Abstract
OBJECTIVE: This study sought to explore the relationships between sex hormones and obesity-related markers, while utilizing Mendelian randomization (MR) to evaluate their causal links. METHODS: Data from 5179 participants of the 2013-2016 National Health and Nutrition Examination Survey (NHANES) were analyzed. Multivariate linear regression and generalized additive models were utilized to examine linear and non-linear associations between sex hormones and obesity-related parameters. Causal inference was explored via a two-sample MR analysis, with inverse-variance weighting (IVW) as the primary approach. RESULTS: Cross-sectional assessments showed that in males, testosterone levels, SHBG, and T/E2 ratio were negatively correlated with obesity, while estradiol displayed a non-linear, inverted U-shaped relationship. In females, estradiol and SHBG had negative associations with adiposity measures. MR analysis revealed that genetically increased Bio-T was associated with lower obesity risk in males, while SHBG showed a negative correlation with visceral and waist obesity. In females, Bio-T and FAI were associated with higher visceral adipose tissue volume and waist circumference, findings consistent with potential causal effects. CONCLUSIONS: Sex hormones show sex-specific associations with obesity. Testosterone is protective against obesity in males, whereas female hyperandrogenism elevates the risk of obesity, particularly central obesity. These results offer novel perspectives for obesity-related health interventions.