PURPOSE OF REVIEW:Progesterone plays a pivotal role in implantation and ongoing pregnancy. In the context of assisted reproductive technology (ART), and in particular frozen embryo transfer (FET), increasing evidence suggests that serum progesterone levels strongly influence reproductive outcomes. This review summarizes the latest findings regarding the optimal range of serum progesterone concentrations and their implications across different endometrial preparation protocols. RECENT FINDINGS:Observational and interventional studies have consistently shown that suboptimal mid-luteal serum progesterone levels are associated with lower implantation and live birth rates in FET cycles. The need for exogenous supplementation varies according to the type of endometrial preparation, with hormone replacement therapy cycles being particularly sensitive to inadequate progesterone exposure. Emerging data support the concept of individualized luteal support, including serum-guided adjustments or alternative routes of administration in cases of suboptimal serum progesterone levels. SUMMARY:Adequate serum progesterone concentrations are critical for optimizing outcomes after FET. Evidence supports a paradigm shift from empirical supplementation to precision medicine approaches based on serum thresholds and patient characteristics. Future research should aim to define universally accepted cutoff values, clarify the role of endometrial versus systemic progesterone, and refine strategies for tailoring luteal phase support in ART.

This systematic review (PROSPERO CRD420251151402) examined the relationship between chronic endometritis and recurrent implantation failure, focusing on prevalence, diagnostic strategies, and treatment-related reproductive outcomes. A comprehensive search of MEDLINE, Embase, Global Health, Cochrane Library, Web of Science, Health Technology Assessment Database, and ClinicalTrials.gov was performed. 24 eligible studies were identified, including women exclusively with recurrent implantation failure evaluated for chronic endometritis using CD138 plasma cell immunohistochemistry and/or hysteroscopy. Due to substantial heterogeneity in diagnostic criteria and outcome reporting, the findings were synthesized narratively. The reported prevalence varied widely across studies (7.7%–63.8%) and largely reflected differences in the CD138 plasma cell thresholds applied. Antibiotic therapy achieved histological resolution in most cases; however, whether this translates into higher clinical pregnancy or live birth rates compared to control populations remains uncertain, as the available studies reported mixed and inconsistent findings. The evidence regarding combined therapeutic strategies was similarly limited and heterogeneous, precluding firm conclusions regarding their effectiveness. Overall, the certainty of evidence was low. Standardized diagnostic criteria and adequately powered prospective studies are required to clarify the clinical relevance of screening and treatment for chronic endometritis in this population.

