Highlight
In this large UK Biobank analysis, probable sarcopenia was associated with a 30% higher risk of any incident stroke, with similar increases for ischemic stroke and an even larger relative increase for hemorrhagic stroke.
Lower grip strength, whether assessed as absolute strength or relative to body size, and self-reported slow walking pace were independently associated with stroke risk after multivariable adjustment.
Among participants who developed stroke, probable or confirmed sarcopenia was associated with higher all-cause mortality, suggesting that muscle health may matter both before and after cerebrovascular events.
A 2-sample Mendelian randomization analysis supported a potentially causal association between faster walking pace and lower risk of any stroke and ischemic stroke.
Background
Stroke remains a leading cause of death and long-term disability worldwide. Although traditional risk factors such as hypertension, diabetes, smoking, atrial fibrillation, and dyslipidemia remain central to prevention strategies, interest has increasingly expanded toward functional and body-composition markers that may capture biological aging, frailty, and multisystem vulnerability. Sarcopenia, broadly defined as the age-related loss of muscle strength and muscle mass, has emerged as one such candidate.
The clinical relevance of sarcopenia extends well beyond falls and mobility limitation. Low muscle strength has been linked to cardiovascular events, all-cause mortality, hospitalization, and poorer recovery after acute illness. Walking pace, meanwhile, has long served as a simple but powerful surrogate for cardiorespiratory fitness, neuromuscular integrity, and overall physiologic reserve. Yet whether sarcopenia itself, or its component traits such as reduced grip strength and slower walking pace, independently predicts future stroke has remained less clear, particularly across stroke subtypes.
Tang and colleagues addressed this question using the UK Biobank, a very large prospective population cohort with linked health records. Their analysis is notable for combining conventional epidemiologic modeling with Mendelian randomization, thereby attempting to move beyond association toward causal inference, at least for selected traits.
Study Design
Population and cohort framework
This prospective cohort study included 482,699 UK Biobank participants. The mean age was 56.4 years with a standard deviation of 8.1 years, and 45.37% were men, corresponding to 219,015 male participants. UK Biobank recruited middle-aged and older adults from the general UK population, with extensive baseline phenotyping and subsequent linkage to hospital, death, and other health records.
Exposure assessment
Sarcopenia status was defined according to the EWGSOP2 criteria, the 2019 framework from the European Working Group on Sarcopenia in Older People 2. Under EWGSOP2, low muscle strength is considered the primary indicator of probable sarcopenia, while low muscle strength plus low muscle quantity or quality defines confirmed sarcopenia. In this cohort, probable sarcopenia was present in 4.7% of participants and confirmed sarcopenia in 0.4%.
The investigators also examined individual muscle-function markers. Grip strength was analyzed both as absolute grip strength and relative grip strength. Walking pace was categorized using baseline assessment data, with brisk walking pace used as the reference group and slow pace compared against it.
Outcomes
The primary outcomes were incident any stroke, ischemic stroke, and hemorrhagic stroke, identified through linked health records. The study also examined all-cause mortality after stroke among participants who subsequently experienced a stroke event.
Statistical methods
Associations between sarcopenia-related measures and incident stroke were estimated using multivariable-adjusted Cox proportional hazards models. Poststroke survival was evaluated using Kaplan-Meier methods and log-rank testing. To explore causality, the investigators also performed 2-sample Mendelian randomization using genetic instruments derived from public genome-wide association studies, focusing particularly on walking pace and stroke outcomes.
Key Findings
Sarcopenia status and incident stroke
Compared with participants without sarcopenia, those with probable sarcopenia had a significantly higher risk of incident stroke across all major categories examined. The adjusted hazard ratio for any stroke was 1.30, with a 95% confidence interval of 1.21 to 1.39. For ischemic stroke, the adjusted hazard ratio was 1.31, with a 95% confidence interval of 1.22 to 1.42. For hemorrhagic stroke, the adjusted hazard ratio was 1.41, with a 95% confidence interval of 1.20 to 1.67.
These effect sizes are clinically meaningful. A roughly 30% increase in relative hazard for any stroke and ischemic stroke, and a roughly 40% increase for hemorrhagic stroke, suggests that probable sarcopenia captures a substantial burden of cerebrovascular vulnerability not fully explained by standard covariates included in the model. The study abstract does not provide the full adjusted covariate list, but the phrasing indicates a multivariable model intended to address confounding.
Grip strength as a continuous stroke risk marker
The study then examined grip strength more granularly. Lower absolute grip strength was associated with higher risk of any stroke, with an adjusted hazard ratio of 1.07 per 5-kg decrement and a 95% confidence interval of 1.06 to 1.09. Relative grip strength showed an even stronger association, with an adjusted hazard ratio of 1.36 per 1 kg divided by body mass index in kg per m2, and a 95% confidence interval of 1.28 to 1.44.
