Seventy-three adult female cynomolgus monkeys fed an atherogenic diet were studied to determine the effect of two different combination contraceptive steroid preparations containing equivalent amounts of estrogen but different progestin components on plasma lipids and lipoproteins. Our hypothesis was that any high density lipoprotein (HDL) lowering effect of the contraceptive steroid preparations was proportional to the rise in total serum cholesterol caused by the progestins. For 2 years, one group (Ovral [Wyeth Laboratories], n = 23) received 75 micrograms norgestrel and 7.5 micrograms ethinyl estradiol daily, while another (Demulen [Searle & Co.], n = 25) received 150 micrograms ethynodiol diacetate and 7.5 micrograms ethinyl estradiol daily. The control group (n = 24) received no treatment. On average, the two oral contraceptive groups had higher total serum cholesterol and triglyceride concentrations but lower HDL cholesterol concentrations and smaller low density lipoproteins (LDL) compared with the control group. There was an inverse relationship between total serum cholesterol and HDL cholesterol for all three groups, but at any given total serum cholesterol concentration between 350 and 500 mg/dl, the Ovral group had HDL cholesterol concentrations that averaged 37% and 14% lower than the control and Demulen groups, respectively. The decrease in HDL concentrations with oral contraceptive treatment was associated with a sharp decrease in (HDL2b)gge protein (82% for Ovral and 59% for Demulen) and a corresponding increase in (HDL3b,c)gge protein as determined by gradient gel electrophoresis. Of 23 animals in the Ovral group, six had HDL subfractions greater than 10 nm diameter (HDL2b)gge compared with 22 of 24 animals in the control group. Although LDL size, on average, was smaller and plasma triglycerides were greater with oral contraceptive treatment compared with controls, there was no apparent relationship between LDL size and plasma triglyceride concentrations.(ABSTRACT TRUNCATED AT 250 WORDS)