Objective: To systematically evaluate the effectiveness of the gonadotropin-releasing hormone (GnRH) antagonist on in vitro fertilisation (IVF) in patients with polycystic ovary syndrome (PCOS). Methods: Nine types of databases were searched by computer, and nine types of relevant journals were searched manually. Randomized, controlled trials of the effects of the GnRH antagonist and GnRH agonist on IVF-ET treatment in the patients with PCOS were included. A meta-analysis was conducted following a quality evaluation. Results: Seven published studies (755 patients) were included. A meta-analysis was conducted following a quality evaluation. There were no significant differences in the amount of gonadotropin (Gn) (MD = −2.05; 95% CI: −4.14–0.05], E2 levels on the day of hCG administration (MD = −156.13; 95% CI: −389.91–77.64), the number of oocytes retrieved (MD = −0.38; 95% CI: −2.32–1.56), the clinical pregnancy rate (Peto OR = 1.08; 95% CI: 0.80–1.45), and the abortion rate (Peto OR = 0.91; 95% CI: 0.54–1.53) between the GnRH antagonist group and the GnRH agonist group. The OHSS rate of the GnRH antagonist group was lower than that of the GnRH agonist group, and the difference was statistically significant (Peto OR = 0.36; 95% CI: 0.25–0.52). Conclusions: Compared with the GnRH agonist protocol, the GnRH antagonist protocol could significantly reduce the risk of OHSS. The clinical pregnancy rates for these two protocols were similar.