Comment: The quality of evidence is downgraded by study quality (lack of blinding, inadequate intention-to-treat) and by imprecise results (few patients and wide confidence intervals).
A Cochrane review [Abstract] 1 included 44 trials involving a total of 2253 women. Metformin was less effective in improving hirsutism compared to combined oral contraceptives (COC) in the subgroup BMI 25 kg/m² to 30 kg/m² (MD 1.92, 95% CI 1.21 to 2.64, 5 RCTs, n = 254), but there were no clear difference in the subgroup BMI under 25. Either metformin alone or COC alone may be less effective in improving hirsutism compared to metformin combined with COC. Metformin resulted in a higher incidence of gastrointestinal and a lower incidence of non-gastrointestinal severe adverse effects requiring stopping of medication.
A comparative study 2 divided women with polycystic ovary syndrome (PCOS) either to receive COC containing 3 mg drospirenone/30 mcg EE (ethinylestradiol) (DRSP group, n=60) or 2 mg chlormadinone acetate/30 mcg EE (CMA group, n=60). After 6 months of continuous treatment, a significant improvement was observed in hirsutism in both groups, the drop in Ferriman-Gallwey score was around 8 units (scale 0 to 36). also in acne lesions (scale 0-3) the imrovement was significant: with DRSP from2.32 ± 0.89to 0.31 ± 0.78 and with CMA 2.32 ± 0.89 to0.45 ± 0.98. Among glucose, insulin levels and HOMA-IR, there were statistically significant higher levels and the hormonal parameters (LH, FSH, prolactin, testosterone and DHEA-S) were statistically significant lower in both groups.
A systematic review and meta-analysis 3 included 19 RCTs. Fourth-generation COC resulted in lower BMI (mean difference [MD] 1.17 kg/m2, 95% CI 0.33 to 2.02]) and testosterone (MD 0.60 nmol/L, 95% CI 0.13 to 1.07) compared with third-generation agents, but no difference was seen in hirsutism. EE/cyproterone acetate was better in reducing hirsutism as well as biochemical hyperandrogenism and BMI (MD 0.62 kg/m2, 95% CI 0.05 to1.20) compared with conventional COCs.There was no difference in hirsutism between high and low EE doses.
Another systematic review and meta-analysis 4 included 36 RCTs. For hirsutism, no differences were seen when comparing metformin vs COC, nor when comparing COC vs combination treatment with metformin and COC. Metformin was inferior on free androgen index (FAI), sex hormone binding globulin and testosterone compared with COC. COC was inferior for FAI and SHBG compared with combination treatment, whereas testosterone did not differ.
Another systematic review and meta-analysis 5 included 56 studies. In adolescents, metformin was better than COC for BMI (-4.02, 95% CI -5.23 to -2.81); COC was better than metformin for menstrual regulation (MD -0.19, 95% CI -0.25 to -0.13, P < 0.00001). In low-quality evidence in adults, metformin was better than placebo for BMI; metformin was better than COC for fasting insulin, whereas COC was better than metformin for irregular cycles (MD 12.49, 95% CI 1.34 to 116.62). COC alone was better than the combination with an anti-androgen for BMI (MD -3.04, 95% CI -5.45 to -0.64).
Primary/Secondary Keywords