We sought to assess the associations of testosterones and sex hormone–binding globulin (SHBG) with metabolic syndrome and insulin resistance in men.
We defined metabolic syndrome according to the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Among men aged ≥20 years who participated in the Third National Health and Nutrition Examination Survey ( n = 1,226), the Cox proportional hazards model was used to estimate the prevalence ratio and 95% CI of metabolic syndrome according to circulating concentrations of testosterones and SHBG.
After adjustment for age, race/ethnicity, smoking status, alcohol intake, physical activity level, LDL cholesterol, C-reactive protein, and insulin resistance, men in the first quartile (lowest) (prevalence ratio 2.16 [95% CI 1.53–3.06]) and second quartile of total testosterone (2.51 [1.86–3.37]) were more likely to have metabolic syndrome than men in the fourth quartile (highest, referent group) ( P < 0.001 for linear trend). Similarly, men in the first quartile of SHBG (2.17 [1.32–3.56]) were more likely to have metabolic syndrome than men in the fourth quartile ( P = 0.02 for linear trend). No significant associations of calculated free testosterone ( P = 0.31 for linear trend) and bioavailable testosterone ( P = 0.11 for linear trend) with metabolic syndrome were detected after adjustment for all possible confounders.