Abstract
CONTEXT: In men, circulating 17β-estradiol originates primarily from peripheral aromatization of testosterone particularly in adipose tissue. The effect of ageing and obesity on circulating estradiol remains unclear. OBJECTIVE: Determine five-year changes in serum estradiol and the association with testosterone and fat mass in Australian men. DESIGN: Longitudinal cohort study. At baseline and five-year follow-up, socio-demographic and health-related data including behaviors, chronic conditions, and medication use were collected by questionnaire. Estradiol and testosterone were assayed by liquid chromatography-tandem mass spectrometry and sex hormone-binding globulin by immunochemiluminescent assay. Fat mass was assessed by dual-energy X-ray absorptiometry. PARTICIPANTS: Community-dwelling men aged 35 years and older at enrollment, resident in the northern and western suburbs of Adelaide without established disease of, or medications affecting, the hypothalamus-pituitary-gonadal axis (n = 725). MAIN OUTCOME MEASURES: The dependence of change in serum estradiol over five years on age, testosterone and fat mass after adjustment for multiple confounders. RESULTS: At baseline, mean age was 53.0 ± 10.8 years. Mean serum estradiol levels at baseline and five-year follow-up were 94.9 ± 34.8 and 89.4 ± 30.4 pmol/L respectively (-1.1 pmol/L/year). On multivariable analyses, estradiol change was associated with changes in testosterone (B-estimate = 2.719, standard error = 0.369, p˂0.001), but not age or total fat mass. Change in testosterone/estradiol ratio was inversely associated with change in fat mass (B = -1.450, SE = 0.575, p = 0.012), and this was consistent across quartiles of fat mass change. CONCLUSIONS: In healthy men, circulating estradiol is primarily dependent on testosterone. With increasing fat mass, estradiol decreases less than testosterone. From a clinical standpoint these data indicate that obesity is associated with a change in the testosterone to estradiol ratio, but a change in estradiol does not occur unless some other pathology is present.