In a double-blind, controlled, randomized, four-arm, bicentric clinical study, the effect of four oral contraceptives (OCs) on lipid metabolism was investigated. Four groups composed of 25 volunteers each (mean age 26.1 +/- 4.5 years; body mass index 21.9 +/- 2.8 kg/m(2)) were treated for six cycles with monophasic combinations containing 21 tablets with either 30 microg ethinyl estradiol (EE) + 2 mg dienogest (DNG) (30 EE/DNG), 20 microg EE + 2 mg DNG (20 EE/DNG), 10 microg EE + 2 mg estradiol valerate (EV) + 2 mg DNG (EE/EV/DNG), or 20 microg EE + 100 microg levonorgestrel (LNG; EE/LNG). The study was completed by 91 women. Blood samples were taken by venipuncture after at least 12 h fasting on Days 21-26 of the control cycle and Days 18-21 of the first, third, and sixth treatment cycle. There were clear differences between the effects of EE/LNG and the formulations containing estrogens and DNG. Although EE/LNG did not change the triglycerides levels, a significant increase was observed during treatment with the DNG-containing preparations. Although EE/LNG significantly reduced HDL-CH and HDL(2)-CH, there was a nonsignificant increase with the DNG-containing OCs. No change was observed in the levels of HDL(3)-CH. A significant rise in apolipoprotein A1 occurred during intake with the three DNG-containing formulations, but not with EE/LNG. In contrast to the women treated with combinations of estrogens and DNG, apolipoprotein B rose significantly in the women in the EE/LNG group. Lipoprotein (a) was significantly reduced by 30 EE/DNG and EE/LNG and remained unaltered with 20 EE/DNG and EE/EV/DNG. Altogether, the changes in lipid metabolism caused by the DNG-containing formulations appeared to be more favorable than those observed with EE/LNG. In OCs with DNG, the EE dose does not seem to play a major role with respect to the effect on lipids.