A randomized prospective study of the metabolic effects of four low-estrogen oral contraceptives

Healthy, nonsmoking, normotensive, well-motivated young women were assigned at random to one of four different commercial, low-estrogen oral contraceptives. Measurements of biochemical parameters were made on two blood specimens collected from fasting subjects late in the pretreatment cycle, then again late in each treatment cycle for 12 months. All the women assigned to one product (0.5 mg norethindrone + 35 micrograms ethinyl estradiol) dropped out of the study before the end of the fifth cycle, but discontinuations with the other three products were few. Though the numbers of subjects were small, the groups were closely matched, and most of the metabolic differences were statistically significant. Products containing ethynodiol diacetate and norethindrone were associated with increases in serum total cholesterol and triglycerides but decreases in high-density-lipoprotein-cholesterol. In contrast, the levonorgestrel-containing preparation produced significantly less of an effect on these tests. A similar pattern was seen with a range of blood coagulation and fibrinolytic factors. Minimal alterations were seen with the levonorgestrel preparation, whereas those containing norethindrone or ethynodiol diacetate showed marked increases in factors I, VII, VIII and X and plasminogen, associated with a decrease in antithrombin III. Differences in the metabolic impact of the various commercially available low-estrogen preparations, combined with their effects on intermenstrual bleeding, allow a choice of the progestogen component most suitable for general use.