–68,77,80,81,84,85,89,90,92,96}; bioelectrical impedance analysis (BIA)^{26–29,32,33,36–45,53,55,58,59,62,64,70,73,75,76,82,83,86,88,93,97,99,101,102}; ultrasonography^44,94; magnetic resonance imaging (MRI)⁵¹; computed tomography (CT)³⁴; BIA and DXA^35,49,74; CT and BIA^31,54; BIA, DXA, and MRI⁵⁶; BIA and ultrasonography ^78,79; and BIA combined with calf circumference⁸⁷. Nine studies did not assess skeletal muscle mass^{50,60,69,71,72,91,95,98,100}.

Muscle strength was assessed using the following methods: one-repetition maximum (1RM)^{30,62,70,85,89}; handgrip dynamometer^{26,27,31–33,36,38,40,42,44,46–48,51,53–55,57–59,64,67,69,71,73,75,76,81,83,84,86,92,93,97}; unspecified dynamometer^{43,77,101,102}; handgrip and isokinetic dynamometer^{28,37,61,87,91}; 1RM and handgrip dynamometer^29,39,78,99; handgrip dynamometer combined with leg extension machine and leg press⁴¹; handgrip dynamometer and submaximal leg press test⁴⁵; 1RM, handgrip dynamometer, and isokinetic dynamometer⁵²; handgrip dynamometer and 30-s sit-to-stand test⁶³; isokinetic dynamometer^{65,66,68,74,94,96,98}; handgrip dynamometer and isometric knee extension^79,80; handgrip dynamometer and back extension strength⁸²; handgrip dynamometer, back extension strength, and isokinetic dynamometer⁸⁸; maximal isometric test¹⁰⁰. Nine studies did not assess muscle strength^{34,35,49,50,56,60,72,90,95}.

Physical performance was assessed using the following instruments: Short Physical Performance Battery (SPPB)^{26,29,39,41,42,44,46,47,52,69,78,98,99}; chair rise test^26,75,91; walking tests (3, 4, 5, 6, and 10 m; 6-min walk)^{27–29,32,36–43,45,46,53–55,57–61,63,64,67,69,74–77,79–81,83,84,87,92,93,95,97,98,101,102}; Timed Up and Go test (TUG)^{27,28,32,37,38,43,45,47,50,57,60,63,64,69,75,80,81,84,85,87,91,92,95,101}; 30-s arm curl test^28,37; sit-to-stand test^{28,29,37,38,40,47,53,57,64,80,81,84,87,93,95}; Senior Fitness Test (SFT)⁶¹; 8-Foot Up-and-Go test (8FUAG)⁶¹; step-up test^64,78; sit-and-reach test⁶⁴; Functional Ambulation Category (FAC)⁷¹; and Get Up and Go test (GUG)⁷⁸. Twenty-four studies did not assess physical performance^{30,31,33–35,48,49,51,56,62,65,66,68,70,72,73,82,86,88–90,95,96,100}.

Quality of life was assessed using the following instruments: Short Form-36 (SF-36)^{27,29,32,37,39,59,84,98}; MacNew Questionnaire ²⁶; EQ-5D^42,71,72; Sarcopenia Quality of Life questionnaire (SarQoL)^51,87; numerical rating scale (0–10)⁶⁰; and Short Form-12 (SF-12)⁸⁰. Sixty-one studies did not evaluate this outcome^{28,30,31,33–36,38,40,41,43–50,52–58,61–70,73–79,81–83,85,86,88–97,99–102}. Regarding mortality, only one study reported this outcome⁴⁶.

Intergroup results

The following intergroup results were observed for skeletal muscle mass:

• In favor of resistance exercise vs. control^{31,34,38,41,42,48,82,84,88,89,96};

• In favor of aerobic exercise vs. control⁸⁸;

• In favor of multicomponent exercise vs. control^{35,46,49,64,87};

• In favor of resistance exercise plus supplementation vs. control^48,79;

• In favor of resistance exercise plus Tai Chi vs. control^31,40;

• In favor of resistance exercise plus Yi Jin Jing vs. control^33,54;

• In favor of resistance exercise plus postural exercise vs. control⁴³;

• In favor of resistance exercise plus aerobic exercise vs. control^{53,58,62,70,74,80,88,92};

• In favor of resistance exercise plus aerobic exercise plus supplementation vs. control^58,80.

