Comparative Effectiveness of Mild or Conventional GnRH-Antagonist Protocols for Ovarian Stimulation in Poor Responders (Poseidon Group 4)

Abstract

Background and Aims: A panel of experts (the Poseidon Group) introduced a new and more detailed stratification for poor ovarian responders in order to predict the prognosis of IVF outcome according to the sensitivity to FSH. However, various arguments about the management strategy of these patients still remain, including the convenience and the cost. Therefore, this study was conducted to compare the efficacy of mild and conventional GnRH antagonist ovarian stimulation prescribed in patients classified in Poseidon Group 4. Methods: This retrospective cohort study included 359 poor responder patients (Poseidon Group 4) treated with mild or conventional GnRH antagonist stimulation regimens from 8/2017 to 7/2019 at Tam Anh Hospital ART Center. The main outcomes were the index of Follicular Output Rate (FORT) or Follicle to Oocyte Index (FOI), the number of day-2 embryos and top-quality embryos obtained. The t-test and Mann–Whitney U test in SPSS v25.0 was used to analyze the continuous data and Chi-squared/Exact test was used for binary variables. Multiple linear regression analysis was done by using Stata versions 15.0 to measure association between primary endpoints with stimulation regimen controlled for covariates and possible confounding factors. Results: In the overall group of poor responders, the conventional GnRH antagonist protocol performed better than the mild protocol. Subsequently, data were analyzed according to the AFC. In women with AFC < 3, no significant differences were observed between the 2 regimens regarding FORT (p = 0.71), FOI (p = 0.12), the number of day-2-embryos (p = 0.052) and the number of top-quality embryos (p = 0.26). In contrast, in women with AFC ≥ 3, mild stimulation regimen resulted in significantly poorer outcome compared to the conventional GnRH antagonist regimen, regarding FORT (p < 0.01), FOI (p < 0.01), the number of day-2-embryos (p < 0.01) and top-quality embryos (p = 0.01). Conclusions: Considering poor responders classified in Poseidon Group 4, both ovarian stimulation regimens resulted in similar outcome for patients with a very low ovarian reserve (AFC < 3). In contrast, the GnRH conventional antagonist protocol with maximum initial FSH dose (300–375 IU/day) and supplementary LH (75–150 IU/day) was more effective than the mild one for patients whose ovarian reserve was less reduced. The Clinical Trial was approved by the Ethnical Biomedical Research Committee Tam Anh General Hospital.