Four RCTs, with a total of 1,066 participants (sample size 100 to 626) were included in the review. Allocation concealment was adequately explained in all studies and both patients and investigators were blinded to treatment in three out of four studies (neither were blinded in the remaining study). Only one study analysed data on an intention-to-treat basis.
No difference was seen between the intention-to-treat and per-protocol analyses; meta-analyses were, therefore, reported per-protocol only.
Clomiphene citrate versus metformin (three studies, 841 participants):
No significant differences between clomiphene citrate and metformin were observed for any outcome measure. The odds ratios were 1.17 (95% CI: 0.16 to 8.61) for live birth, 1.22 (95% CI: 0.23 to 6.55) for pregnancy rate and 1.55 (95% CI: 0.40 to 5.99) for cumulative ovulation rate. Significant heterogeneity was detected for all three outcomes. The odds ratios were 1.58 (95% CI: 0.77 to 3.25) for abortion and 0.71 (95% CI: 0.22 to 2.25) for adverse events leading to discontinuation. No significant heterogeneity was observed for either outcome.
Metformin plus clomiphene citrate versus clomiphene citrate (three RCTs, 966 participants):
No significant differences between metformin plus clomiphene citrate and clomiphene citrate alone were observed for any outcome measure. The odds ratios were 0.99 (95% CI: 0.70 to 1.40) for live birth, 0.85 (95% CI: 0.62 to 1.15) for pregnancy and 0.84 (95% CI: 0.60 to 1.18) for cumulative ovulation rate. No significant heterogeneity was detected for any of these outcomes. The odds ratios were 0.74 (95% CI: 0.43 to 1.26) for abortion and 0.23 (95% CI: 0.04 to 1.24) for adverse events leading to discontinuation. Significant heterogeneity was observed for adverse events.
Metformin plus clomiphene citrate versus metformin (two RCTs, 741 participants):
Combination therapy was more effective than metformin alone with respect to live births (odds ratio 0.23, 95% CI: 0.13 to 0.40), pregnancy rates (odds ratio 0.23, 95% CI: 0.14 to 0.37) and ovulation rates (odds ratio 0.23, 95% CI: 0.15 to 0.34). No significant heterogeneity was observed for any outcome. There was no significant difference between metformin and combination therapy in either abortion rate (odds ratio 0.47, 95% CI: 0.22 to 1.0) or adverse events leading to discontinuation (odds ratio 0.86, 95% CI: 0.23 to 3.04). No significant heterogeneity was detected for either outcome.