Muscle quality index combined with plasma metabolomics: enhancing risk stratification for post-stroke dementia

Abstract

Post-stroke dementia is a severe long-term stroke complication with heavy medical burdens, and early risk identification is critical. This study explored the association of muscle quality index (MQI) and plasma metabolomics with dementia among stroke survivors. This study included 7,047 UK Biobank participants with a history of stroke who were free of dementia at the baseline assessment. MQI was defined as dominant handgrip strength (HGS) divided by corresponding arm fat-free mass. Plasma metabolomics was analyzed via high-throughput nuclear magnetic resonance (NMR). Metabolomics risk scores (MRS-ACD, MRS-AD, and MRS-VaD) were constructed using metabolite features screening with LASSO regression and XGBoost algorithms. Statistical analyses included Cox proportional hazards models, time-dependent receiver operating characteristic (TD-ROC), restricted cubic spline (RCS) regression, interaction analyses, Mendelian randomization (MR), and several sensitivity analyses. Higher MQI was associated with lower risks of all-cause dementia (ACD), Alzheimer’s disease (AD) and vascular dementia (VaD) (all p for trend <0.05). Each 1-SD increment in MRS was associated with a higher risk of ACD (HR = 1.63, 95% CI: 1.36–1.96), AD (HR = 2.40, 95% CI: 1.74–3.31), and VaD (HR = 1.92, 95% CI: 1.46–2.54). Individuals with high MRS and lowest MQI had the highest ACD (HR = 3.29, 95% CI: 1.91–5.68), AD (HR = 5.83, 95% CI: 1.75–19.37), and VaD (HR = 4.57, 95% CI: 2.28–9.14) risk. MQI was linearly negatively associated with ACD ( p for non-linear = 0.240) and VaD ( p for non-linear = 0.665), and non-linearly associated with AD ( p for non-linear = 0.013). Two-sample MR using genetically proxied grip strength and lean mass provided suggestive genetic evidence for muscle-dementia relationships: higher genetically predicted lean mass reduced AD risk (OR = 0.92, 95% CI 0.86–0.99, p  = 0.027), while grip strength showed an independent protective effect on VaD after accounting for lean mass (OR = 0.16, 95% CI 0.04–0.64, p  = 0.009). Extensive sensitivity analyses confirmed the robustness of the main findings. Joint assessment of MQI and plasma metabolomics provides suggestive evidence for stratifying post-stroke dementia risk and may assist early risk screening in stroke survivors.