Women are at greater risk than men of sustaining certain kinds of injury and diseases of collagen-rich tissues. To determine whether a high level of oestradiol has an acute influence on collagen synthesis in tendons at rest and in response to exercise, one-legged kicking exercise was performed for 60 min at 67% of maximum power by healthy, young oral contraceptive (OC) users when circulating synthetic (ethinyl) oestradiol was high (n = 11, HE-OC) and compared to similar women who had never used OCs when circulating endogenous oestrogen was low (n = 12, LE-NOC). Interstitial fluid was collected 24 h post-exercise through microdialysis catheters placed anterior to the patellar tendon in both legs and subsequently analysed for the amino-terminal propeptide of type I collagen (PINP), a marker of tendon collagen synthesis. To determine the long-term effect of OC usage, patellar tendon cross-sectional area (CSA) was measured by magnetic resonance imaging (MRI). A lower exercise-induced increase in tendon collagen synthesis was observed in HE-OC than in LE-NOC (DeltaPINP (mean +/- s.e.m.) 1.5 +/- 5.3 versus 24.2 +/- 9.4 ng ml(-1), P <0.05). Furthermore, serum and the interstitial peritendinous tissue concentrations of insulin-like growth factor I (IGF-I) and IGF-binding proteins showed a reduced bioavailability in HE-OC compared with results in LE-NOC. No difference in patellar tendon CSA was observed between groups. In conclusion, the selective increase in tendon collagen synthesis in LE-NOC but not HE-OC 24 h post-exercise is consistent with the hypothesis that oestradiol inhibits exercise-induced collagen synthesis in human tendon. The mechanism behind this is either a direct effect of oestradiol, or an indirect effect via a reduction in levels of free IGF-I. However, the data did not indicate any long-term effect on tendon size associated with chronic OC use.