For muscle strength, the following intergroup differences were observed:

• In favor of resistance exercise vs. control^{31,38,41,42,48,52,61,63,82–85,88,89,96};

• In favor of whole-body vibration vs. control^81,94;

• In favor of multicomponent exercise vs. control^26,46,87;

• In favor of aerobic exercise vs. control⁸⁸;

• In favor of resistance exercise plus supplementation vs. control^48,85;

• In favor of resistance exercise plus Tai Chi vs. control^31,40;

• In favor of resistance exercise plus Yi Jin Jing vs. control³³;

• In favor of resistance exercise plus aerobic exercise vs. control^{37,55,58,62,70,74,80,88,91–93};

• In favor of resistance exercise plus postural exercise vs. control⁴³;

• In favor of resistance exercise plus aerobic exercise plus supplementation vs. control^58,80;

• In favor of multicomponent exercise plus supplementation vs. control¹⁰².

For physical performance, the following intergroup differences were observed:

• In favor of resistance exercise vs. control^{38,61,78,84,85};

• In favor of whole-body vibration vs. control⁹⁵;

• In favor of multicomponent exercise vs. control^{26,44,46,60,64,87,102};

• In favor of resistance exercise plus supplementation vs. control^79,85;

• In favor of resistance exercise plus aerobic exercise vs. control^{37,55,74,80,91–93};

• In favor of resistance exercise plus Tai Chi vs. control⁴⁰;

• In favor of resistance exercise plus postural exercise vs. control⁴³;

• In favor of resistance exercise plus aerobic exercise plus supplementation vs. control⁸⁰;

• In favor of multicomponent exercise plus supplementation vs. control^101,102.

For quality of life, the following differences were observed:

• In favor of resistance exercise vs. control⁴²;

• In favor of multicomponent exercise vs. control^60,87.

Regarding mortality, no significant differences were observed between the multicomponent exercise and control groups.

Methodological quality of the studies

The methodological quality assessment, presented in Table 1, identified 36 studies with a low risk of bias (PEDro score ≥ 6). The mean PEDro score across the included RCTs was 5.71 ± 1.72.

Table 1.

Methodological quality of the included studies.

