An inverse relationship between serum total testosterone levels and NAFLD was shown in all models (p value: 0.0004 in model 1, 0.0006 in model 2, and 0.004 in model 3) (Table 2).
An inverse association between serum total testosterone levels and NAFLD was also shown in all models (p value: 0.0036 in model 1, 0.0044 in model 2, and 0.0292 in model 3).
The association between serum total testosterone levels and NAFLD was examined by employing multiple logistic regression analysis with potential confounding variables such as age, smoking, diabetes, exercise, BMI, TGs, HDL-C, HOMA-IR, hs-CRP, and VAT controlled.
We fitted conditional logistic regression models to examine the associations between SHBG and total testosterone levels with diabetes by sex.
These findings led to the assumption that the association between low serum total testosterone levels and NAFLD might be mediated by the effect of VAT and insulin resistance.
a Multiple logistic regression was used to analyze the associations between NAFLD and serum total testosterone levels.
There were no differences between groups at baseline, and total testosterone levels increased in all treatment arms compared with pre-treatment levels.
For a sensitivity analysis of our main finding, we examined the relationship between daily BPA excretion and total testosterone levels in men, excluding outlier BPA values above 25 μg/day (n = 7 removed, ranging from 25.29 41.12 μg/day) (see Figure 1 for unadjusted model).
Results: Total testosterone levels were measured in 19 men.
In group M, there were no significant statistical differences between intermediate-severe sexual dysfunction groups and mild-intermediate sexual dysfunction groups in terms of total testosterone levels.