STUDY QUESTION:Can natural cycles (NC) be effectively utilized in advanced maternal age (AMA) undergoing oocyte donation, without compromising live birth rates (LBRs) and miscarriage outcomes, when compared to artificial cycles (ACs)? SUMMARY ANSWER:In donor oocyte embryo transfer cycles, NC demonstrated superior outcomes in reproductive efficacy and obstetrical safety compared to AC, independent of the recipient's age. WHAT IS KNOWN ALREADY:Previous studies have posited that NC may result in better outcomes when compared to AC embryo transfer, including a lower risk of miscarriage and hypertensive disorders of pregnancy. Recent studies support that NC-frozen embryo transfer (FET) decreases obstetrical and neonatal complications compared to AC-FET, even if LBR differences remain controversial in some general populations. There is limited research on the use of NCs in women of AMA. STUDY DESIGN, SIZE, DURATION:This retrospective, multicentre, cohort study included all single blastocyst embryo transfers following oocyte donation performed between January 2010 and December 2023, subdivided according to the type of endometrial preparation performed (NC or AC). The oocyte donation model was chosen to minimize the potential confounding effect related to poor oocyte competence in older women and the influence of ovarian stimulation performed during autologous IVF on endometrial receptivity prior to a fresh embryo transfer. PARTICIPANTS/MATERIALS, SETTING, METHODS:The main objective of the study was to compare LBR. Secondary outcomes included hCG-positive pregnancy rate, clinical pregnancy rate, miscarriage rate, obstetric, and perinatal outcomes. Confounder-adjustment was performed using a multivariable generalized estimating equations model regression analysis, adjusting for multiple confounders. A sub-analysis compared results when the AC protocol was optimized with progesterone (P4) monitoring and rescue therapy. Additionally, an interaction variable was added to the final multivariable model to assess whether female recipient age may modify the effect of each type of endometrial preparation on LBRs. MAIN RESULTS AND THE ROLE OF CHANCE:In total, 67 048 embryo transfers were analysed, including NC (n = 6922) and AC (n = 60 126). The NC group demonstrated consistent superiority over AC after adjustment for confounders across all transfers. NC was associated with a higher LBR (aOR 1.38, 95% CI 1.29-1.47; P < 0.01) and significantly lower miscarriage rate per hCG-positive pregnancy (aOR 0.68, 95% CI 0.61-0.76; P < 0.01). This superiority persisted even in optimized AC cycles with P4 monitoring and rescue therapy (LBR aOR 1.42, 95% CI 1.31-1.54; P < 0.01). Furthermore, NC was associated with significantly lower obstetrical risks in singleton pregnancies, including hypertensive disorders of pregnancy (aOR 0.72, 95% CI 0.56-0.94; P = 0.01), Caesarean delivery (aOR 0.86, 95% CI 0.77-0.96; P < 0.01), and large for gestational age (aOR 0.77, 95% CI 0.67-0.89; P < 0.01). The interaction between endometrial preparation method and female recipient age was not statistically significant (aOR 1.02, 95% CI 0.99-1.03). LIMITATIONS, REASONS FOR CAUTION:The retrospective nature of the study and the inherent risk of bias related to unmeasured confounding factors may have impacted the results. Another limitation is the low percentage of NC included in the study (10.32% of all cycles), which could be related to the low uptake to this treatment modality in real-life practice. WIDER IMPLICATIONS OF THE FINDINGS:NC may offer superior reproductive outcomes and is associated with lower obstetrical risks, with differences unlikely to be modified by female age. Therefore, it seems reasonable to suggest NC for older women, as they could benefit from the decreased risk of miscarriage and hypertension during pregnancy. STUDY FUNDING/COMPETING INTEREST(S):No specific funding was obtained for this study. A.R.N. has received research grants (to institution) from Theramex; Consulting and Speakers' fees and travel support from Organon and Merck KgaA; S.S.-R. has received consulting fees from Organon, IBSA, and Besins; Speakers' fees and travel support from Organon, Ferring Pharmaceuticals, Theramex, IBSA, Gedeon Richter, Abbott, and Besins. He has also received travel support from Organon, Ferring, Theramex, IBSA, Gedeon-Richter, Abbott, and Besins. He holds stocks/shares with IVIRMA Lisboa. He is a member of the ESHRE Executive Committee and was the Senior Deputy of Safety and Quality for ESHRE. TRIAL REGISTRATION NUMBER:N/A.

The route of administration significantly influences the pharmacokinetics of progestogens. For luteal phase support (LPS) after fresh embryo transfer in IVF, ESHRE recommends vaginal, intramuscular, subcutaneous progesterone or oral dydrogesterone. This network meta-analysis evaluated the relative effectiveness of progestogen administration routes on clinical pregnancy rate (CPR) and live birth rate (LBR) in fresh IVF cycles. A systematic review identified peer-reviewed, published randomized controlled trials (RCTs) comparing individual progestogens (versus placebo or other progestogens) and reporting CPR. Studies involving frozen-thawed transfers, non-progestogenic LPS or non-available formulations were excluded. Of the 24 RCTs included, oral administration was the only route reaching statistical significance for increased CPR, although overlapping confidence intervals across comparisons indicate uncertainty in clinically relevant superiority. For LBR, only oral and intramuscular routes demonstrated statistically significant improvement versus placebo; vaginal and subcutaneous did not. These findings, supported by sensitivity analyses excluding low-quality and older studies, suggest that oral and intramuscular progestogens may be more effective in improving IVF outcomes, but outcome differences require further studies. This study substantially advances the evidence base by integrating recent data, applying rigorous quality assessment standards and using state-of-the-art network meta-analytic methodology, contributing to refining recommendations for optimal progestogenic LPS in fresh IVF cycles.