Study	C1†	C2	C3	C4	C5	C6	C7	C8	C9	C10	C11	Score‡ (0–10)
Li et al., 2025	+	+	+	+	−	-	+	+	+	+	+	8
Wei et al., 2025	+	+	-	+	-	-	-	-	-	+	+	4
Zhang et al., 2025	+	+	+	+	-	-	+	+	-	+	+	7
Zhuang et al., 2025	+	+	+	+	-	-	+	+	+	+	+	8
Dos Santos et al., 2024	+	+	-	-	-	-	-	+	+	+	+	5*
Guo et al., 2024	+	+	-	+	-	-	-	+	-	+	+	5
He et al., 2024	+	+	+	+	-	-	-	+	-	+	+	6*
He et al., 2024b	+	+	+	+	-	-	-	-	-	+	+	5*
Heo & Jee, 2024	+	+	-	+	-	-	-	+	-	+	+	5*
Jung et al., 2024	-	+	-	+	-	-	-	+	-	+	+	5*
Liao et al., 2024	-	+	-	+	-	-	-	-	-	+	+	4
Liu et al., 2024	+	+	+	+	-	-	+	-	+	+	+	7*
Rezaei et al., 2024	+	+	-	+	-	-	-	-	-	+	+	4*
Zhang et al., 2024	+	+	+	+	-	-	+	+	+	+	+	8
Chen et al., 2023	+	+	+	+	-	-	+	-	+	+	+	7*
Flor-Rufino et al., 2023, 2023b	+	+	+	+	-	-	+	-	-	+	+	6*
Magtouf et al., 2023	+	+	-	+	-	-	+	+	-	+	+	6*
Monti et al., 2023	-	+	-	+	-	-	-	-	-	+	+	4*
Valdes-Badilla et al., 2023	-	+	-	+	-	-	-	+	-	+	+	5*
Bernabei et al., 2022	+	+	-	+	-	-	+	-	+	+	+	6*
Courel-Ibanez et al., 2022	+	+	-	+	-	-	-	+	-	+	+	5*
Han et al., 2022	+	+	-	+	-	-	-	+	-	+	+	5*
Jung et al., 2022	-	+	-	+	-	-	-	+	+	+	+	6*
Li et al., 2022	+	+	+	+	-	-	-	+	+	+	+	7*
Nambi et al., 2022	+	+	+	+	-	-	+	+	-	+	+	7*
Souza et al., 2022	+	+	-	+	-	-	+	+	+	+	+	7*
Wang et al., 2022	+	+	-	-	-	-	-	+	-	+	+	4
Wei et al., 2022	+	+	-	+	-	-	-	+	-	+	+	5*
Yuenyongchaiwat & Akekawatchai 2022	-	+	-	+	-	-	+	+	+	+	+	7*
Ghasemikaram et al., 2021	+	+	-	+	-	-	-	-	-	+	+	4*
Lee et al., 2021	+	+	+	+	-	-	+	+	+	+	+	8*
Li et al., 2021	+	+	-	+	-	-	-	-	+	+	+	5*
Mori & Tokuda 2021	-	+	-	+	-	-	-	+	-	+	+	5
Sen et al., 2021	+	+	+	+	-	-	-	+	-	+	+	6*
Seo et al., 2021	+	+	-	+	-	-	-	-	-	+	+	4*
Bagheri et al., 2020	-	+	-	+	-	-	-	-	-	+	+	4*
Banitalebi et al., 2020	+	+	+	+	-	-	+	-	+	+	+	7*
Jeon & Kim 2020	+	+	+	+	-	-	-	+	-	+	+	6*
Kemmler et al., 2020, 2020b, 2020c, 2020d	+	+	+	+	-	-	+	+	+	+	+	8*
Liang et al., 2020	-	+	-	+	-	-	+	+	+	+	+	7*
Moghadam et al., 2020	+	+	-	+	-	-	-	-	-	+	+	4
Oh et al., 2020	+	+	+	+	-	-	+	+	-	+	+	7*
Pinheiro et al., 2020	+	+	+	+	-	-	-	-	-	+	+	5
Freitas et al., 2019	-	+	-	+	-	-	-	+	+	+	+	6
Jung et al., 2019	+	+	-	+	-	-	-	-	-	+	+	4*
Letieri et al., 2019	-	+	-	+	-	-	-	+	-	-	-	3*
Lichtenberg et al., 2019	-	+	+	+	-	-	+	+	+	+	+	8*
Lu et al., 2019	-	+	-	+	-	-	-	+	+	+	+	6*
Vezzoli et al., 2019	-	+	-	+	-	-	-	+	-	+	+	5*
Yamada et al., 2019	+	+	-	+	-	-	+	-	+	+	+	6*
Zhu et al., 2019	+	+	-	+	-	-	-	-	+	+	+	5*
Zhu et al., 2019b	+	+	-	+	-	-	-	+	-	+	+	5*
Chen et al., 2018	-	+	-	+	-	-	-	-	-	+	+	4*
Iranzo et al., 2018	+	+	-	+	-	-	+	-	-	+	+	5*
Liao et al., 2018	+	+	+	+	-	-	+	+	+	+	+	8
Mafi et al., 2018	-	+	-	+	-	-	-	+	-	+	+	5*
Piastra et al., 2018	+	+	-	+	-	-	-	+	+	+	+	6*
Tsekoura et al., 2018	+	+	+	+	-	-	-	+	+	+	+	7*
Chen et al., 2017	-	+	-	+	-	-	+	-	-	+	-	4*
Cunha et al., 2017	+	+	+	+	-	-	+	+	-	+	+	7*
Huang et al., 2017	+	+	+	+	-	-	+	+	-	+	+	7*
Jung et al., 2017	-	+	-	+	-	-	-	-	-	+	+	4
Liao et al., 2017	+	+	-	+	-	-	+	+	+	+	+	7*
Park et al., 2017	+	+	-	+	-	-	-	+	-	+	+	5*
Wei et al., 2017, 2017b	+	+	-	+	-	-	-	+	+	+	+	6*
Gadelha et al., 2016	+	+	-	+	-	-	-	+	-	+	+	5*
Kim et al., 2016	+	+	-	+	-	-	-	+	-	+	+	5*
Vasconcelos et al., 2016	+	+	+	+	-	-	+	+	+	+	+	8*
Balachandran et al., 2014	+	+	-	+	-	-	+	-	-	+	+	5*
Bellomo et al., 2013	-	+	-	-	-	-	+	+	-	-	+	4*
Kim et al., 2013	+	+	+	+	-	-	+	+	-	+	+	7*
Kim et al., 2012	+	+	+	+	-	-	+	+	-	+	+	7*

