Exercise for Sarcopenia: Which Training Helps Lung Function?

For older adults with sarcopenia, the best current bet for improving measured lung function is combined aerobic and resistance training. In a 2026 meta-analysis of 11 randomized trials involving 655 older adults with sarcopenia, combined programs improved forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and maximal inspiratory pressure (MIP). Resistance-only training improved one measure, while aerobic-only training did not reach statistical significance on the spirometric outcomes pooled in that review.

That ranking is useful, but it is not a license to dismiss walking. The aerobic-only estimate for MIP was almost identical to the combined-training estimate: +9.40 versus +9.42 cmH₂O. The difference was statistical precision, not an obvious difference in magnitude. A separate randomized trial also found that a walking-based home program improved cardiorespiratory performance in sarcopenic older adults. The evidence favors combination training when pulmonary function is the stated target; it does not show that aerobic exercise is pointless.

Why muscle loss can affect breathing

Sarcopenia reduces skeletal muscle mass and strength with age. The respiratory-sarcopenia concept extends that problem to the muscles that support breathing, including the diaphragm and other respiratory muscles. In a 2021 cross-sectional study of community-dwelling older adults, sarcopenia was independently associated with respiratory muscle weakness after adjustment for covariates. That finding shows an association, not proof that limb-muscle loss causes respiratory weakness.

Still, the pattern has biological plausibility. A 2024 diagnostic study found that diaphragmatic mobility and maximal respiratory pressures declined alongside sarcopenia. A separate randomized trial of inspiratory muscle training in elderly women found measurable increases in diaphragm thickness. Together, these studies support the idea that respiratory muscles can lose capacity with age and can respond to training. They do not prove that every improvement in spirometry comes from diaphragm hypertrophy.

The evidence-ranked exercise options

The most direct comparison comes from Pan et al.'s 2026 systematic review and meta-analysis. It pooled 11 randomized controlled trials with 655 sarcopenic older adults and compared aerobic, resistance, and combined aerobic-plus-resistance programs on pulmonary measures.

Combined training produced statistically significant mean improvements in FVC of 0.30 L (95% CI, 0.09 to 0.52; P=0.005), FEV1 of 0.28 L (95% CI, 0.14 to 0.44; P=0.0002), and MIP of 9.42 cmH₂O (95% CI, 0.85 to 17.99; P=0.03). Against aerobic-only control programs, combined training also showed additional gains in FVC of 0.14 L (P=0.0003), FEV1 of 0.28 L (P<0.00001), and FEV1/FVC of 8.52 (P<0.00001).

Resistance-only training reached statistical significance for FEV1/FVC, with a mean difference of 2.74 (95% CI, 0.36 to 5.13; P=0.02). It did not reach significance for FVC, FEV1, or MIP. Aerobic-only training did not reach significance for FVC, FEV1, FEV1/FVC, or MIP in the pooled analysis.

For pulmonary measurements, the provisional order is combined training first, resistance training next, and aerobic-only training last. The word "provisional" does real work here.

Why the aerobic result deserves a closer look

The aerobic-only MIP result was +9.40 cmH₂O, with a 95% CI from −1.09 to +19.89 and P=0.08. Because the confidence interval crossed zero and the P value did not meet the conventional threshold, the meta-analysis classified it as non-significant. But its point estimate was nearly the same as the combined-training MIP gain of +9.42 cmH₂O.

This does not prove that aerobic and combined training work equally well. It does mean the MIP finding cannot support a hard claim that aerobic training fails biologically. Smaller samples and wider uncertainty can turn a similar estimated effect into a non-significant result. The pooled literature is also small enough that subgroup conclusions remain unstable.

More importantly, a 2023 randomized trial of a walking-based home program found improved cardiorespiratory performance and physical activity in sarcopenic older people. Its outcome was cardiorespiratory performance, not the spirometric endpoints pooled by Pan et al. Different outcome measures, program characteristics, and small-study variation could explain why it appears to conflict with the meta-analysis. The studies answer related questions, but not the identical question.

What may be clinically meaningful—and what remains unknown

The combined-training gains of 0.28 L in FEV1 and 0.30 L in FVC exceed benchmarks sometimes used in chronic obstructive pulmonary disease research: roughly 100–200 mL for FEV1 and 150 mL for FVC. The MIP gain of 9.42 cmH₂O also approaches a 10 cmH₂O threshold used in inspiratory-muscle-weakness research.

Those comparisons provide context, not validation. No minimal clinically important difference has been established for sarcopenic older adults without primary lung disease. A change in a breathing test is also a surrogate outcome; it does not automatically mean fewer hospitalizations, less breathlessness, or better survival. The available trials do not establish those patient-level outcomes.

Resistance probably matters, but the mechanism is not settled

A 2022 randomized trial in older women with sarcopenia found that peripheral resistance training improved respiratory parameters even though the program did not directly train the inspiratory muscles. That result suggests that training limb muscles may improve breathing through broader changes in muscle function, thoracic mechanics, or the effort required for daily activity. It cannot identify which pathway caused the change, and its all-female sample limits conclusions about older men.

Direct respiratory training may offer another route. The inspiratory-muscle-training trial in elderly women increased diaphragm thickness, but it did not establish that this approach outperforms whole-body combined exercise for spirometric outcomes. For now, the evidence supports including progressive resistance work when lung function is a priority, rather than treating it as optional decoration around a walking plan.

Can this be done at home?

Probably. A 2025 randomized trial of home-based virtual-reality exercise plus resistance training found the approach feasible and safe for community-dwelling older people with sarcopenia. It offers a practical delivery option, especially for people who cannot attend a gym or rehabilitation center. Pulmonary-specific outcome data from that study are not available here, so it cannot show whether virtual-reality delivery improves FVC, FEV1, or MIP more than conventional training.

Home walking also has randomized-trial support for cardiorespiratory performance. It remains a sensible option for people unable to perform resistance training, but it has less direct evidence for improving spirometric lung measures in this population.

What limits confidence in the ranking

The central meta-analysis includes only 655 participants across 11 trials. Its trials likely used different definitions of sarcopenia, which can enroll meaningfully different groups of older adults. Diagnostic uncertainty extends to respiratory sarcopenia itself: a 2025 comparative analysis found that maximal respiratory pressure and peak expiratory flow identify different probable respiratory-sarcopenia phenotypes.

The review also does not provide enough protocol detail to prescribe a minimum effective dose, a best resistance-to-aerobic ratio, or a specific program duration. It reports no long-term follow-up after training stops, no sex-stratified results, and no available funnel-plot or Egger-test data to assess publication bias. Its FEV1/FVC comparison between combined and aerobic training also varied substantially by control condition, so that estimate should not be treated as fixed.

The practical verdict

If an older adult with sarcopenia wants the exercise approach with the broadest randomized evidence for measured lung-function improvement, choose a program that combines aerobic activity with progressive resistance training. The pooled gains in FVC, FEV1, and MIP favor that choice.

Walking still has value, especially when resistance exercise is not feasible. It improved cardiorespiratory performance in a trial of sarcopenic older adults, and the pooled aerobic MIP estimate was too imprecise to call it a biological failure. What current research cannot provide is a precise prescription: the optimal intensity, frequency, duration, and long-term maintenance plan remain unknown. For an older person with lung disease, chest symptoms, major frailty, or a high fall risk, a clinician or physical therapist should tailor the program before the first hard session.