Effects of Metformin on Spontaneous and Clomiphene-Induced Ovulation in the Polycystic Ovary Syndrome

Abstract

Obese women with the polycystic ovary syndrome are relatively unresponsive to the induction of ovulation by clomiphene. We hypothesized that reducing insulin secretion by administering metformin would increase the ovulatory response to clomiphene. We performed oral glucose-tolerance tests before and after the administration of 500 mg of metformin or placebo three times daily for 35 days in 61 obese women with the polycystic ovary syndrome. Women who did not ovulate spontaneously were then given 50 mg of clomiphene daily for five days while continuing to take metformin or placebo. Serum progesterone was measured on days 14, 28, 35, 44, and 53, and ovulation was presumed to have occurred if the concentration exceeded 8 ng per milliliter (26 nmol per liter) on any of these days. Twenty-one women in the metformin group and 25 women in the placebo group were given clomiphene because they did not ovulate spontaneously during the first phase of the study. Among the 21 women given metformin plus clomiphene, the mean (±SE) area under the serum insulin curve after oral glucose administration decreased from 6745±2021 to 3479±455 μU per milliliter per minute (40.5±12.1 to 20.9±2.7 nmol per liter per minute, P=0.03), but it did not change significantly in the 25 women given placebo plus clomiphene. Nineteen of the 21 women (90 percent) who received metformin plus clomiphene ovulated (mean peak serum progesterone concentration, 23.8±3.4 ng per milliliter [7.6±10.9 nmol per liter]). Two of the 25 women (8 percent) who received placebo plus clomiphene ovulated (P<0.001). Overall, 31 of the 35 women (89 percent) treated with metformin ovulated spontaneously or in response to clomiphene, as compared with 3 of the 26 women (12 percent) treated with placebo. The ovulatory response to clomiphene can be increased in obese women with the polycystic ovary syndrome by decreasing insulin secretion with metformin.