^†: Item not included in the scoring; ‡ Overall mean (SD): 5.71 (1.72); + : Criterion met; -: Criterion not met; C1: eligibility criteria were specified; C2: subjects were randomly allocated to groups; C3: allocation was concealed; C4: the groups were similar at baseline regarding the most important prognostic indicators; C5: there was blinding of all subjects; C6: there was blinding of all therapists who administered the therapy; C7: there was blinding of all assessors who measured at least one key outcome; C8: measures of at least one key outcome were obtained from more than 85% of the subjects initially allocated to groups; C9: all subjects for whom outcome measures were available received the treatment or control condition as allocated or, where this was not the case, data for at least one key outcome was analysed by “intention to treat”; C10: the results of between-group statistical comparisons are reported for at least one key outcome; C11: the study provides both point measures and measures of variability for at least one key outcome; *: Item scored according to the assessment available in the PEDro database.

Quantitative synthesis of the studies (meta-analysis)

All forest plots corresponding to the primary analyses are presented in Additional File A6. For each meta-analysis, the certainty of evidence was assessed using the GRADE system. Across the different analyses, the certainty of evidence ranged from very low to moderate, with the primary reasons for downgrading being risk of bias, inconsistency, imprecision, and potential publication bias (Additional File A7).

Publication bias was assessed by visual inspection of funnel plots for the following comparisons: exercise vs. control (skeletal muscle mass, muscle strength, and physical performance), exercise plus supplementation vs. supplementation (skeletal muscle mass, muscle strength, and physical performance), and exercise vs. supplementation (skeletal muscle mass and muscle strength). Evidence of publication bias was identified in the analyses of exercise plus supplementation vs. supplementation (skeletal muscle mass and muscle strength) and exercise vs. supplementation (muscle strength) (Additional File A8).

Figures 2, 3, 4, 5 and Table 2 present the primary analyses. Moderate-certainty evidence (downgraded for inconsistency) indicated that physical exercise is significantly superior to control conditions in improving skeletal muscle mass (small effect size) (Fig. 2), muscle strength (Fig. 3), and physical performance (moderate effect size) (Fig. 4). For QoL (Fig. 5), no significant differences were observed between physical exercise and control groups, with very low-certainty evidence (downgraded for inconsistency and imprecision). In the comparative analysis between exercise modalities, resistance exercise was superior to aerobic exercise for skeletal muscle mass (large effect size), whereas no significant differences were found for muscle strength; both analyses were rated as very low-certainty evidence (downgraded for risk of bias and imprecision).

Fig. 2.

Meta-analysis comparing exercise vs. control for skeletal muscle mass.

Fig. 3.

Meta-analysis comparing exercise vs. control for muscle strength.

Fig. 4.

Meta-analysis comparing exercise vs. control for physical performance.

Fig. 5.

Meta-analysis comparing exercise vs. control for QoL.

Table 2.

Primary analyses.

Primary analyses	Number of studies	Number of participants	Effect size (95% CI)			p
Exercise plus supplementation vs. supplementation
Skeletal muscle mass	13	774		0.45 (0.16, 0.74)	0.002
Muscle strength	12	738		0.45 (0.10, 0.81)	0.01
Physical performance	10	552		0.13 (-0.14, 0.40)	0.35
Exercise vs. supplementation
Skeletal muscle mass	10	646		0.04 (-0.27, 0.34)	0.82
Muscle strength	10	646		-0.00 (-0.16,0.15)	0.96
Physical perfomance	8	469		0.17 (-0.22, 0.55)	0.40
Resistance exercise vs. aerobic exercise
Skeletal muscle mass	2	70		0.93 (0.43, 1.43)	0.0001
Muscle strength	2	70		0.46 (-0.01, 0.94)	0.06

When physical exercise was combined with supplementation, it was superior to supplementation alone in improving skeletal muscle mass and muscle strength, with low-certainty evidence (downgraded for inconsistency and publication bias) and a small effect size. For physical performance, no significant differences were observed between these interventions, supported by low-certainty evidence (downgraded for risk of bias and imprecision). In the direct comparison between physical exercise and supplementation alone, no significant differences were found for any of the analyzed outcomes. Skeletal muscle mass and physical performance were supported by very low-certainty evidence (downgraded for risk of bias, inconsistency, and imprecision), whereas muscle strength was supported by low-certainty evidence (downgraded for imprecision and publication bias).

