Association of serum testosterone with chronic obstructive pulmonary disease (COPD) in a nationally representative sample of White, Black, and Hispanic men

Background

The association between total testosterone (T) and chronic obstructive pulmonary disease (COPD), remains poorly understood. We aim to investigate this association and how it varies by smoking status, body fatness, and race/ethnicity in a nationally representative sample of American men.

Methods

Data included a full sample (NHANES 1988–1991, 1999–2004, 2011–2012) and subset sample (excluding 2011–2012, no estradiol and SHBG levels available) of 2748 and 906 men (≥20 years), respectively. COPD was measured by self-report or spirometry test. Total T (ng/mL) was measured among men who participated in a morning examination session. Weighted multivariable-adjusted logistic regression models were conducted.

Results

Low T was positively associated with self-reported COPD in the full sample (OR = 2.10, 95% CI = 1.18–3.74, P_trend = 0.010), and when stratified by current smokers and body fatness. When examined across race and ethnicity strata, this association persisted among White men (OR = 2.50, 95% CI = 1.30–4.79, P_trend = 0.002) but not among Hispanic or Black men. In the subset sample, low T was positively associated with self-reported COPD (OR = 1.42, 95% CI, 0.57,3.55, P_trend = 0.04), including among smokers and White men, but not body fatness. No significant associations were observed with COPD defined with spirometry plus self-report.

Conclusion

Low levels of T were associated with an increased prevalence of self-reported COPD in the full and subset samples. Similar associations were observed after stratifying by smoking status, body fatness, and race/ethnicity in the full sample and subset sample. Prospective studies are warranted to confirm these significant associations among understudied and underserved populations.