Ageing in male adults is typically accompanied by adiposity accumulation and changes in circulating sex hormone concentrations. We hypothesized that an ageing-associated increase in oestrogens and decrease in androgens would correlate with an increase in adiposity. 10-year prospective, observational study. A total of 190, community-dwelling men in the Japanese American Community Diabetes Study. At 0 and 10years, CT scanning quantified intra-abdominal fat (IAF) and subcutaneous fat (SCF) areas while plasma concentrations of oestradiol, oestrone, testosterone and dihydrotestosterone were measured by liquid chromatography-tandem-mass spectrometry at each time point. Multivariate linear regression analyses assessed correlations between 10-year changes in hormone concentrations and IAF or SCF, adjusting for age and baseline fat depot area. Participants were middle-aged [median 54.8 years, interquartile range (IQR) 39.9-62.8] men and mostly overweight by Asian criterion (median BMI 24.9, IQR 23.3-27.1) and with few exceptions had normal sex-steroid concentrations. Median oestradiol and dihydrotestosterone did not change significantly between 0 and 10years (P=.084 and P=.596, respectively) while median oestrone increased (P<.001) and testosterone decreased (P<.001). Median IAF and SCF increased from 0 to 10years (both P<.001). In multivariate analyses, change in oestrone positively correlated (P=.019) while change in testosterone (P=.003) and dihydrotestosterone (P=.014) negatively correlated with change in IAF. Plasma oestradiol and oestrone positively correlated with change in SCF (P=.041 and P=.030, respectively) while testosterone (P=.031) negatively correlated in multivariate analysis. Among 190 community-dwelling, Japanese American men, increases in IAF were associated with decreases in plasma androgens and increases in plasma oestrone, but not oestradiol, at 10years. Further research is necessary to understand whether changing hormone concentrations are causally related to changes in regional adiposity or whether the reverse is true.
Read full abstract