Statins, Grip Strength, and the Genetics That Don’t Predict Safety

Your genetics may help predict how you process a statin, but they did not identify a subgroup protected from declining grip strength or appendicular lean mass in a large UK Biobank analysis. That finding deserves attention. It does not prove that statins cause muscle loss.

Gentreau et al. (2025) analyzed 297,977 UK Biobank participants cross-sectionally and followed 35,557 of them across two assessment visits. Statin use was associated with statistically significant declines over time in bilateral grip-strength sum and appendicular lean mass, a measure derived from bioelectrical impedance analysis. The study adjusted for body mass index, activity, diet, smoking, alcohol, depression, hypertension, diabetes, coronary heart disease, COPD, and renal failure.

Those are objective endpoints, not a questionnaire asking whether someone felt achy after a prescription refill. But the study remains observational. People receive statins because they have, or are at high risk for, cardiometabolic disease. Those same conditions can accelerate muscle loss. A very large cohort can narrow random error; it cannot make that problem disappear.

The genetic "escape hatch" did not appear

The study's most unusual feature was its pharmacogenomic analysis. The researchers built a polygenic score using variants in genes involved in statin pharmacokinetics and pharmacodynamics, including SLCO1B1, CYP3A4, CYP2C9, ABCG2, APOE, and PCSK9. The score was intended to capture differences in LDL-lowering response, hepatotoxicity risk, and susceptibility to myopathy.

It did not meaningfully modify the longitudinal association between statin use and either muscle endpoint. In the cross-sectional analysis, the interaction results were borderline rather than conventionally statistically significant: p=0.058 for grip strength and p=0.068 for appendicular lean mass. In the longitudinal analysis, the interaction was null.

That result does not mean genes play no role in statin muscle symptoms. It means this particular score did not identify a genetically protected group for the functional outcomes measured here. The distinction matters because the score was built from genetic associations with statin efficacy and hepatotoxicity, not from studies designed specifically to predict statin-related weakness, lean-mass loss, or myopathy. A null result can reflect the wrong genetic tool as easily as the absence of a genetic effect.

What the UK Biobank study measured—and what it did not

Grip strength came from dynamometry, with strength summed across both hands. Appendicular lean mass was estimated by bioelectrical impedance analysis; the authors report validation against dual-energy X-ray absorptiometry. These are useful measures of muscle function and body composition, although appendicular lean mass is not the same as a clinical diagnosis of frailty, disability, or statin myopathy.

The longitudinal findings were statistically significant after adjustment, but the available report does not provide the absolute change estimates or confidence intervals needed to judge clinical size here. Statistical significance in a cohort this large does not tell a reader how much strength or lean mass changed in practical terms.

The exposure was baseline statin use, not randomized treatment assignment. That distinction is the center of the causal question. A person prescribed a statin is more likely to have diabetes, hypertension, coronary heart disease, or a related risk profile. Each can affect muscle through inflammation, vascular disease, reduced activity, insulin resistance, hospitalization, and aging-related illness. Adjusting for yes-or-no comorbidity variables helps, but it cannot fully capture severity, duration, treatment history, or deterioration during follow-up.

The study identifies a population-level association. It cannot separate a drug effect from the muscle consequences of the diseases that commonly lead to a statin prescription.

The follow-up sample also included only 35,557 people, about 12% of the original cross-sectional cohort. That 88% reduction raises the possibility of selection or survivorship bias: people who return for repeat assessment may be healthier or differ in adherence and disease progression from those who do not. The direction of any resulting bias cannot be assumed.

A plausible mechanism is not proof in living humans

Statin-associated muscle symptoms are real. A 2025 systematic review and meta-analysis of coenzyme Q10 supplementation describes myalgia, weakness, and cramping in roughly 5% to 20% of statin users. One leading explanation starts with the mevalonate pathway: inhibiting HMG-CoA reductase may reduce coenzyme Q10 production, impair mitochondrial energy production, increase reactive oxygen species, and injure muscle fibers.

This is biologically plausible, but plausibility is not a biopsy result. The review found mixed evidence that coenzyme Q10 supplementation improves statin-associated muscle symptoms. It does not establish that coenzyme Q10 prevents the objective grip-strength or lean-mass changes seen in UK Biobank.

A separate 2025 hiPSC-derived myocyte study by Zhao et al. found that statin myotoxicity in cultured human-derived muscle cells involved impaired proteostasis and dysregulation of the ubiquitin-proteasome pathway. Restoring proteostatic balance reduced toxicity in that model. Cell culture can identify hypotheses worth testing; it cannot show that the same pathway drives long-term muscle change in people taking standard clinical doses.

Human evidence has not yet closed that gap. The available literature here includes no randomized trial with grip strength and appendicular lean mass as prespecified statin endpoints, no Mendelian-randomization study using HMGCR variants for these muscle phenotypes, and no direct human muscle-biopsy, in-vivo mitochondrial-function, or coenzyme Q10 data that settle the mechanism.

Muscle protection does not need to wait for a causal verdict

Regardless of why a statin user loses muscle—or whether they lose any at all—resistance training has stronger evidence for improving muscle outcomes than supplements have for preventing a hypothetical statin effect. A 2024 meta-analysis of 12 randomized trials in secondary sarcopenia found improvements in muscle strength, muscle mass, and physical function with resistance training.

In adults over 70 with sarcopenia, a 2024 meta-analysis of 13 trials found that more than 10 weeks of resistance training improved handgrip strength by 1.67 kg (p=0.02), isometric strength (standardized mean difference 0.53; p=0.02), and skeletal muscle mass index by 1.67 kg/m² (p=0.0002). Those participants were not a statin-specific population, so these results do not prove that training offsets statin-related changes. They do show that muscle strength and mass remain modifiable in older adults with substantial comorbidity burdens.

Nutrition can support training, but the evidence needs more restraint than supplement marketing usually permits. A 2025 meta-analysis of leucine in older adults with sarcopenia included 566 participants receiving leucine and 567 receiving placebo. A 2023 review of whey protein, leucine, and vitamin D found improved appendicular muscle mass, but no significant difference in grip strength or Short Physical Performance Battery score when exercise co-intervention was not considered. Neither analysis tested statin-associated functional decline directly.

The practical verdict

The UK Biobank study is a credible warning signal: statin use tracked with worsening objective muscle measures over time, and the tested pharmacogenomic score offered no evidence of protection. It is not a verdict that statins destroy muscle.

If you take a statin and develop new muscle pain, weakness, or cramping, discuss it with the prescribing clinician rather than stopping treatment on your own. Cardiovascular risk reduction can be lifesaving, and statin choice, dose, timing, alternative agents, and other causes of symptoms all require clinical context. For muscle health, progressive resistance training has direct randomized-trial support; coenzyme Q10 remains a biologically reasonable but unproven answer for this specific problem.

A randomized trial measuring grip strength and appendicular lean mass in statin users, or a well-designed Mendelian-randomization study targeting these outcomes, would materially change the evidence. Until then, genetics does not offer reassurance here, and the association does not establish blame.