Aging muscle still rebuilds its power plants when you train

As muscle ages, its mitochondria falter: energy production becomes less efficient, the regulation of oxidative stress shifts, the balance between mitochondrial fusion and fission is disturbed, and the turnover that clears out damaged mitochondria slows down. This cellular decline is one of the suspected drivers of the loss of strength and stamina in older age. Exercise is the most promising lever available, but the evidence on whether training can genuinely remodel aged mitochondria has been scattered across human and animal studies that each measure different things. This review set out to pull that evidence into a single picture.

The authors searched PubMed, EMBASE and Web of Science up to June 26, 2026, following PRISMA 2020 guidance, for studies comparing aged exercise groups with sedentary counterparts. Twenty-eight studies qualified, spanning both human and rodent work. Outcomes covered oxidative stress, mitochondrial dynamics, mitophagy, mitochondrial biogenesis and function, skeletal muscle morphology and physical performance. Where the data allowed, results were pooled statistically; where they did not, the authors mapped the direction of findings across studies to look for consistent patterns.

The pooled analysis showed a significant improvement in VO2max or VO2peak after exercise training, with the included studies in close agreement. By contrast, pooled changes in citrate synthase, a common marker of mitochondrial content, and in BMI were not significant. The directional synthesis found mitochondrial biogenesis and function consistently responsive across species, particularly OXPHOS proteins, mitochondrial volume density, ATPmax and SIRT1/SIRT3 signalling. Human and rodent studies assessed only partly overlapping markers of mitophagy and mitochondrial dynamics, so differences between species could not be separated from differences in what was measured.

The result supports the broad view that training counters some of the mitochondrial decline of aging, and it extends that view by showing which cellular pathways respond most consistently. What it cannot do is crown a best training modality: patterns varied by protocol and study design, and the comparisons were indirect. The authors are explicit that human and rodent findings should be read as complementary evidence rather than equivalent estimates. Which protocols and durations remodel aged mitochondria most effectively remains an open question.

For anyone watching their fitness decline with age, this is a reason to think the underlying machinery is still trainable: aerobic capacity improved measurably, and the cellular systems that build and maintain mitochondria responded to exercise. Because no style of training proved superior here, the finding supports being consistently active in general rather than any specific routine. The absence of a BMI change is also a useful reminder that the benefits of training in older age are not captured by the bathroom scale.