Association between visceral adipose tissue and total testosterone among the United States male adults: a cross-sectional study.

Abstract

Visceral adipose tissue (VAT) is regarded as an important risk factor for obesity-related diseases. The results of the association between VAT and total testosterone (TT) are controversial and whether this association is nonlinear is still unknown. 3971 male participants who were aged 20-59 years from the National Health and Nutrition Examination Surveys 2011-2016 were included. VAT area was measured by dual-energy x-ray absorptiometry. TT in serum was assessed utilizing the isotope dilution liquid chromatography-tandem mass spectrometry technique. Linear regression models assessed the associations between VAT area and TT. A restricted cubic spline model was employed to investigate nonlinear relationships. A two-piecewise linear regression model was applied to determine the threshold effect. Subgroup analyses were conducted. The weighted methods were utilized in all analyses. VAT area was inversely associated with TT in the crude and adjusted models. In the fully adjusted model, VAT area was associated with TT (β = -0.59, 95% confidence interval [CI] = -0.74, -0.43) and compared to the first tertile of VAT area, the second and the third tertile had a lower TT level, the β and 95% CI = -65.49 (-83.72, -47.25) and -97.57 (-121.86, -73.27) respectively. We found these inverse associations were nonlinear. The cutoff point of the VAT area was 126 cm2. When the VAT area was <126 cm2, VAT area was significantly associated with a lower TT level (β = -1.55, 95% CI = -1.93 to -1.17, p < 0.0001). However, when the VAT area was ≥126 cm2, this association was less apparent (β = -0.26, 95% CI = -0.52 to 0.01, p = 0.06). No significant interactions among different ages (<50 or ≥50 years), marital, and physical activity status were found. These findings underscore the potential for VAT area as a modifiable indicator for improving testosterone deficiency.