Association of the triglyceride-glucose muscle-loss index with incident stroke risk: a large-scale prospective cohort study in the UK Biobank

Background Insulin resistance and skeletal muscle depletion are recognized as synergistic contributors to vascular pathogenesis, yet their combined prognostic value for cerebrovascular outcomes has not been established. This study investigated the longitudinal association between the triglyceride-glucose muscle-loss index (TyG-MLI), a novel composite biomarker integrating these two pathological dimensions, and the risk of incident stroke and its subtypes. Methods We analyzed 400,624 UK Biobank participants free of stroke at baseline. TyG-MLI was derived from fasting triglycerides, glucose, cystatin C, and creatinine. Primary endpoints were incident overall stroke, acute ischemic stroke (AIS), intracerebral hemorrhage (ICH), and subarachnoid hemorrhage (SAH). Multivariable cause-specific Cox regression was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Results Over a median follow-up of 13.64 years, 8,354 participants developed strokes (AIS: 6,321; ICH: 1,411; SAH: 844). After full adjustment, each 1-SD increment in TyG-MLI was independently associated with overall stroke (HR 1.15, 95% CI 1.12–1.17), AIS (HR 1.16, 95% CI 1.13–1.19), and ICH (HR 1.13, 95% CI 1.07–1.19), but not SAH (HR 1.07, 95% CI 0.99–1.16; p = 0.083). Adding TyG-MLI to the base clinical model modestly improved discrimination for overall stroke (Δ C-statistic = 0.0023), AIS (Δ C-statistic = 0.0026), and ICH (Δ C-statistic = 0.0020), but not SAH. Smoking status modified the association with overall stroke. Conclusions TyG-MLI was consistently associated with incident overall stroke and AIS, whereas its association with ICH should be interpreted cautiously and requires further validation. This integrated metabolic-muscular biomarker may provide modest additional information for cerebrovascular risk stratification pending external validation.