Sensitivity analysis

To assess whether studies with low methodological quality were influencing the primary analyses, sensitivity analyses were conducted retaining only studies with a low risk of bias (PEDro score ≥ 6) (Additional File A9). In the comparison between physical exercise and control, the previously observed benefits for skeletal muscle mass, muscle strength, and physical performance were maintained. In the comparison between physical exercise combined with supplementation and supplementation alone, the effects identified in the primary analysis for skeletal muscle mass and muscle strength remained unchanged; however, physical performance also became significant in favor of the exercise plus supplementation combination. In the direct comparison between physical exercise and supplementation, the non-significant result for muscle strength was maintained, whereas skeletal muscle mass became favorable to supplementation and physical performance showed a significant effect in favor of exercise.

Subgroup analysis

Subgroup analyses examining different exercise modalities are presented in Table 3, with corresponding forest plots in Additional File A10. When resistance exercise, multicomponent exercise, whole-body vibration, and resistance exercise plus other interventions (aerobic exercise, Tai Chi, and postural exercise) were analyzed separately versus control, all modalities except whole-body vibration demonstrated significant effects for skeletal muscle mass (small effect size), muscle strength (small to large effect size), and physical performance (moderate effect size). The exception was resistance exercise plus other interventions, which did not reach statistical significance for physical performance.

Table 3.

Subgroup analyses.

Subgroup	Number of studies	Number of participants	Effect size (95% CI)	p	Q statistic
Exercise vs. control for skeletal muscle mass
Resistance exercise	21	1053	0.45 (0.22, 0.67)	0.0001	Q = 3.86 df = 3 p = 0.28
Multicomponent exercise	7	290	0.35 (0.11, 0.58)	0.004
Whole-body vibration	2	95	-0.01 (-0.42, 0.39)	0.94
Resistance exercise plus other interventionsa	11	622	0.38 (0.04, 0.71)	0.03
Exercise vs. control for muscle strength
Resistance exercise	21	951	0.96 (0.69, 1.22)	< 0.00001	Q = 20.46 df = 3 p = 0.0001
Multicomponent exercise	6	319	0.43 (0.20, 0.65)	0.0002
Whole-body vibration	2	119	-0.06 (-0.43, 0.31)	0.75
Resistance exercise plus other interventionsa	12	661	0.54 (0.22, 0.85)	0.001
Exercise vs. control for physical performance
Resistance exercise	13	424	0.60 (0.14, 1.06)	0.01	Q = 1.72 df = 3 p = 0.63
Multicomponent exercise	7	427	0.68(0.30, 1.07)	0.0005
Whole-body vibration	1	80	0.33 (-0.18, 0.84)	0.20
Resistance exercise plus other interventionsa	9	505	0.37 (-0.09, 0.83)	0.11
Exercise plus supplementation vs. supplementation alone for skeletal muscle mass
Resistance exercise	11	506	1.00 (0.49, 1.52)	0.0001	Q = 9.12 df = 2 p = 0.01
Multicomponent exercise	2	129	0.32 (-0.03, 0.67)	0.07
Resistance Exercise plus Aerobic	3	268	0.11 (-0.16, 0.38)	0.43
Exercise plus supplementation vs. supplementation alone for muscle strength
Resistance exercise	10	470	1.31 (0.69, 1.93)	< 0.0001	Q = 16.25 df = 2 p = 0.0003
Multicomponent exercise	2	129	0.43 (0.08, 0.78)	0.02
Resistance Exercise plus Aerobic	3	268	-0.21 (-0.63, 0.21)	0.33
Exercise plus supplementation vs. supplementation alone for physical performance
Resistance exercise	6	254	0.08 (-0.40, 0.57)	0.74	Q = 3.38 df = 2 p = 0.18
Multicomponent exercise	2	129	0.41 (0.06, 0.76)	0.02
Resistance Exercise plus Aerobic	2	169	-0.02 (-0.32, 0.28)	0.89
Exercise vs. supplementation for skeletal muscle mass
Resistance exercise	5	259	-0.04 (-0.41, 0.33)	0.84	Q = 22.61 df = 2 p < 0.0001
Multicomponent exercise	2	132	0.73 (0.37, 1.08)	< 0.0001
Resistance Exercise plus Aerobic	3	255	-0.32 (-0.57, -0.07)	0.01
Exercise vs. supplementation for muscle strength
Resistance exercise	5	259	-0.04 (-0.29, 0.20)	0.74	Q = 2.59 df = 2 p = 0.27
Multicomponent exercise	2	132	0.24 (-0.10, 0.59)	0.16
Resistance Exercise plus Aerobic	3	255	-0.09 (-0.34, 0.15)	0.46
Exercise vs. supplementation for physical performance
Resistance exercise	4	170	0.05 (-0.54, 0.65)	0.86	Q = 8.09 df = 2 p = 0.02
Multicomponent exercise	2	132	0.72 (0.37, 1.07)	< 0.0001
Resistance Exercise plus Aerobic	2	167	-0.17 (-0.77, 0.42)	0.56

^a.Aerobic exercise, Tai Chi, and postural exercise.

