Diver · Clinical endocrinology 2003 · cross-sectional comparative study · n=18

Diurnal rhythms of serum total, free and bioavailable testosterone and of SHBG in middle-aged men compared with those in young men.

Cited 340 times in the scientific literature.

Level 4 - case-series / case-control

Cross-sectional comparative physiological study of two age groups

PubMed 12780747 · doi:10.1046/j.1365-2265.2003.01772.x · record verified 2026-08-27

What was done

Researchers measured serum total testosterone, free testosterone, bioavailable testosterone, and sex hormone-binding globulin (SHBG) from blood samples collected every 30 minutes over a 24-hour period in 10 young men and 8 middle-aged men. Rhythms were evaluated using time series analysis by least squares estimation to determine mesor, amplitude, and acrophase (timing of peak).

What was found

Both groups exhibited a significant diurnal rhythm in all measured hormones, with at least a 43% decrease in total testosterone from peak to nadir across all participants. Comparing young versus middle-aged men for total testosterone showed no significant differences in mesor (P = 0.306), amplitude (P = 0.061), or acrophase (P = 0.972). Bioavailable testosterone displayed significant circadian rhythmicity in both cohorts (middle-aged P < 0.001, young P = 0.014) with no difference in mesor (P = 0.175) or acrophase (P = 0.978), though amplitude differed significantly (P = 0.031). Free testosterone rhythms were significant in young (P < 0.001) and middle-aged men (P = 0.002), with no group difference in mesor (P = 0.094) or acrophase (P = 0.698). Acrophases for all testosterone forms occurred between 07.00 h and 07.30 h. SHBG peaked in mid-afternoon (15.12 h in young vs 15.40 h in middle-aged; acrophase P = 0.477; mesor P = 0.083), with older men exhibiting a larger amplitude (P = 0.044).

Why it matters

These findings counter the belief that diurnal testosterone variations flatten or disappear during normal aging, demonstrating that healthy men retain morning androgen peaks into their seventh decade.

Limits

The study is constrained by a very small sample size (n = 18 total). The abstract does not define the exact age ranges, baseline health metrics, sleep duration, or lifestyle controls.

Cited by