Abstract
BACKGROUND: Thyroid stimulating hormone (TSH) has been proven to be closely associated with metabolic abnormalities, including obesity. The objective of this investigation was to scrutinize the intricate association between TSH concentration and obesity, within the adult population. The study focused on a comprehensive examination of the relationship, delving into specific adiposity parameters such as total percent fat (TPF), android percent fat (APF), and gynoid percent fat (GPF). METHODS: This study included 809 participants aged 20 and above with normal TSH values from the National Health and Nutrition Examination Survey (NHANES) 2007-2012. Multivariable linear regression models examined the associations of TSH concentration with TPF, APF, and GPF. Subgroup analyses, stratified by sex, were performed using multivariable linear regression. Fitted smoothing curves and generalized additive models addressed non-linear relationships between TSH and TPF, APF, and GPF. RESULTS: In fully adjusted models, a significant positive association was observed between TPF and TSH (β = 0.01, 95% CI: 0.00-0.02, p<0.05), while no such association was evident in APF and GPF. Upon sex stratification, females exhibited significant positive correlations between TSH and TPF, APF, and GPF (all p < 0.001), contrasting with males where no such correlations were found. Notably, a non-linear association was identified in males, specifically a U-shaped curve (inflection point: 32.6%) for TSH and APF. CONCLUSION: The study unveiled a statistically significant positive association between TSH and TPF in adults. Upon sex stratification, similar statistically significant relationships were observed between TSH and adiposity (TPF, APF, GPF) in females, while males exhibited a U-shaped non-linear relationship between TSH and APF.