In the subgroup analysis comparing different modalities of exercise combined with supplementation versus supplementation alone, significant effects in favor of resistance exercise (large effect size) and multicomponent exercise (small effect size) were observed for both skeletal muscle mass and muscle strength, while for physical performance, only multicomponent exercise (small effect size) reached statistical significance. Resistance plus aerobic exercise did not reach statistical significance for any of the three outcomes.

In the analysis of physical exercise versus supplementation alone, when resistance exercise, multicomponent exercise, and resistance plus aerobic exercise were examined separately, significant effects were observed only in favor of multicomponent exercise for skeletal muscle mass and physical performance (moderate effect size). When resistance plus aerobic exercise was compared with supplementation, a significant effect was observed in favor of supplementation for skeletal muscle mass (small effect size).

Regarding the subgroup analyses examining whether the effects of exercise are consistent across different diagnostic definitions of sarcopenia, a significant between-subgroup difference was observed in only one analysis, specifically the comparison of exercise plus supplementation vs. supplementation alone for muscle strength (Additional File A11).

Discussion

The present findings indicate that physical exercise was, overall, superior to control conditions in improving skeletal muscle mass, muscle strength, and physical performance, aligning with previous systematic reviews and meta-analyses that have consistently reported benefits of these interventions in individuals with sarcopenia^11,104–106. However, consistent with the observations by Shen et al.¹⁰³, no significant effects on QoL were identified, a result that may be explained by the multidimensional nature of this outcome, which encompasses physical, psychological, and social components, as well as by the limited number of trials assessing it, often using heterogeneous and insufficiently standardized instruments^17,103.

In this regard, it is noteworthy that most studies employed generic instruments such as the SF-36^{27,29,32,37,39,59,84,98}, which may lack the sensitivity required to detect sarcopenia-specific changes in QoL. Only two studies used the SarQoL questionnaire^51,87, a disease-specific tool recommended by major international consensus groups and considered more likely to capture clinically meaningful variations in this population^12,17. Furthermore, the duration of most included interventions may have been insufficient to produce meaningful effects on the psychological and social dimensions of QoL, which typically require longer periods to respond to physical training. Collectively, these factors may explain why improvements in muscle mass and strength did not translate into statistically significant gains in QoL. Future studies should prioritize longer follow-up periods and validated disease-specific instruments to more adequately capture this outcome.

The superiority of resistance exercise over aerobic exercise for increasing skeletal muscle mass reinforces established evidence in the literature. Resistance training is widely regarded as the primary stimulus for hypertrophy, as it enhances neuromuscular activation and promotes favorable metabolic adaptations, and is considered the gold-standard approach for preserving lean mass in older adults^107,108. In contrast, isolated aerobic exercise, although important for cardiovascular health¹⁰⁹, generally fails to provide a sufficient hypertrophic stimulus, which accounts for its more modest effects on muscle mass^110,111. These results also corroborate findings from subgroup analyses of other reviews demonstrating greater effectiveness of resistance or multicomponent programs for all assessed outcomes^53,112–115. Conversely, the lack of significant effects observed for whole-body vibration likely reflects the considerable methodological heterogeneity of the available studies, which hampers direct comparisons and reduces the ability to identify consistent benefits^116–118.

Regarding combined interventions, the association of physical exercise with nutritional supplementation produced superior effects compared with supplementation alone for muscle mass and strength, and, in sensitivity analyses, also for physical performance. This synergistic effect has been suggested by prior studies and international guidelines, which propose that exercise provides the mechanical stimulus necessary for muscle protein synthesis while supplementation offers the substrates that potentiate this response^12,17,18. Nevertheless, the direct comparison between exercise and supplementation alone did not reveal significant differences for any of the analyzed outcomes, a finding consistent with network meta-analyses that similarly identified no marked differences between these isolated strategies ¹⁸. However, when analyses were restricted to studies with high methodological quality, supplementation was superior for muscle mass whereas exercise yielded better results for physical performance, a pattern that may reflect the distinct nature of the underlying mechanisms: supplementation more directly influences body composition parameters ¹¹⁹, while exercise promotes both structural and neuromuscular adaptations that translate into improved functional performance^113–115.