Is there a relationship between serum estradiol (E2) levels on the day of embryo transfer and ongoing pregnancy rate (OPR) in artificial endometrium preparation cycles? Patients with serum E2 >256.9 pg/ml the day of embryo transfer (ET) had significantly lower ongoing pregnancy rates compared to patients with serum E2:190.6-256.9 pg/ml. Estradiol and progesterone are key hormones in endometrial preparation for ET in artificial cycles. While it is well-established that progesterone levels on the day of ET are crucial for optimizing pregnancy outcomes, the role of E2 levels on the day of ET remains unclear. Currently, there is only limited evidence suggesting an optimal E2 range for favorable reproductive outcomes; therefore, further research is needed to address this topic. Post hoc analysis of two prospective cohort studies including 1098 embryo transfers after an artificial endometrial preparation cycle with estradiol valerianate and vaginal micronized progesterone (400 mg/12 hours). Study was conducted between February 2016 and November 2018. Patients undergoing ET (ET), aged <50, BMI <30 Kg/m2, with a normal uterine cavity, a triple layer endometrium >6.5 mm, and being transferred 1-2 good quality blastocysts; in a private infertility centre. Serum E2 determination was performed the day of ET and results were blinded during study period. Primary endpoint was ongoing pregnancy rate beyond the 12th week of pregnancy. A total of 1098 cycles were analysed. The mean age of the included women was 39.7±4.6; BMI: 23.6±4.1; Endometrial thickness: 8.8±1.6 mm. Serum E2 the day of ET was 205.7±87.4 pg/ml (p25: 144.1; p50:190.6; p75:256.9). The ongoing pregnancy rates according to serum E2 levels were: p75: 50.6% (p = 0.106). Taking p50-p75 (Q2 to Q3) as a reference considering that the mean E2 levels were within this range, women with serum E2 >256.9 pg/mL (>p75) had a significantly lower OPR compared to women in p50-p75 (50.6% vs. 60.2%; p = 0.016; OR (95% CI): 0.90 (0.83-0.98), after adjusting for confounding factors. The LBR according to serum E2 levels were: p75:43.6% (p = 0.152). The multivariate logistic regression showed that women with serum E2 >256.9 pg/mL (>p75) had a significantly lower LBR compared to women in p50-p75 (p = 0.031; OR (95% CI): 0.91 (0.84-0.99), after adjusting for confounding factors. The Youden index estimated that the optimal E2 level with the highest sensitivity and specificity for having a live birth was 193.5 pg/ml. There were not statistical differences in LBR below or above this threshold (49.1% vs 46.4%, p = 0.41). This is a post hoc analysis, and sample size was not calculated for this purpose. Only women with a normal uterine cavity, an appropriate endometrial thickness and good quality blastocysts transfer were included. Extrapolation to an unselected population needs to be validated. The present study suggests that there is a critical upper threshold of serum E2 values the day of embryo transfer in artificial endometrial preparation cycles. A lower threshold could not be defined. No

RESEARCH QUESTION:Does mid-luteal serum progesterone concentration influence pregnancy outcomes in modified natural cycle frozen embryo transfer (mNC-FET) with ovulation triggering and luteal phase support (LPS)? DESIGN:This prospective, blinded, single-centre cohort study was conducted at IVIRMA Valencia (Spain) between February 2020 and June 2021. A total of 241 infertile patients under 50 years old, with body mass index (BMI) <40 kg/m2, undergoing mNC-FET were included. Ovulation was triggered with recombinant human chorionic gonadotrophin when the dominant follicle measured ≥16 mm and endometrial thickness was >6.5 mm. LPS was provided with micronized vaginal progesterone 200 mg every 12 h. Serum progesterone was measured within 2 h of embryo transfer, and results were blinded until study completion. RESULTS:Mean serum progesterone on embryo transfer day was 26.2 ± 9.0 ng/ml, with 99% of patients exceeding the clinically relevant threshold (>10 ng/ml). Participants were stratified into quartiles by progesterone concentration, with no significant differences in ongoing pregnancy rates: Q1 (<20.2 ng/ml), 56.7%; Q2 (20.2-24.8 ng/ml), 47.5%; Q3 (24.9-31.1 ng/ml), 51.7%; Q4 (>31.1 ng/ml), 48.3% (P = 0.74). Live birth rates were also comparable (P = 0.66). In multivariable logistic regression, mid-luteal serum progesterone was not significantly associated with ongoing pregnancy (odds ratio 0.86, 95% CI 0.40-1.85; P = 0.70) after adjusting for clinical covariates. Female age, oocyte age, BMI, number of previous transfers, embryos transferred, embryo quality, and oocyte origin were also unrelated to pregnancy probability. CONCLUSIONS:In mNC-FET cycles with HCG triggering and LPS, mid-luteal serum progesterone concentrations were consistently adequate and not associated with pregnancy outcomes. Routine progesterone monitoring appears unnecessary in this setting.

