Counts describe this library, not the strength of evidence.
How to read this evidence
Strength, muscle size and function are related but distinct outcomes. Trials and reviews support progressive resistance exercise, while longevity cohorts cannot establish exact safe-dose limits. More complicated prescriptions do not consistently outperform achievable basics.
Among 191 pre-frail adults aged 65 or older, the highest tertile of fat-free mass index (FFMI) went with better physical function, higher cognitive scores and lower odds of sarcopenia after adjustment, while higher BMI was associated with lower odds of sarcopenia. The abstract gives no effect sizes, and the study is cross-sectional and small.
In a meta-analysis of 16 observational studies, people with diabetes had higher odds of sarcopenia than people without it (combined OR 2.09, 95% CI 1.62 to 2.70, from 7 studies). Because the studies are observational and the abstract repeats the same OR for the complications analysis, the exact size and its causal meaning are uncertain.
In 26 studies of 1191 older people with sarcopenia, resistance training (SMD 1.36) and mixed training (SMD 0.62) improved knee extension strength versus controls, while whole body vibration did not (SMD 0.65, 95% CI -0.02 to 1.31). Heterogeneity was high (I² 56% to 84% for the significant results), so the size of the benefit is uncertain.
In a meta-analysis of 14 trials with 561 older adults with sarcopenia, resistance training improved handgrip strength (SMD 0.81), knee extension strength (SMD 1.26) and gait speed (SMD 1.28) and reduced body fat mass (SMD -0.53). Muscle mass measures did not clearly change, and heterogeneity was high (I2 up to 89%) for several of the strength and performance outcomes.
European review of aging and physical activity : official journal of the European Group for Research into Elderly and Physical Activity
Across 43 studies, adding supervised aerobic training to an identical strength program made no clear difference to maximal strength (SMD -0.06) or muscle hypertrophy (SMD -0.01) but lowered explosive strength gains (SMD -0.28, 95% CI -0.48 to -0.08, p = 0.007). The explosive-strength effect was clearer when both were done in the same session, and the abstract reports no information on study quality.
Across 28 studies of 16,800 people with type 2 diabetes, the pooled prevalence of sarcopenia (loss of muscle mass and function) was 18% (95% CI 0.15-0.22), with older age the clearest risk factor (OR 4.73). Heterogeneity was extreme (I2 = 97.4%) and the abstract repeats one set of estimates for several factors, so most risk-factor numbers are unreliable.
In an open-label pilot of eight older adults, 24 weeks of GlyNAC (glycine plus N-acetylcysteine) corrected glutathione deficiency and improved markers of oxidative stress, inflammation, strength, gait speed and cognition, with benefits declining after 12 weeks off the supplement. With eight people, no placebo group and no numbers in the abstract, this is a hypothesis for larger trials rather than evidence of benefit.
A systematic review of 44 studies on commercially available resistance-training monitors found that most did not use a gold-standard criterion measure, so device errors may be under- or over-reported. Linear transducers appeared more accurate and reliable than other device types, but the abstract gives no per-device numbers.
In 887 community-dwelling adults aged 50 or older, sit-to-stand scores tracked gait speed, endurance and balance far more closely than grip strength, and did not track muscle mass. Confirmed sarcopenia was less common when defined by sit-to-stand (4.6% and 4.1%) than by grip strength (7.1%), a cross-sectional comparison that cannot say which definition predicts outcomes better.
European review of aging and physical activity : official journal of the European Group for Research into Elderly and Physical Activity
Among 1851 Japanese adults aged 65+, sarcopenia affected 11.5% of men and 16.7% of women, and was linked to higher mortality (HR 2.0 in men, 2.3 in women) over 5.8 years. Low muscle mass alone, without weak grip or slow walking, was not linked to higher risk, but the confidence intervals are wide.
Using data from 11 448 adults in Italy, the authors built age- and sex-specific percentile curves for handgrip strength and the five-repetition chair-stand test, showing a rise to a peak at 18-24 years, preservation at 25-44 and decline from 45 onwards. Participants were volunteers recruited at exhibitions, malls and health campaigns, so they may not mirror the whole population.
Among 1732 older adults with type 2 diabetes, 10.37% had sarcopenia, and metformin users had lower odds than people taking no medication (OR 0.510 for metformin alone, 0.398 combined with other drugs). This is a cross-sectional snapshot, so it cannot show that metformin protects muscle.
In a trial of 140 hospitalized older adults with sarcopenia, a whey protein, leucine and vitamin D formula raised gait speed by 0.061 m/s per month versus -0.001 with placebo (difference 0.063 m/s/month, 95% CI 0.043 to 0.082, P < 0.001). It was a single-site trial, and results were analyzed in 127 patients who were reassessed at discharge.
A meta-analysis of six studies with 223 older adults with sarcopenia found that whole-body vibration (SMD 0.69, 95% CI 0.28 to 1.11, P = 0.001) and local vibration (SMD 3.78, 95% CI 2.29 to 5.28, P < 0.001) increased muscle strength. The authors say the small number of studies calls for caution.
European review of aging and physical activity : official journal of the European Group for Research into Elderly and Physical Activity
This umbrella review of 30 meta-analyses (54 outcomes) found sarcopenia, loss of muscle mass and function, linked to worse cancer prognosis, more postoperative complications and, in older adults, more falls, fractures and all-cause mortality. Most outcomes were rated low or very low quality evidence, and observational data cannot show cause.
Pooling 8 studies, older adults with sarcopenia had a summarized odds ratio (OR) of 1.29 (95% CI 1.00-1.67) for hypertension, while handgrip strength showed no association (OR 0.99, 95% CI 0.80-1.23). The pooled estimate sits right at the edge of statistical significance, and the studies were mostly cross-sectional-type associations, so they cannot show which came first.
In 294 community-dwelling adults aged 65 and over, the Short Physical Performance Battery (SPPB) had moderate accuracy for sarcopenia (AUC 0.644 to 0.770), and a cut-point of 8 points or less caught 82-100% of severe cases but had only 36-41% specificity. Sarcopenia was rare (4 to 16%), so many people flagged would not have it.
Among 4500 adults aged 50 or older in western China, 19.31% had sarcopenia, and sarcopenia was linked to mild (odds ratio 1.41) and moderate/severe (OR 3.05) cognitive impairment. It is a single snapshot, so it cannot show which came first.