Metformin versus Ethinyl Estradiol-Cyproterone Acetate in the Treatment of Nonobese Women with Polycystic Ovary Syndrome: A Randomized Study

Abstract

Insulin resistance and hyperinsulinemia are characteristic of polycystic ovary syndrome (PCOS), especially in obese women. Insulin-sensitizing drugs such as metformin have improved ovulation and lowered serum testosteronelevels in nonobese women with PCOS-even those with normal insulin sensitivity. This trial randomized 17 nonobese women (body mass index [BMI] less than 25 kg/m 2 ) with PCOS to receive either 500 mg metformin twice a day for 3 months, followed by 1 gm twice daily for 3 months longer, or oral contraceptive pills containing 35 μg ethinyl estradiol and 2 mg cyproterone acetate (EE-CA). Contraceptive pills were given for 3 weeks each month. Polycystic ovaries were demonstrated by vaginal ultrasonography in all patients. BMI declined significantly at 3 months and slightly at 6 months in metformin-treated women, as did the waist-to-hip ratio (WHR). BMI increased in the EE-CA group, with no change in the WHR. Hirsutism decreased slightly at 6 months on EE-CA but did not change during metformin therapy. Mean ovarian volume decreased significantly with EE-CA, but did not change substantially during metformin therapy. In neither group did the area under the curve for glucose change significantly after 6 months of treatment. Fasting insulin levels decreased significantly in metformin-treated women, as did early-phase insulin secretion. Early-phase C-peptide secretion tended to increase in these women. Hepatic insulin extraction in the fasting state rose significantly, but did not change in the EE-CA group. Fasting energy expenditure decreased with metformin therapy but increased significantly in the EE-CA group. Insulin-stimulated serum-free fatty acid levels tended to decrease after 6 months of metformin therapy but increased significantly with EE-CA treatment. Women in both groups had significantly decreased levels of serum testosterone, androstenedione, and dehydroepiandrosterone. The free-androgen index also declined significantly. In the EE-CA group. the LDL/HDL ratio decreased while serum triglycerides increased. Metformin therapy reduced hyperinsulinemia and hyperandrogenism in these nonobese women with PCOS. Menstrual cycles were improved in some instances. EE-CA treatment also improved hyperandrogenism and did not significantly worsen insulin sensitivity. Metformin offers an effective approach to treating such women for anovulation.