Can quantitative PCR (qPCR) serve as a reliable, cost-efficient alternative to sequencing-based methods for accurately analysing microbial composition and relative abundance in the vagina? qPCR outperformed sequencing in species detection and quantification, demonstrating high accuracy and providing a reliable alternative for vaginal microbiome analysis at a fraction of cost. The vaginal microbiome is increasingly recognised as a key factor in reproductive health, with emerging research linking its imbalance to ART failure and adverse pregnancy outcomes. This has driven efforts to characterise its composition and role in fertility. 16S rRNA and whole-genome sequencing are the gold standard for microbiome analysis but are very costly, labour-intensive, and require extensive computational analysis, limiting feasibility for large-scale studies and clinical application. As only a limited number of species are known to inhabit the vagina, sequencing may not always be necessary, and qPCR may serve as a cost-effective and rapid alternative for microbiome analysis. First, a qPCR assay was developed for vaginal microbiome profiling, with target species selected through literature review. Accuracy in species detection and quantification was assessed using 10 artificial genomic DNA mixtures (3-6 species per mixture; mean: 3.9) with known bacterial DNA proportions (1–88%), representing 39 individual species measurements. qPCR results were compared with 16S rRNA and whole-genome sequencing. Subsequently, all three methods were applied to vaginal swabs from 56 ART patients, to compare their performance. qPCR employed specific assays targeting 22 bacterial species, including commensal Lactobacillus spp. and dysbiosis-associated species linked to bacterial vaginosis and aerobic vaginitis. Additionally, four pan-genus assays covering the most common vaginal genera (Lactobacillus, Bifidobacterium, Streptococcus and Gardnerella) ensured a comprehensive microbiome assessment. 16S rRNA sequencing analysed all nine hypervariable regions, unlike conventional approaches targeting selected regions, enabling a comparison of sequencing variability. Whole-genome sequencing captured entire microbial genomes, allowing for high-resolution species identification and quantification. For artificial gDNA mixtures, qPCR demonstrated high accuracy in species detection and quantification. Results closely matched theoretical values, showing a strong correlation (r = 0.98, p < 0.0001) and minimal deviation (2.1%±2.0%) across 39 species measurements, confirming its reliability in reflecting known bacterial DNA proportions.16S rRNA sequencing showed lower accuracy in species-level detection, with relative quantification also being unreliable and varying widely across hypervariable regions (r = 0.43-0.79, p < 0.0001). It frequently misclassified both beneficial Lactobacillus spp. and dysbiotic species. Additionally, species resolution depended on the hypervariable region analysed, often failing to distinguish closely related species and sometimes lacking genus-level classification. On average, 23.0%±9.1% of sequencing output (range: 14.2-34.6%) was either misclassified or unassignable at the species level. Whole-genome sequencing was more reliable than 16S rRNA sequencing for species identification, providing improved species-level resolution. However, 23.4%±22.9% of sequencing reads remained unassigned at species level. Additionally, species quantification accuracy was only moderate (r = 0.53, p < 0.0001).In vaginal swabs from ART patients, 16S sequencing showed substantial variability between hypervariable regions, particularly at the species level. Such misclassification is problematic, as bacterial species, not just genera, impact vaginal health and ART success. qPCR and whole-genome sequencing showed greater concordance, with whole-genome sequencing detecting few additional species, though not at clinically relevant proportions. qPCR can only detect preselected species, potentially missing novel or unexpected microbes that sequencing might identify. Additionally, its targeted nature may overlook low-abundance species. However, given the well-defined composition of the vaginal microbiome, these limitations are unlikely to significantly affect its effectiveness for clinical and research use. Besides its accuracy in vaginal microbiome analysis, qPCR costs three to four times less than sequencing and avoids complex bioinformatics and high-performance computing. With a 4-hour processing time versus 40–50 hours for sequencing, its cost-effectiveness and speed make it ideal for large-scale vaginal microbiome studies and clinical use. No

