OBJECTIVES: To compare the effects of metformin and orlistat in terms of reduction in weight or BMI, and improvement of ovulation rates, endocrinologic and lipid profiles, and occurrence of adverse events among overweight or obese women diagnosed with PCOS.

SEARCH METHODS: We searched Medline, OVID, HERDIN, EMBASE, Cochrane Library and ClinicalTrials.gov for head to head clinical trials of metformin versus orlistat for the treatment of overweight and obese women with PCOS. We also contacted the pharmaceutical companies and did hand-searching to look for related studies.

SELECTION CRITERIA: Only randomized controlled trials comparing metformin and orlistat as treatment for overweight and obese PCOS women were included. Other inclusion criteria included: trial period of at least 3 months duration, participants, of any ethnicity, 18-40 years old, who are overweight or obese, and studies with or without non-pharmacologic interventions as part of the treatment regimen.

DATA COLLECTION AND ANALYSIS: Titles and abstracts identified through the search strategies were screened by two reviewers. Two authors extracted data on publication characteristics, inclusion and exclusion criteria, intervention and co-intervention, primary and secondary outcomes, and details of study design. Two authors assessed the quality and risk bias of each RCT based on random sequence generation, allocation concealment, blinding of participants, caregivers, and assessors, attrition bias, incomplete outcome data, selective reporting, and publication bias.

MAIN RESULTS: We included 5 RCTs (n=221). Overall, treatment effects of orlistat and metformin showed no significant difference in the following outcomes: ovulation rates (RR 0.78; 95% CI 0.41, 1.49), reduction of BMI (MD -0.47; 95%CI:-1.53,0.59), serum testosterone levels (MD -2.15;95% CI -9.64, 5.33), free androgen index MD 3.26; 95% CI -7.91, 14.43), homeostatic model assessment-insulin resistance (3.70; 95% CI -6.74, 14.15), fasting insulin (MD 7.86; 95% CI -3.09, 18.81), HDL-C (MD -1.19 ; 95% CI -4.78, 7.16) and triglycerides (MD -1.95; 95% CI -8.81, 4.90). Orlistat was significantly better than metformin in reducing total cholesterol (MD -6.60; 95% CI -10.79, -2.41), and LDL (MD -5.04; 95% CI -9.64, 5.33), free androgen index (MD 3.26; 95% CI -7.91, 14.43), homeostaic model assessment-insulin resistance (3.70; 95% CI -6.74, 14.15), fasting insulin (MD 7.86; 95% CI -3.09, 18.81), HDL-C (MD -1.19 ; 95% CI -4.78, 7.16) and triglycerides (MD -1.95; 95% CI -8.81, 4.90). Orlistat was significantly better than metformin in reducing total cholesterol (MD -6.60; 95% CI -10.79, -2.41), and LDL (MD -5.04); 95% CI  -9.99, -0.09), and had less adverse events (RR 0.37, 95% CI 0.14, 0.96).

AUTHORS' CONCLUSIONS: Metformin and Orlistat have similar effects on weight loss, ovulation rates, and endocrinologic profiles of obese women with PCOS. Orlistat is more effective than metformin in decreasing total cholesterol and LDL -C levels, and has less adverse events than metformin. Therefore, we may recommend orlistat to overweight or obese women with PCOS who also have dyslipidemia. However, caution is given to our interpretations since small sample size, low quality of RCTs, and wide confidence intervals of pooled estimates significantly influence interpretation and recommendations. RCTs with adequately powered study populations are recommended to confirm findings of this review.