Mediation of fasting blood glucose between relative muscle strength and hypertension: insights from two cohorts

Abstract Background Hypertension and type 2 diabetes are major global health burdens and often coexist due to shared metabolic risk factors. Relative muscle strength (RMS), a composite measure of muscle function, shows an inverse association with hypertension. The mechanisms underlying this link remain unclear. Elevated fasting blood glucose (FBG) reflects impaired glucose metabolism and insulin resistance and is associated with both reduced muscle strength and higher blood pressure. This suggests that FBG may partly mediate the RMS–hypertension relationship. Yet, large-scale population studies have rarely tested this mediating pathway, leaving an important knowledge gap. Methods We analyzed data from two nationally representative cohorts: NHANES (2011–2014;adults aged ≥ 18 Years, mean age 48 years ) and CHARLS (2011–2012; adults aged ≥ 45 Years, mean age 58 years). Hypertension prevalence was nearly 30.0% in NHANES and 42.1% in CHARLS. RMS was calculated as grip strength divided by appendicular skeletal muscle mass (ASM). Hypertension was defined as systolic/diastolic blood pressure ≥ 140/90 mmHg or self-reported diagnosis. Logistic regression examined the RMS–hypertension associations, and causal mediation analysis quantified fasting glucose’s mediating role, and restricted cubic spline models were applied to explore potential non-linear relationship. Results In NHANES (n = 9,652; Hypertension prevalence 30.0%), RMS was inversely associated with hypertension across quartiles (Q2–Q4 vs. Q1: adjusted ORs (95% CIs) of 0.82, 0.78, and 0.72, respectively). Mediation analysis showed fasting glucose partially mediated this association, accounting for 13.1% of this association. In CHARLS (n = 12,946; Hypertension prevalence 42.1%), similar trends were observed (Q2–Q4 vs. Q1: adjusted ORs (95% CIs) 0.91, 0.76, and 0.66), But fasting glucose partially mediated this association, explaining only 2.0% of the association. Restricted cubic spline models revealed significant nonlinearity in CHARLS (P < 0.001) but not in NHANES (P = 0.921). Conclusion This study is the first to examine the partially mediating role of fasting blood glucose in the RMS–hypertension relationship across diverse populations. RMS was consistently associated with lower hypertension risk, with varying degrees of glucose mediation between cohorts. These findings support integrating RMS assessment into cardiovascular risk screening and highlight muscle strength as a potential target for non-pharmacological prevention. Given the cross-sectional nature of this study, longitudinal research is needed to clarify causal pathways and inform public health strategies.