Hill · Journal of endocrinological investigation 2008 · Controlled crossover physiological trial · n=12

Exercise and circulating cortisol levels: the intensity threshold effect.

Level 3 - non-randomized controlled study

Within-subject repeated measures controlled experimental trial without explicit mention of randomization.

PubMed 18787373 · doi:10.1007/BF03345606 · record verified 2026-08-26

What was done

Twelve active, moderately trained men completed four separate 30-minute sessions on different days: resting control, and exercise at 40%, 60%, and 80% of maximal oxygen uptake (VO2max). Blood was sampled immediately before and after each session to measure serum cortisol, ACTH, and plasma volume shifts, while controlling for time of day, diet, psychological stress, and prior activity.

What was found

Cortisol changed pre- to post-session by -6.6 ± 3.5% (resting control), +5.7 ± 11.0% (40% VO2max), +39.9 ± 11.8% (60% VO2max), and +83.1 ± 18.5% (80% VO2max). Changes at 60% and 80% were significantly greater than resting control and 40%, and differed significantly from each other. ACTH mirrored cortisol changes, but only reached a statistically significant pre-to-post increase at 80% VO2max (p < 0.05). Plasma volume shifted by +2.2 ± 3.0%, -9.9 ± 5.0%, -15.6 ± 3.5%, and -17.2 ± 3.3% across the control, 40%, 60%, and 80% sessions, respectively. After accounting for plasma volume contraction and circadian decline, 40% VO2max exercise resulted in an actual net reduction in circulating cortisol.

Why it matters

The findings establish an exercise intensity threshold (at or above approximately 60% VO2max for 30 minutes) required to stimulate the hypothalamic-pituitary-adrenal axis, demonstrating that low-intensity exercise does not provoke a true physiological stress hormone response.

Limits

The study evaluated a very small cohort (n = 12) restricted exclusively to young, moderately trained men, limiting generalizability to women, untrained individuals, or elite athletes. Only acute responses to 30 minutes of exercise were measured, without post-exercise recovery kinetics beyond immediate post-exercise sampling. The abstract does not specify whether testing session orders were randomized.

Cited by