The effectiveness of metformin, oral contraceptives, and lifestyle modification in improving the metabolism of overweight women with polycystic ovary syndrome: a network meta-analysis

We designed a network meta-analysis that investigated relatively different interventions that included the effects of metformin, oral contraceptives, and lifestyle modification on the metabolic parameters of patients with polycystic ovary syndrome. In addition, we searched for eligible interventions that improved the metabolism of glucose and lipids.

We searched the PubMed, EMBASE, and Cochrane Central databases from inception to May 2018. Publication types that were categorized as randomized controlled trials met our inclusion criteria. The main outcome included the homeostasis model assessment of insulin resistance, total cholesterol, low-density lipoprotein cholesterol, and total triglycerides. We performed both a pairwise meta-analysis and a network meta-analysis to evaluate the mean difference value and 95% credibility intervals, and we calculated the surface cumulative rank curve.

There were a total of 12 kinds of interventions: metformin, 2 mg cyproterone acetate plus 0.05 mg ethinylestradiol (EE/CA), 0.15 mg desogestrel plus 0.03 mg ethinylestradiol (EE/DSG), and 3 mg drospirenone plus 0.03 mg ethinylestradiol (EE/DRSP), lifestyle, exercise, diet, metformin + lifestyle, metformin + diet, EE/CA + lifestyle, metformin + EE/CA, and EE/DRSP + lifestyle from the 20 eligible RCTs that were included in this study. Our meta-analysis results showed that metformin + lifestyle (MD = −2.04, 95% CrI = −3.64 to −0.41), EE/CA + lifestyle (MD = −2.23, 95% CrI = −4.11 to −0.35), and EE/DRSP + lifestyle (MD = −2.59, 95% CrI = −4.66 to −0.50) resulted in lower in the levels of total cholesterol. Women treated with metformin + lifestyle (MD = −1.82, 95% CrI = −2.88 to −0.79), EE/CA + lifestyle (MD = −2.25, 95% CrI = −3.58 to −1.08), or EE/DRSP + lifestyle (MD = −2.29, 95% CrI = −3.69 to −1.07) exhibited significantly lower low-density lipoprotein cholesterol when compared with the placebo group. There was no significant difference between any of the interventions compared with a placebo in the levels of homeostasis model assessment of insulin resistance and total triglycerides. The surface cumulative rank curve revealed that metformin + lifestyle might be the best intervention with respect to the improvement of the homeostasis model of assessment insulin resistance and EE/DRSP + lifestyle appeared to be the best intervention for the reduction of total cholesterol and low-density lipoprotein cholesterol. Moreover, the metformin + diet intervention was more effective in reducing the level of total triglycerides.

For overweight polycystic ovary syndrome patients, our evidence revealed that EE/CA and EE/SRSP combined with metformin or lifestyle changes can reduce the adverse effects on glucose and lipid metabolism of the use of oral contraceptive agents alone. Conventional PCOS treatments, such as metformin, EE/CA, and EE/DRSP, combined with lifestyle control can be particularly effective in improving the homeostasis model assessment of insulin resistance and lipid metabolism.