Treating Polycystic Ovaries with Rosiglitazone

Abstract & Commentary

Synopsis: Rosiglitazone treatment effectively restores ovulation in very heavy women with polycystic ovaries and insulin resistance.

Source: Sepilian V, Nagamani M. J Clin Endocrinol. Metab. 2005;90:60-65.

Sepilian and Nagamani from the University of Texas in Galveston reported the results of rosiglitazone treatment in 12 obese women with polycystic ovaries and insulin resistance. These women were very heavy, with an average BMI of 40.4. The fasting insulin levels (an average of 46.0 µU) were clearly in the range of hyperinsulinemia due to an increase in insulin resistance. Treatment with rosiglitazone, 4 mg daily for 6 months, improved insulin resistance, lowered total and free testosterone levels and increased the levels of sex hormone binding globulin. There was no significant change in body weights, but 11 of the 12 women began to ovulate.

Comment by Leon Speroff, MD

The value of this case series is the success with rosiglitazone in contrast to the reported failure of metformin treatment in women who are very obese. The effect of metformin has been controversial, with some suggestion that the ovulatory response is the result of the weight loss that often accompanies its use. In a study designed to control the effect of body weight, the administration of metformin was without effect on insulin resistance in extremely overweight women with polycystic ovaries.1 In another well-designed study, metformin again had no effect on insulin resistance when body weights remained unchanged, and in this study baseline weights and hyperinsulinemia were only modestly increased.2 In lean, anovulatory women with hyperinsulinemia, metformin treatment reduces hyperandrogenemia although there may be no change in body weight; however, a decrease in the waist-to-hip ratio accompanies a reduction in the hyperinsulinemia.3,4 This indicates that both obese and nonobese patients with hyperinsulinemia can respond to metformin treatment. The reasons for the differences among the studies are not apparent. Response is rapid and any weight loss is very modest, indicating that the salutary effect is a reduction in hyperinsulinemia.

A significant number (about 40%) of anovulatory women respond to metformin with ovulation and pregnancy.5-7 In a group of obese women with polycystic ovaries, 90% of the women treated with metformin and 50 mg clomiphene ovulated compared with 8% in the group treated with placebo and clomiphene.8 Most randomized clinical trials have indicated that metformin treatment increases ovulatory rates, especially in women resistant to clomiphene,9-11 although this has not been a uniform experience.12 The reasons for differing results are not clear, but ethnicity and patient selection are possible explanations.

Women with hyperinsulinemia have been reported to have a greater risk of early miscarriage and gestational diabetes, although a study that carefully matched patients for age and weight found no increase in gestational diabetes or pregnancy-induced hypertension.13-15 The risk of early miscarriage in these women is associated with the increased levels of PAI-1 that accompany hyperinsulinemia, suggesting the possibility that placental thrombosis induces miscarriage. Supporting the existence of these complications in pregnancy are the results with metformin treatment maintained throughout pregnancy, yielding a low rate of miscarriage and a profound reduction in gestational diabetes.16-18 Metformin has been used during pregnancy for decades in South Africa with no evidence of teratogenicity or neonatal hypoglycemia.19,20

The thiazolidinediones markedly improve insulin sensitivity and insulin secretion mainly by improved peripheral glucose utilization. Troglitazone decreases hyperinsulinemia, and produces improvements in metabolic abnormalities (decreased androgens, increased SHBG, decreased PAI-1 consistent with improved fibrinolytic capacity, and decreased LH) and a return to ovulation has been reported with this agent.21-23 However, troglitazone was withdrawn from the market because of liver toxicity. Pioglitazone and rosiglitazone have been reported to be free of liver toxicity; nevertheless, monitoring liver function every 2 months is recommended, a requirement that has limited the use of these agents for the purpose of treating anovulation and infertility. In addition, the glitazone drugs, when used for the treatment of diabetes, are associated with weight gain. Nevertheless, clinical studies indicate that rosiglitazone, 4 mg daily or b.i.d., is similar to metformin in inducing ovulation and improving insulin resistance,24-26 and pioglitazone, 45 mg/day, achieved a response in women who were resistant to metformin.27

Some clinicians choose to reserve diabetic drug treatment for those women who fail to respond to clomiphene. In view of the reduction in early pregnancy losses associated with metformin treatment, we prefer to start with metformin and add clomiphene if the patient has not ovulated after 3 months. The experience reviewed above indicates that rosiglitazone or pioglitazone should be considered for those patients who are very obese and fail to respond to metformin.


1. Ehrmann DA, et al. J Clin Endocrinol Metab. 1997;82:524-530.

2. Acaby O, et al. Fertil Steril. 1996;65:946-949.

3. Nestler JE, et al. J Clin Endocrinol Metab. 1997;82: 4075-4079.

4. Morin-Papunen L, et al. J Clin Endocrinol Metab. 2003;88:148-156.

5. Velazquez EM, et al. Metabolism. 1997;46:454-457.

6. Diamanti-Kandarakis E, et al. Eur J Endocrinol. 1998;138:269-274.

7. Fleming R, et al. J Clin Endocrinol Metab. 2002;87: 569-574.

8. Nestler JE, et al. N Engl J Med. 1998;338:1876-1880.

9. Vandermolen DT, et al. Fertil Steril. 2001;75:310-5.

10. Kocak M, et al. Fertil Steril. 2002;77:101-106.

11. George SS, et al. Hum Reprod. 2003;18:299-304.

12. Ng Eh, et al. Hum Reprod. 2001;16:1625-1631.

13. Mikola M, et al. Hum Reprod. 2001;16:226-229.

14. Bjercke S, et al. Gynecol Obstet Invest. 2002;54:94-98.

15. Haakova L, et al. Hum Reprod. 2003;18:1438-1441.

16. Glueck CJ, et al. Fertil Steril. 2001;75:46-52.

17. Glueck CJ, et al. Fertil Steril. 2002;77:520-525.

18. Jakubowicz DJ, et al. J Clin Endocrinol Metab. 2002; 87:524-529.

19. Coetzee EJ, et al. Diabetologia. 1979;16:241-245.

20. Coetzee EJ, et al. Diabetes Res Clin Pract. 1985;1: 281-287.

21. Dunaif A, et al. J Clin Endocrinol Metab. 1996;81: 3299-3306.

22. Ehrmann DA, et al. J Clin Endocrinol Metab. 1997;82: 2108-2116.

23. Azziz R, et al. J Clin Endocrinol Metab. 2001;86: 1626-1632.

24. Ghazeeri G, et al. Fertil Steril. 2003;79:562-566.

25. Shobokshi A, et al. J Soc Gynecol Investig. 2003;10: 99-104.

26. Belli SH, et al. Fertil Steril. 2004;81:624-629.

27. Glueck CJ, et al. Hum Reprod. 2003;18:1618-1625.

Leon Speroff, MD, Professor of Obstetrics and Gynecology, Oregon Health Sciences University, Portland is Editor for OB/GYN Clinical Alert.