RESUMO
OBJECTIVE: The aim of the study was to explore the optimal timing of gonadotropin initiation and the reasonable interval of luteinizing hormone (LH) levels in the gonadotropin-releasing hormone antagonist (GnRH-A) protocol. METHODS: A retrospective cohort study was conducted to analyze the data concerning the oocyte retrieval cycles from 1,361 cases with the GnRH-A protocol implemented. The ovarian responses (including AMH, AFC) in these patients were divided into the poor ovarian response group (an antral follicle count [AFC] ≤ 6, n = 394), the normal ovarian response group (an AFC > 6 and < 15, n = 570), and the high ovarian response group (an AFC ≥ 15, n = 397), according to the AFC. The patients were sub-grouped according to LH levels on the protocol initiation day, and the clinical outcomes (including dose of Gn initiation, Gn administration days, GnRH-ant administration days, P levels on the HCG day, E2 levels on the HCG day, LH levels on the HCG day, number of embryos transferred, total fertilization rate, embryo implantation rate(%), proportion of 2PN, proportion of good-quality embryos, endometrial thickness on the hCG injection day(mm), moderate to severe OHSS, AFC on the initiation day, proportion of type A endometrium on the hCG injection day, clinical pregnancy rate, biochemical pregnancy rate, early abortion rate, ectopic pregnancy rate) were compared. RESULTS: On the GnRH-A protocol initiation day, among all patients with different ovarian responses, the body mass index (BMI) in those with an LH ≥ 5 IU/L was lower. The differences in pregnancy outcomes between the LH < 5 IU/L group and the LH ≥ 5 IU/L group were not statistically significant across the different ovarian response groups, but the LH < 5 IU/L group had a higher proportion of good-quality embryos (80.3±24.9 vs. 74.8±26.9, P =0.035) than the LH≥5IU/Lgroup in those with poor ovarian response. The total fertilization rate (82.2±18.1 vs 85.4±15.1, P =0.021) and proportion of two pronuclei (2PN) (69.0±20.9 vs 72.7±19.9, P =0.035) were higher in the LH ≥ 5 IU/L group than the LH<5 IU/L group for those with normal ovarian responses. The embryo implantation rate (41.4±41.3 vs 52.6±43.4, P =0.012) was higher in the LH ≥ 5 IU/L group than in the LH<5 IU/L group in those with high ovarian response. The results of the multivariate logistic analysis showed that the age of the female partner, number of embryos transferred, proportion of good-quality embryos, endometrial thickness on the hCG injection day, and moderate- to-severe ovarian hyperstimulation syndrome (OHSS) were independent factors correlated with the outcome of live births (P < 0.05). CONCLUSION: The LH levels on the gonadotropins (Gn) initiation day in the GnRH-A protocol will not affect pregnancy outcomes.