Background: Serum testosterone concentration declines and sex hormone-binding globulin (SHBG) rises during male ageing. However, whether these changes reflect physiological ageing or accumulation of age-related comorbidities remains uncertain.
Objective: We examined the longitudinal changes in serum testosterone and SHBG concentrations in middle-aged to older men from the U.K. Biobank, the concordance between baseline and follow-up values, and their relationships with concomitant changes in sociodemographic and lifestyle factors.
Methods: Immunoassay serum total testosterone (n=7,813) and SHBG (n=6,491) were measured at baseline (2006-2010) and follow-up (2012-2013). Bland-Altman analyses and concordance correlation of repeated hormone measurements were conducted. Associations of changes in hormone concentrations with lifestyle and medical factors were explored using Spearman’s rank correlation and bivariate hexbin plots.
Results: Over 4.3 years follow-up, there was a negligible mean change in serum total testosterone concentration (±SE) of +0.06±0.03 nmol/L, whereas mean SHBG concentration increased by +3.7±0.12 nmol/L. Concordance between measurements were 0.67 (95% confidence interval [CI]: 0.66-0.69) for total testosterone and 0.83 (CI=0.82-0.84) for SHBG concentrations. Changes in serum testosterone correlated with changes in SHBG (Spearman’s rank ρ=0.33, CI=0.30-0.35), and inversely with changes in BMI (ρ=-0.18, CI=-0.20 to -0.16), and waist circumference (ρ=-0.13, CI=-0.15 to -0.11).
Conclusions: In relatively healthy middle-aged to older men, there was no evidence of a decline in mean testosterone concentrations over time, although mean SHBG concentrations increased. Although there was negligible net change observed, concomitant changes in SHBG, BMI, and waist circumference explained some of the variation in repeat measurements of testosterone. The relative stability of total testosterone concentration during follow-up supports the concept that healthy men can preserve endogenous testosterone production during ageing. These findings facilitate future research investigating associations of baseline testosterone with prospective health outcomes in ageing men.