These findings matter because grip strength is easy to measure in routine care, inexpensive, and reproducible. Relative grip strength may better account for body size and adiposity than absolute grip strength alone, potentially explaining the larger observed effect estimate. Clinically, this means muscle weakness may carry prognostic significance even when overt sarcopenia is not formally diagnosed.
Walking pace and stroke risk
Self-reported slow walking pace was also associated with substantially increased stroke risk. Compared with brisk walking pace, slow pace was associated with an adjusted hazard ratio of 1.64 for any stroke, with a 95% confidence interval of 1.54 to 1.75.
Among the exposures examined, slow walking pace appears to have one of the largest effect sizes in the abstract. Walking pace is likely acting as an integrated physiologic marker, reflecting neuromuscular performance, cardiorespiratory fitness, frailty, subclinical vascular disease, and perhaps behavioral patterns such as lower physical activity. Because walking pace is simple to ask about, it may have practical value in risk stratification, though it should not be interpreted in isolation.
Poststroke mortality
Among participants who experienced a stroke, probable or confirmed sarcopenia was associated with increased all-cause mortality. The abstract does not provide hazard ratios or absolute survival differences for this poststroke analysis, but the direction of effect is consistent with prior literature linking frailty, low muscle reserve, and impaired rehabilitation potential to poorer outcomes after acute neurologic events.
From a clinical standpoint, this suggests that muscle health may influence more than primary prevention. It may also shape the trajectory after stroke, potentially affecting recovery capacity, susceptibility to complications, nutritional resilience, and tolerance of rehabilitation.
Mendelian randomization findings
The Mendelian randomization component adds an important layer of interpretation. Genetically predicted faster walking pace was associated with a lower risk of any stroke, with an odds ratio of 0.94 per standard deviation increase and a 95% confidence interval of 0.90 to 0.97. For ischemic stroke, the odds ratio was 0.95 per standard deviation increase, with a 95% confidence interval of 0.91 to 0.98.
These results support, though do not prove definitively, a potentially causal protective role for faster walking pace or for biological pathways captured by genetic determinants of walking pace. Notably, the Mendelian randomization findings were reported for walking pace rather than for sarcopenia status overall, which likely reflects stronger or more suitable genetic instruments for this trait.
Clinical Interpretation
The study’s central message is that muscle health and physical function are not peripheral geriatric concerns; they appear closely linked to cerebrovascular risk. This has at least three practical implications.
First, clinicians should consider low grip strength and slow walking pace as meaningful risk markers, especially in middle-aged and older adults who may not otherwise appear frail. These measures may help identify patients whose risk is underestimated by conventional cardiovascular profiling alone.
Second, the findings reinforce the value of functional assessment in preventive medicine. A dynamometer test and a simple question about usual walking pace are far easier to implement than many laboratory or imaging-based tools. Whether these measures improve prediction models beyond established risk factors requires formal testing, but they are appealing because of their low cost and broad feasibility.
Third, the poststroke mortality signal suggests that clinicians caring for stroke survivors may need to pay closer attention to pre-existing sarcopenia and ongoing muscle loss. In practice, that could mean earlier nutritional assessment, resistance exercise where feasible, swallow-safe protein optimization, mobilization, and rehabilitation planning tailored to limited physiologic reserve.
Biological Plausibility
The observed associations are biologically plausible. Sarcopenia is linked to chronic inflammation, insulin resistance, oxidative stress, endothelial dysfunction, and reduced physical activity, all of which may promote vascular disease. Low muscle mass and strength may also reflect cumulative exposure to adverse cardiometabolic conditions.
Walking pace likely captures multiple overlapping domains. Slower pace may signal poorer aerobic capacity, impaired autonomic regulation, early neurologic dysfunction, subclinical atherosclerosis, or frailty-related inflammation. In this sense, walking pace may function as a systems-level phenotype rather than a pure musculoskeletal measure.
There may also be reverse-pathway considerations. Preclinical cerebrovascular disease could contribute to declining gait performance or grip strength before a formal stroke diagnosis. That possibility does not negate the predictive value of these traits, but it does complicate causal interpretation in standard observational analyses and partly explains the value of the Mendelian randomization component.
Strengths of the Study
The study has several major strengths. Its sample size is exceptionally large, allowing robust estimation of associations for overall stroke as well as ischemic and hemorrhagic subtypes. The prospective design reduces some forms of recall bias and clarifies temporality between baseline exposures and subsequent stroke outcomes.
The use of EWGSOP2 criteria strengthens clinical interpretability because this framework aligns with contemporary sarcopenia definitions used in geriatric and rehabilitation medicine. The examination of both syndrome-level classification and component traits such as grip strength and walking pace also helps clinicians understand where the signal is coming from.
Finally, the inclusion of Mendelian randomization is an important methodological asset. Although not definitive, it provides supportive evidence that the walking pace association may not be entirely explained by residual confounding.