INTRODUCTION:To determine for the first time the incidence of preeclampsia and other pregnancy complications according to the mode of endometrial preparation for frozen embryo transfer (FET) in a randomized fashion. METHODS:Women about to undergo FET of a single euploid blastocyst were randomly assigned to a modified natural cycle (MNC) or an artificial cycle (AC). Inclusion criteria were as follows: Caucasian; non-obese; 18-43 years of age; nulliparity; regular menstrual cycles; and autologous oocytes. Exclusion criteria were as follows: uterine alterations; moderate-heavy smokers; gamete donation; and chronic diseases. A pilot sub-study of first-trimester markers of preeclampsia was performed in 60 of the patients. RESULTS:Of the 1260 patients estimated, 591 met the inclusion criteria and were willing to participate; of these, 306 and 285 were randomly assigned to a MNC or AC, respectively. After exclusion, 242 and 227 patients finally underwent a MNC or AC, resulting in 131 and 103 clinical pregnancies, 121 and 92 live births, and 91 and 70 questionnaires obtained concerning pregnancy complications, respectively. The incidence of preeclampsia was double in the AC group (10.00 % versus 4.39 %), though not significantly different. In the AC group, the prevalence of first trimester bleeding was significantly higher (42.85 % versus 15.38 %), and there was a significant reduction in live birth rates (40.53 % versus 50.00 %) in the per protocol analysis. No differences in early markers of preeclampsia were detected between the two groups. DISCUSSION:Although this randomized study was truncated, our findings are in accordance with previous reports of a higher risk of preeclampsia and other complications when the endometrium is artificially prepared for FET.

Infertility affects 15% of the population in developed countries, and its prevalence is increasing. Fertility can be influenced by different factors. Although key factors like maternal age cannot be changed, there is growing evidence that other modifiable factors, such as diet, can have an impact on fertility. Diet has become increasingly important in recent years for a number of reasons: the new trend toward a healthy lifestyle, the higher prevalence of certain digestive disorders, a lack of time that leads people to consume more prepared and processed food, and personal choice to not eat meat, among others. To meet these needs, several diets have recently become popular, such as the Mediterranean diet, known as the gold standard of health; the DASH diet, known for preventing hypertension; the Western diet, characterized by processed food; the ketogenic diet, characterized by low carbohydrate intake; and the vegetarian diet, which is the choice for people who do not eat meat or animal by-products. Diets present a unique composition characterized by the presence or absence of specific nutrients, which have also been associated with male and female fertility individually. This review assesses the impact of these diets and of macro- and micronutrients on both female and male fertility.

To elucidate if morning or evening start of exogenous progesterone (P4) five days before blastocyst embryo transfer (ET) impacts ongoing pregnancy rates (OPR) in artificial cycles. Single-centre retrospective cohort study of 6493 artificial cycles for an ET (oestrogens and luteal phase support [LPS] with micronized vaginal progesterone [MVP] 400 mg/12 h), conducted at an infertility clinic, December 2018—July 2022. LPS was given from five days before ET. Until March 2021, LPS was started in the evening of day 0 of P4 exposure (“evening start”); since April 2021, LPS was started in the morning of day 0 (“morning start”). Morning start of LPS (n = 2482 cycles); evening start (n = 3983 cycles). Morning or evening start did not exert any impact in OPR (46.9

INTRODUCTION:A pilot study was carried out to test the efficacy of the autologous mitochondrial transfer therapy (AUGMENT) technique. No improvements in pregnancy rate, development, or embryo quality were observed in the AUGMENT-treated group versus the Control group in this study. The main objective of this research is to analyze whether AUGMENT technology did have any impact on the obstetric and perinatal outcomes of pregnancies and children resulting from treated oocytes. METHODS:Follow up study of women with a livebirth who participated in a pilot randomized controlled trial in which sibling MII oocytes were randomly allocated to AUGMENT + intracytoplasmic sperm injection (ICSI) (AUGMENT group) or ICSI alone (control group). Preimplantation genetic testing for aneuploidy was performed in both groups. Pregnancy and neonatal outcomes of 14 women (15 pregnancies) and their 18 children were analyzed. The information was retrieved by reviewing the medical records or through questionnaires sent to the patients. RESULTS:No differences were found in this small case series between the AUGMENT and control groups regarding the rate of gestational complications, birth defects, gestational age at delivery (271.4 ± 12.56 vs 278 ± 10.4 days), birthweight (3.1 ± 0.6 kg vs. 3.1 ± 0.4 kg) and neonatal outcome. DISCUSSION:The few pregnancies achieved using AUGMENT oocyte therapy had similar outcomes than controls in this very small series. Our very preliminary data need to be confirmed in larger samples. The long term follow up of these children also needs to be analyzed.

