Abstract B118: Modifiable correlates of low testosterone in U.S. men

Abstract

Abstract Background: In some studies, men with low total, free, or SHBG-adjusted testosterone possibly had a higher risk of cancer death. Because testosterone replacement therapy for men with clinically low testosterone has an uncertain risk versus benefit ratio, identifying modifiable factors that influence testosterone level and that are public health targets is needed. Obesity (lower testosterone) and cigarette smoking (higher testosterone) are two behavioral factors that are recognized to influence testosterone levels; given the great magnitude of the adverse effects of cigarette smoking, it is not a prevention strategy. Thus, in this study we investigated obesity as well as dietary and lifestyle factors in association with low total and free testosterone. Methods: The Third National Health and Nutrition Examination Survey (NHANES III), conducted 1988–1994, was a cross-sectional study of the U.S. civilian non-institutionalized population. Subjects participated in an interview, underwent a physical examination, and provided a blood specimen. Serum was available for 1470 men who participated in the morning session of Phase I (1988–1991) and were ≥20 years old. Testosterone and sex hormone binding globulin (SHBG) were measured by competitive electrochemiluminescence immunoassay. Free testosterone was estimated by mass action. We excluded men whose testosterone level was ≤2nd percentile because these men may have had congenital hypogonadism. We defined low total and free testosterone as ≤ 3.0 ng/mL and ≤ 0.07 ng/mL, respectively. We estimated the associations of the modifiable factors with low total and free testosterone using logistic regression adjusting for age and race/ethnicity. Results: Measures of overall and central adiposity were positively associated with low total and free testosterone. After mutual adjustment, men with high waist circumference (ethnic specific cutpoints; vs. low) were 2.59 (95% CI: 1.12–6.00) times more likely to have low total testosterone and 2.65 (95% CI: 1.44–4.89) times more likely to have low free testosterone, whereas body mass index and percent body fat did not remain statistically significantly associated. Independent of waist circumference, percent of calories from protein, fruit and vegetable intake, alcoholic beverage consumption, and physical activity were not statistically significantly associated with low total and free testosterone. These results were similar in men who did not have comorbidities at time of blood draw and in non-smokers. Discussion: Cross-sectionally, high waist circumference was the most consistent correlate of low total and free testosterone. These findings suggest that avoiding or reducing central adiposity may be one way to prevent clinically low androgen levels. Given that men with low androgens may have a higher risk of cancer death, our results may have import for cancer prevention and/or survival. Citation Information: Cancer Prev Res 2010;3(1 Suppl):B118.