Search for a command to run...
Abstract Study question In FET cycles, which are the mid-luteal values of progesterone and estradiol that better correlate with the live birth rate? Summary answer The optimal values associated with the highest live birth rate range between 10.8 and 18.0 ng/ml for progesterone and 350.0 and 490.3 pg/ml for estradiol. What is known already In the last years, the FET cycles are dramatically increasing for the great popularity gained by the freeze-all, the best strategy for minimizing OHSS. Endometrial preparations for FET include natural cycles (NC), modified NC or, more frequently, artificial cycles in which the endometrium is prepared with exogenous steroids. While there is a general agreement on the criteria for deciding the optimal timing for scheduling FET (an endometrial thickness ≥7 mm with a trilaminar pattern), no consensus has so far been reached on the endogenous progesterone and oestradiol values useful for individualizing the luteal phase support and ensuring the best success rates. Study design, size, duration A meta-analysis, based on PubMed, CENTRAL, Scopus, Web of Science, ClinicalTrials.gov, Embase, was conducted to evaluate the optimal mid-luteal progesterone and estradiol levels correlating with the highest live birth rate. Following PICO, inclusion criteria were: Population: infertile patients undergoing FET. Intervention: groups or subgroups of patients with different values of serum progesterone and/or estradiol. Comparison: not applicable. Outcome: live birth rate (LBR). Study design: prospective and retrospective cohort studies. This meta-analysis was submitted to PROSPERO (ID644713). Participants/materials, setting, methods Electronic and manual search, up to December 2024, identified 147 studies. Two researches (A.M.,S.G.) reviewed 115 studies independently (after removing 32 duplicates), excluded 59 studies after the first (title-abstract) screening and 44 studies after the second (full-text) screening. 12 studies were selected. Meta-regression is presented in terms of rate (for the outcome) and estimation of moderation effects (for progesterone and estradiol) with 95%CI and p-value (significance level at = 0.05); heterogeneity was assessed using I² statistics. Main results and the role of chance Serum mid-luteal progesterone was reported in all the studies (8.445 patients) (Alsbjerg et al, 2024; Álvarez et al, 2021; Alyasin et al, 2021; Cédrin-Durnerin et al, 2019; Gaggiotti-Marre et al, 2020; Kofinas et al, 2015; Labarta et al, 2021; Liu et al, 2023; Maignien et al, 2022; Yovich et al, 2015; Zhao et al, 2024; Zhu et al, 2023), while estradiol in 10 studies (excluding Kofinas et al, 2015; Zhu et al, 2023). 31 data points for progesterone (range 6.3-55.9 ng/ml) and 25 for estradiol (range 142.0-871.7 pg/ml), were obtained and they were classified into 4 categories for progesterone (<10.0; 10.0-19.9; 20.0-29.9; ≥30.0) and into 3 categories for estradiol (<300.0; 300.0-499.9; ≥500.0). The pooled estimate of LBR was 43.7% (95%CI=0.389-0.484, I2=71.2%); serum progesterone showed a significant effect of moderation (F(df1=3, df2=28) = 98.528, p<0.001), indicating that the highest LBR (56.6%) was found for values comprised between 10.8-18.0 ng/ml (estimate=0.566, 95%CI=0.427-0.796; p<0.01). Results from the studies reporting estradiol values showed a LBR of 44.2% (95%CI=0.385-0.499, I2=66.6%); estradiol showed a significant effect of moderation (F(df1=3, df2=22)=80.223,p<0.001), with the highest LBR (51.8%) for values comprised between 350.0-490.3 pg/ml (estimate=0.518, 95%CI=0.376-0.664; p < 0.01). The results were controlled for publication bias and effects of potential variables affecting the outcome (female’s age, endometrial preparation, endometrial thickness, number and stage of transferred embryos). Limitations, reasons for caution Although the majority of the included studies were prospective, 5 studies were retrospective; this could represent a reason for caution. The level of heterogeneity among the studies was quite high. Different routes of progesterone administration were considered. Only one study analysed the NC for the endometrial preparation (Gaggiotti-Marre et al,2020). Wider implications of the findings This meta-analysis demonstrated that, in FET cycles, the LBR can be reduced both for values below and above the identified range of mid-luteal progesterone and estradiol levels. For this reason, it is important to dose the mid-luteal hormones in FET cycles in order to individualize the luteal phase support. Trial registration number No