It is hypothesized that the association between physical activity or cardiovascular fitness and chronic disease risk might be mediated, in part, through sex hormones in men. Limited observational data exists on the potential association of either physical activity or cardiovascular fitness with androgens.
PURPOSE: To examine the cross-sectional association of serum androgen concentrations with self-reported physical activity and cardiorespiratory fitness in black and white young men.
METHODS: Data were collected from 391 black and 604 white male participants of the Coronary Artery Risk Development in (Young) Adults (CARDIA) Male Hormone Study who were aged 28 to 42 years. Cross-sectional associations of serum total testosterone (TT), bioavailable testosterone (BT) and sex hormone binding globulin (SHBG) levels with self-reported total physical activity score (activity), and in a subset of men (n=617) with cardiorespiratory fitness measured via duration on a treadmill test (fitness) were assessed. Analyses were jointly stratified by body mass index (BMI) (30 kg/m2) and race.
RESULTS: Age-adjusted activity scores were higher PMong black than white men at all levels of BMI. Conversely, age-adjusted fitness was higher PMong whites than blacks for all BMI groups. Age-adjusted TT, SHBG and BT tended to be higher in blacks than whites across levels of BMI. For both black and white men, BMI was inversely associated with TT and SHBG, but not with BT. TT was not associated with activity for any BMI-race group, except obese white men where a positive association (p=0.01) was observed. The age and BMI adjusted mean TT concentration was 4.5 ng/ml for obese white men in activity quartile 1 and 5.6 ng/ml for obese white men in quartile 4; BT also was positively associated (p=0.0004) with activity only PMong obese white men. SHBG was unassociated with activity across BMI-race groups. TT and BT were not associated with fitness for any BMI-race group. SHBG was positively associated with fitness in obese white men (p=0.0008) and marginally positively associated with fitness in normal weight white men (p=0.06). PMong obese white men, TT and SHBG levels increased, but BT levels were similar across quartiles of fitness.
CONCLUSION: Physical activity is modestly positively associated with BT in obese white men. However, physical fitness was not significantly associated with BT or TT despite significant positive associations with SHBG. The different associations of fitness and activity with BT suggest different biological mechanisms. These results suggest physical activity might be a determinant of androgen levels independent of its association with BMI in obese white men only.