Metformin and glitazones: do they really help PCOS patients?

BACKGROUND

Polycystic ovarian syndrome (PCOS) leads to a multitude of clinical and biochemical alterations in patients. Metformin and the thiazolidenediones (TZDs)--which have insulin-sensitizing properties--are believed to be effective in minimizing the changes caused by this syndrome.

OBJECTIVES

Our goal was to assess the evidence for the use of TZDs or metformin in the treatment of PCOS patients. In addition, we sought to assess and compare the effectiveness of metformin vs TZDs in the clinical and biochemical regression of PCOS based on available randomized controlled trials (RCTs).

SEARCH STRATEGY

We searched Medline (January 1966 to January 2007), PubMed (January 1954 to January 2007), Google Scholar search engine (through January 2007), and reference lists of articles. We also contacted researchers and clinicians in the field.

SELECTION CRITERIA

We reviewed RCTs involving women diagnosed with PCOS (based on 1990 the National Institutes of Health Criteria) who were treated with either metformin or TZDs. Trials were limited to those that were testing only the effects of either of these agents as their primary endpoint.

MAIN RESULTS

A total of 115 trials were obtained, of which only 33 trials met the inclusion criteria. Ultimately, 31 trials involving total 1892 patients were included in the analysis (23 metformin, 2 rosiglitazone, 1 pioglitazone, 5 troglitazone) with 2 unobtainable trials. There was insufficient data to compare metformin to the TZDs in any parameter because the literature often contained inadequate quantitative data, or there were too few published trials. As a result, we performed the meta-analysis for metformin only. Among the outcomes examined, the only statistically significant changes were minimal decreases with metformin in ovulation rates and luteinizing hormone to follicle stimulating hormone ratio (LH/FSH), and an increase in fasting insulin. There was no clinically significant change with metformin in ovulation rate, pregnancy rate, body mass index, waist-to-hip ratio, hirsutism (F-G score), LH/FSH, fasting insulin, fasting blood glucose, total testosterone, free testosterone, androstenedione, and dehydroepiandrosterone sulfate.

AUTHORS' CONCLUSIONS: There is a paucity of data from RCTs to compare the effectiveness of metformin vs TZDs as well as the effects of either agent in treating the clinical and biochemical features of PCOS. Further research involving RCTs with larger sample sizes is needed before any recommendation can be made on the usefulness of these agents in the treatment of PCOS.