To assess whether testosterone mediates the association between body fat percentage (BFP) and diabetes mellitus (DM) risk.
Approach:
Data Collection: Clinical data from Shengjing Hospital (N = 4433) and NHANES (N = 4772) were analyzed for the BFP-DM association.
Statistical Analysis: Restricted cubic spline (RCS) modeling and Mendelian randomization (MR) were employed to evaluate causal relationships and mediation effects.
Key Findings:
BFP was associated with DM (OR = 2.684, 95% CI 1.981-3.638).
Testosterone may mediate the BFP-DM association, accounting for 6.24% (95% CI 2.84%-10.68%) of the total effect, significant in males (14.92%, 95% CI 6.49%-25.29%) but not in females.
Higher BFP was linked to elevated blood glucose (OR = 1.147, 95% CI 1.097-1.200) and T2DM risk (OR = 2.456, 95% CI 2.120-2.845), while lower testosterone was associated with increased DM risk (OR = 0.614, 95% CI 0.521-0.724).
Testosterone partially mediated the effect of BFP on blood glucose (8.63%, 95% CI 2.19%-15.10%) and T2DM risk (3.80%, 95% CI 0.62%-6.98%), with no significant mediation for T1DM.
Interpretation:
Limitations:
The study is cross-sectional, limiting causal inference.
Further prospective studies are needed to confirm findings.
