〔Abstract〕 Objective To investigate the clinical effects and pregnancy outcomes of using luteal phase long protocol and GnRH antagonist protocol in patients with polycystic ovary syndrome(PCOS) who have failed their first GnRH antagonist protocol therapy. Methods The clinical data of 163 PCOS patients who underwent IVF/ICSI-ET were retrieved. After the failure of their first GnRH antagonist protocol treatment, they were divided into two groups in the second controlled ovarian hyperstimulation(COH) cycle: Luteal phase long protocol group(n=95) and GnRH antagonist protocol group(n=68). A retrospective analysis and comparison of basic clinical data, clinical and laboratory indicators, and pregnancy outcomes between two groups were conducted. Results (1) There was no statistically significant difference in basic clinical indicators between two group except LH.(2) Compared the first and second cycle treatments of patients in the luteal phase long protocol group, the initiation dose of gonadotropin(Gn), total number of Gn days, total Gn usage, estradiol(E2) on the day of hCG injection, number of retrieved eggs, oocyte maturation rate, 2PN fertilization rate, 2PN cleavage rate, blastocyst formation rate, high-quality blastocyst formation rate, and moderate to severe OHSS rate were significantly higher than those in the first GnRH antagonist cycle(P<0.05). The GnRH antagonist protocol group also showed similar improvements.(3) The comparison of the second COH cycle between two groups showed that the total number of Gn days, total Gn usage, and total Gn cost in the luteal phase long protocol group were significantly higher(P<0.05), while the E2and LH on the day of hCG injection, and the maturation rate of eggs were significantly lower than those in the GnRH antagonist protocol group(P<0.05). However, there was no statistically significant difference in the number of retrieved eggs, 2PN fertilization, 2PN cleavage, blastocyst formation rate, high-quality blastocyst formation rate, and OHSS rate between the two groups;(4) The comparison of fresh transplantation cycles for the second COH cycle between the two groups showed that the luteal phase long protocol fresh transplantation rate, implantation rate, clinical pregnancy rate, and live birth rate were slightly higher than those of the GnRH antagonist protocol group, but the difference was not statistically significant. Comparing the outcomes of pregnancy following the initial frozen-thawed embryo transfer(FET) between two groups, the biochemical pregnancy rate and clinical pregnancy rate of the GnRH antagonist protocol group were higher than those of the luteal phase long protocol group(P<0.05). However, no significant statistical variations were found in implantation rate, live birth rate, neonatal gestational age, and birth weight. Conclusion For PCOS patients who fail the first GnRH antagonist protocol, an appropriate increase in the initiating dose and usage of Gn can achieve satisfactory pregnancy outcomes with both protocols. Compared with change to a luteal phase long protocol, reusing the GnRH antagonist protocol still maintains its long-standing advantages, such as shorter total Gn days, lower costs, and better patient compliance.