Limitations and Cautions
Several limitations deserve careful attention. UK Biobank participants are known to be healthier and less socioeconomically deprived than the general population, which may affect external validity. Absolute risks may therefore differ in routine clinical populations, although relative associations often remain informative.
Second, sarcopenia assessment in large cohorts often relies on pragmatic surrogates rather than gold-standard muscle imaging or repeated longitudinal measures. The very low prevalence of confirmed sarcopenia in this cohort, only 0.4%, suggests that stricter criteria identified a small subgroup and may limit precision for some estimates not shown in the abstract.
Third, walking pace was self-reported, which introduces measurement error. Even so, self-reported walking pace has repeatedly shown prognostic value in epidemiologic studies, which may speak to its real-world usefulness despite imperfect precision.
Fourth, residual confounding remains possible. Low grip strength and slow pace are correlated with multimorbidity, physical inactivity, undernutrition, depression, and early neurologic disease. Multivariable adjustment can reduce but not eliminate these concerns.
Fifth, Mendelian randomization assumptions must be respected. Genetic instruments for walking pace may capture pathways related not only to locomotion but also to broader health status, body composition, or neurocognitive factors. Pleiotropy therefore remains a potential challenge.
Implications for Practice and Research
This study does not mean sarcopenia should immediately be added to formal stroke prevention guidelines as a causal, modifiable risk factor on par with hypertension or smoking. However, it does suggest that clinicians should pay more attention to muscle strength and mobility as part of holistic vascular risk assessment.
In primary care and neurology clinics, practical steps could include opportunistic grip strength testing in older adults, documentation of usual walking pace or gait speed, and closer review of vascular prevention in patients showing early functional decline. A patient with low grip strength and slow pace may warrant more careful blood pressure control, diabetes management, smoking cessation support, physical activity counseling, and evaluation for frailty or undernutrition.
For researchers, the next questions are interventional. Does improving muscle strength or gait performance reduce future stroke risk, or are these traits primarily markers rather than mediators? Randomized trials of resistance training, protein supplementation, multimodal exercise, and rehabilitation-oriented lifestyle interventions with vascular endpoints would be especially informative. It will also be important to test whether adding grip strength or walking pace improves stroke risk prediction beyond established clinical models.
Funding and ClinicalTrials.gov
The abstract provided does not report specific funding details. The study was conducted using UK Biobank data, and full funding disclosures should be confirmed in the published article.
A ClinicalTrials.gov registration number is not applicable to this observational UK Biobank cohort analysis.
Conclusion
Tang and colleagues provide strong epidemiologic evidence that probable sarcopenia, weaker grip strength, and slower walking pace are associated with higher risk of incident stroke in a very large community-based cohort. The association extends to both ischemic and hemorrhagic stroke, and sarcopenia also appears linked to worse survival after stroke. The Mendelian randomization analysis adds support for a potentially causal relationship between faster walking pace and lower stroke risk, particularly for ischemic stroke.
For clinicians, the study reinforces a simple but important idea: muscle function is a vascular health signal. A weak handshake and a slow walking pace may not merely reflect aging; they may mark elevated cerebrovascular risk and reduced resilience after stroke. The next challenge is to determine whether preserving or improving muscle health can translate into measurable stroke prevention benefits.
References
1. Tang LL, Huang YH, Jin YJ, Yang KC, Lin Q, Zhou Z, Gao F, Xiong B, Yuan C, Tong LS. Sarcopenia, Grip Strength, Walking Pace, and New-Onset Stroke Risk: A UK Biobank Study. Stroke. 2026-05-08. PMID: 42101654. Available at: https://pubmed.ncbi.nlm.nih.gov/42101654/
2. Cruz-Jentoft AJ, Bahat G, Bauer J, Boirie Y, Bruyère O, Cederholm T, Cooper C, Landi F, Rolland Y, Sayer AA, Schneider SM, Sieber CC, Topinkova E, Vandewoude M, Visser M, Zamboni M. Sarcopenia: revised European consensus on definition and diagnosis. Age Ageing. 2019;48(1):16-31. PMID: 30312372.
3. Sattar N, Mark PB, Jukema JW, et al. Perspective on the UK Biobank project and potential applications in cardiovascular research. Eur Heart J. 2019;40(14):1158-1162. PMID: 30844092.
4. O’Donnell MJ, Chin SL, Rangarajan S, et al. Global and regional effects of potentially modifiable risk factors associated with acute stroke in 32 countries: INTERSTROKE case-control study. Lancet. 2016;388(10046):761-775. PMID: 27431356.
5. Aroor AR, Jia G, Sowers JR. Cellular mechanisms underlying obesity-induced arterial stiffness. Am J Physiol Regul Integr Comp Physiol. 2018;314(3):R387-R398. PMID: 29237783.
6. Studenski S, Perera S, Patel K, et al. Gait speed and survival in older adults. JAMA. 2011;305(1):50-58. PMID: 21205966.