It is also worth considering whether the additional benefits observed with the combination of exercise and supplementation may reflect a ceiling effect of exercise alone, particularly among individuals already engaged in structured training programs, in whom further gains in muscle mass and strength may be limited without adequate nutritional support. Moreover, the benefits of supplementation are likely not uniform across all individuals with sarcopenia. Older adults with insufficient protein intake or nutritional deficiencies may represent a subgroup particularly responsive to the combined intervention, given that the anabolic stimulus provided by exercise cannot be fully translated into muscle protein synthesis in the absence of adequate substrate availability^12,17,18. This hypothesis is consistent with the heterogeneity observed across the included studies and underscores the importance of incorporating nutritional screening into the clinical assessment of older adults with sarcopenia, so that supplementation can be targeted to those most likely to benefit.

The setting in which exercise interventions were delivered also deserves consideration, as community-dwelling older adults^{26,27,29–33,35,36,53,56–68,70,72–82,84–91,93–102}, hospitalized individuals^{28,48,51,54,69,71,92}, and long-term care facility residents^34,47,55,83 represent populations with distinct functional profiles, levels of supervision, and responsiveness to training. The majority of included studies were conducted in community settings, where exercise interventions consistently demonstrated benefits across all primary outcomes. In contrast, studies conducted in hospital settings showed more heterogeneous results, possibly reflecting shorter intervention durations, the acute clinical context, and the potential interference of post-surgical recovery on the adaptive response to exercise. Studies involving long-term care facility residents, although fewer in number, suggested that structured exercise programs can produce meaningful improvements in muscle strength and physical performance even in more dependent and frail populations. These observations reinforce the importance of tailoring exercise prescription to the specific clinical setting and functional status of the patient, and highlight the need for future studies to stratify their analyses by care setting to better characterize the effects of exercise across different contexts of sarcopenia management.

Finally, the present findings reinforce the central role of physical exercise in the management of sarcopenia, consistent with the recommendations of the International Clinical Practice Guidelines for Sarcopenia (ICFSR)¹². Nonetheless, the certainty of evidence ranged from very low to moderate, which warrants caution in the interpretation of results and underscores the need for future studies with greater methodological rigor.

Limitations and strengths

Despite the robustness of this review, several limitations must be acknowledged. The search was not conducted across all existing databases, but was restricted to those most relevant to the outcomes of interest. This limitation was partially mitigated by systematically screening the reference lists of all included studies.

The heterogeneity among exercise protocols, encompassing modality, intensity, duration, and frequency, hindered direct comparisons and may have influenced the observed effects. In addition, the use of different diagnostic criteria for sarcopenia and various outcome assessment methods across studies increases the variability of the results. To address this, subgroup analyses stratified by diagnostic criteria were performed, and the effects of exercise were, overall, consistent across the different sarcopenia definitions. It is also worth noting that some included studies diagnosed sarcopenia based on combinations of parameters that did not systematically require reduced muscle mass, which may introduce overlap with the construct of dynapenia, defined as the age-related loss of muscle strength independent of muscle mass loss. Although this distinction was not formally addressed in the included studies, the overall consistency of effects observed across different diagnostic criteria in the subgroup analyses suggests that the clinical relevance of exercise interventions extends across different phenotypes of muscle deterioration in older adults.

Adverse event reporting was also inconsistent across the included studies, with 49 studies providing no information on this outcome^{30–35,38,41–45,47–54,56,58–60,62,64,70–75,77,78,82–86,88–91,93,94,96,100–102}, which limits the ability to fully characterize the safety profile of the interventions analyzed.

The scarcity of RCTs assessing QoL represents an additional limitation. When this outcome was evaluated, the instruments used showed considerable methodological variation and were frequently not aligned with the recommendations of the most widely accepted international consensus groups, such as AWGS, EWGSOP, and IWGS. This heterogeneity may have contributed to the inconsistency of findings in this domain. The limited reporting of mortality as an outcome similarly precluded more comprehensive analyses of this clinically important endpoint.

On the other hand, through a comprehensive and unrestricted search strategy, this review compiled the most complete evidence currently available on exercise interventions for older adults with sarcopenia, encompassing a substantial sample of this population and including exclusively RCTs, thereby reducing the risk of bias. No restrictions regarding language or year of publication were applied, further enhancing the representativeness of the included evidence. An additional strength is that this review encompassed all modalities of physical exercise investigated in older adults with sarcopenia.

