PULMONARY DISEASES Reduced pulmonary function is associated with lower levels of endogenous total and free testosterone. The Tromsø Study Johan Svartberg 1 , Henrik Schirmer 1,2 , Astri Medbø 2 , Hasse Melbye 2 & Ulf Aasebø 1,3 1 Department of Medicine, University Hospital of North Norway, 9038 Tromsø, Norway; 2 Institute of Community Medicine, University of Tromsø, 9037 Tromsø, Norway; 3 Institute of Clinical Medicine, University of Tromsø, 9037 Tromsø, Norway Accepted in revised form 12 December 2006 Abstract. Men with chronic obstructive pulmonary disease have reduced endogenous testosterone levels. Little is known, however, about the relationship be- tween pulmonary function and endogenous testos- terone levels in a general population. In the present study we have examined the cross-sectional associa- tions between sex hormones measured by immuno- assay and pulmonary function assessed with spirometry and oxygen saturation in 2,197 men par- ticipating in the fifth Tromsø study. The data were analyzed by univariate correlations, multiple linear regression analyses and analyses of variance and covariance. Total and free testosterone were posi- tively and independently associated with forced vitality capacity, FVC (% of predicted) (P = 0.001 and P = 0.006, respectively) and forced expiratory volume in 1 second, FEV 1 (% predicted) (P = 0.033 and P = 0.002, respectively), and men with severe pulmonary obstruction (FEV 1 % of predicted < 50) had lower free testosterone levels (P = 0.005). In this cross-sectional data from Tromsø, a reduction in pulmonary function was associated with lower levels of total and free testosterone. We suggest that the reduction of total and free testosterone could be due to an alteration of the hypothalamic-pituitary response. Key words: Cross-sectional, Smoking, Spirometry, Testosterone, Waist circumference Abbreviations: BMI = body mass index; COPD = chronic obstructive pulmonary disease; DHEAS = dehydroepiandrosterone sulphate; FEV 1 = forced expiratory volume in 1 sec; FSH = follicle stimulating hormone; FVC = forced vital capacity; LH = luteinizing hormone; SHBG = sex hormone-binding globulin; WC = waist circumference Introduction It has previously been reported that men with chronic obstructive pulmonary disease (COPD) have reduced endogenous testosterone levels [1–4], and that continuous oxygen treatment can nor- malize testosterone levels [2]. Anthropometrical measurements such as weight are strongly associ- ated with testosterone levels [5] and weight reduc- tion is a well-known feature in COPD patients [6]. Furthermore, testosterone supplementation has been found to improve lean body mass and body composition in men with COPD [7–9]. It is gener- ally agreed that smoking is strongly associated with reduction in pulmonary function, and in several cross-sectional studies smoking has been associated with higher both total and free testosterone levels [10–12]. Thus, there are several strong and inter- esting relationships between testosterone levels and pulmonary functions, but to our knowledge no large population-based study has investigated these associations. Therefore, in the present study we have examined the cross-sectional associations be- tween sex hormones and pulmonary function as- sessed with spirometry and oxygen saturation in 2,197 men participating in the fifth Tromsø study. Subjects and methods All men and women older than 29 years, living in the municipality of Tromsø and who participated in the second phase of the fourth Tromsø study [10] or be- came 30, 40, 45, 60 or 75 years old during 2001, were invited to participate. This fifth survey also consisted of two visits held a few weeks apart. All subjects that participated the second phase of the fourth Tromsø study in 1994–1995 were invited to the second visit in the fifth survey for more extensive examinations, which, for the first time in the Tromsø study, also included assessment of pulmonary function. A total of 5,939 subjects (2,447 men) attended the second visit and 5,110 subjects participated in the pulmonary investigation. European Journal of Epidemiology (2007) 22:107–112 Ó Springer 2006 DOI 10.1007/s10654-006-9095-9