In a 6-month, randomized, double-blind study the effects of two combined oral contraceptives containing 150 micrograms desogestrel and either 20 or 30 micrograms ethinylestradiol on hemostatic parameters were investigated in 1633 healthy women. Compared with baseline, the 30 micrograms ethinylestradiol formulation increased prothrombin fragment 1 + 2 (+72.2%), D-dimer (+42.4%) and protein C activity (+6.1%), whereas antithrombin-III activity (-6.3%) and protein S activity (-19.7%) were decreased. The use of the 20 micrograms ethinylestradiol formulation was associated with the same pattern of changes, but with lower magnitude (F1+2 + 61.1%, D-dimer +36.0%, antithrombin III -5.3%, protein C +4.6% and protein S-16.0%). The changes from baseline were significantly smaller in the 20 micrograms ethinylestradiol group for D-dimer, antithrombin III and protein S than in the 30 micrograms ethinylestradiol group (p = 0.019, p = 0.038 and p = 0.001, respectively). One woman with a combined deficiency of proteins C and S developed deep venous thrombosis while using the 20 micrograms ethinylestradiol formulation. Use of both formulations was associated with a shift of the coagulation-fibrinolysis balance to an enhanced fibrin-generating and fibrin-degradating activity. The less-pronounced effect on hemostasis with the 20 micrograms ethinylestradiol preparation is reassuring with regard to thromboembolic risk in general. However, women with coagulation inhibitor deficiency should be advised not to use oral contraceptives.