Clinical implications and future perspectives

From a clinical standpoint, the findings of this review provide actionable guidance for the management of sarcopenia in older adults. The moderate effect sizes observed for muscle strength and physical performance indicate clinically meaningful improvements, as gains in these outcomes are directly associated with a reduced risk of falls, functional dependency, and hospitalizations^12,120. Resistance training should be considered the cornerstone of treatment given its consistent and large effects on muscle mass, while multicomponent programs may be recommended when the primary goal is to optimize physical performance. When nutritional deficiency is suspected or confirmed, the combination of exercise with supplementation should be prioritized, as it produced superior effects for muscle mass and strength, with additional benefits for physical performance observed in high-quality studies. These findings reinforce the need for individualized, multimodal strategies in which exercise prescription and nutritional support are tailored to the specific profile of each patient.

Regarding safety, the adverse events reported across the included studies were generally mild and transient, including muscle pain^{26,27,37,63,98}, joint pain^{55,63,67,68,99}, and abdominal distension³⁶, with falls and one death reported in the context of a multicomponent exercise program in a frail population⁴⁶. These findings suggest that exercise interventions are broadly safe for older adults with sarcopenia when appropriately prescribed and supervised. Nonetheless, the high proportion of studies that did not report adverse events limits the ability to draw definitive conclusions about the safety profile of each exercise modality, and standardized reporting of adverse events should be prioritized in future trials.

Further research is needed to promote greater standardization of intervention protocols, particularly with respect to training intensity, duration, and the direct comparison of different exercise modalities. Likewise, long-term studies are essential to assess clinically relevant outcomes such as hospitalizations, institutionalization, and mortality. Notably, mortality could not be analyzed in the present review due to the absence of data in the included trials, representing a significant gap that should be addressed in future investigations.

In addition, the diagnostic criteria for sarcopenia adopted across the included studies varied considerably, and not all followed the most widely accepted international consensus recommendations, such as those of AWGS, EWGSOP, and IWGS, underscoring the need for greater methodological standardization in future research.

It is also important to emphasize the use of validated instruments recommended by international consensus groups for the assessment of sarcopenia-related outcomes, given the considerable methodological heterogeneity observed among the included studies. According to these recommendations, skeletal muscle mass should preferably be assessed using DXA, BIA, or magnetic resonance imaging; muscle strength should be evaluated using handgrip dynamometry or the chair stand test; physical performance should be assessed primarily through gait speed or the SPPB; and QoL should be measured using the SarQoL questionnaire, the instrument specifically recommended for older adults with sarcopenia.

Conclusion

Physical exercise is supported as a first-line intervention for older adults with sarcopenia, promoting clinically meaningful gains in muscle mass, strength, and physical performance, particularly through resistance training or multicomponent programs, based on moderate-certainty evidence. The potential additional benefits of combining exercise with nutritional supplementation, as well as the comparative effectiveness of different exercise modalities, should be interpreted with caution given the low to very low certainty of evidence for these analyses, and future studies are likely to modify these conclusions. Long-term studies assessing clinically relevant outcomes such as hospitalizations, institutionalization, and mortality remain a priority for the field.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

QoL

Quality of life

RCT

Randomized controlled trials

EWGSOP

European working group on sarcopenia in older people

TUG

Timed up and go

SMD

Standardized mean difference

MD

Mean difference

AWGS

Asian working group for sarcopenia

IWGS

International working group on sarcopenia

CDC

Centers for disease control and prevention

DXA

Dual-energy X-ray absorptiometry

BIA

Bioelectrical impedance analysis

MRI

Magnetic resonance imaging

CT

Computed tomography

1RM

One-repetition maximum

SPPB

Short physical performance battery

SFT

Senior fitness test

8FUAG

8-Foot up-and-go

FAC

Functional ambulation category

GUG

Get up and go

SF-36

Short form 36

SarQoL

Sarcopenia quality of life

SF-12

Short form 12

ICFSR

International clinical practice guidelines for sarcopenia

Author contributions

MCFL: Writing – original draft, Visualization, Formal analysis, Data curation, Methodology, Investigation. RGO: Writing – review & editing, Validation, Supervision, Project administration, Methodology. LIMA: Writing – original draft, Visualization, Formal analysis, Data curation, Methodology, Investigation. LCO: Writing – review & editing, Validation, Supervision, Project administration, Methodology, Conceptualization.

Data availability

All generated data is available as supplementary material.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.