Objective: To analyze if partial premature ovulation (PPO) detection during oocyte pick-up (OPU) impairs the quality of the retrieved oocyte cohort. Methods: The PPO concept refers to the situation when premature ovulation happens only in some of the follicles and it is detected during OPU. This study constitutes a retrospective analysis performed in an infertility clinic (Spain) during 2016-2021 with patients undergoing OPU after controlled ovarian hyperstimulation for an in vitro fertilization (IVF) treatment. Study code: 2110-VLC-091VG, registered on December 9 2021. Data from women with PPO (n=111) were compared to a matched control sample of cycles without PPO (n=333) at a proportion of 1:3. Results: Cycles were matched for age, body mass index (BMI), treatment year, embryo genetic analysis and stimulation protocol type. The mean numbers of oocytes (6.1 vs. 11.2), mature oocytes (4.7 vs. 8.8), correctly fertilized oocytes (3.6 vs. 6.6) and top-quality blastocysts (0.9 vs. 1.8) were significantly lower in the PPO group than the nonPPO group (p<0.05). However, maturation, fertilization, top-quality blastocyst and pregnancy rates were statistically comparable among groups (p>0.05). Conclusions: Cycles with PPO have fewer available oocytes and, thus, fewer available embryos for transfer, al though their quality is intact, and still offer chances of pregnancy in these cases. Hence cycle cancellation may not be worth associated money, time and morale losses once PPO is detected.

Research question: Is there a difference between the proportion of patients with serum progesterone <8.8 ng/ml on the day of embryo transfer when micronized vaginal progesterone (MVP) for luteal phase support (LPS) is given as pessaries versus capsules? Design: This retrospective, matched -cohort, single -centre study compared pessaries (Cyclogest) versus capsules (Utrogestan, Progeffik) for LPS in hormone replacement treatment -embryo transfer (HRT-ET) cycles. Patients under 50 years old with a triple -layer endometrial thickness of >6.5 mm underwent transfer of one or two blastocysts. Serum progesterone concentrations were measured on the day of transfer; patients with concentrations <8.8 ng/ml received a single 'rescue' dose of additional progesterone by subcutaneous injection. Results: In total 2665 HRT-ET cycles were analysed; 663 (24.9%) used pessaries for LPS and 2002 (75.1%) used capsules. Mean serum progesterone concentrations with standard deviations on the day of embryo transfer were significantly higher in the group using MVP pessaries compared with those using capsules (14.5 +/- 5.1 versus 13.0 +/- 4.8 ng/ml; P = 0.000). The percentage of participants with suboptimal serum progesterone concentrations on the day of embryo transfer (<8.8 ng/ml) was significantly lower in the pessary group than the capsule group (10.3%, 95% confidence interval [CI] 7.9-12.6% versus 17.9%, 95% CI 16.2-19.6%; adjusted odds ratio 0.426, 95% CI 0.290-0.625; P = 0.000). No differences in pregnancy outcome were observed between the groups. Conclusions: Using MVP pessaries rather than capsules for LPS resulted in significantly fewer patients having suboptimal serum progesterone concentrations on the day of embryo transfer. Consequently, almost 50% fewer patients in the pessary group needed rescue treatment.

Does vaginal microbiome have an impact on vaginal absorption of exogenous progesterone (P intake) in artificial cycles? Results suggest that the presence of lactobacillus in the vagina may positively impact the uptake of P. Low serum P levels on ET day decrease significantly live birth rates. About 20% of patients receiving micronized vaginal P show inadequate levels. It is of interest to find out which intrinsic factors might influence on the capacity of absorption (P intake) of vaginal P, and if they could be treated to prevent this situation. Vaginal microbiome hast been suggested as a possible factor, although this has not been addressed yet. The aim of this study was to explore if there is a correlation between vaginal microbiome and pH, and P intake. A prospective single-centre cohort pilot study including 92 ART patients was conducted from February 2022 to January 2023 in IVI-RMA Valencia. Embryo transfer was conducted in the context of a hormonal replacement therapy cycle with use of vaginal P (400mg/12h) for luteal phase support. “Progesterone intake” capacity was evaluated according to the vaginal microbiome status (lactobacillus dominant (LDM, >90%) or non-lactobacillus dominant (NLDM) and pH levels determinations (considered normal when moderately acidic ( < =4.8). Samples were taken twice: A) proliferative phase (day of initiation of exogenous P) and B) mid-luteal phase (day of ET, P + 5). Serum E2 and P levels as well as vaginal microbiome and pH were analysed. A validated qPCR methodology was used to accurately assess the presence and relative abundance of the four major lactobacillus species associated with eubiosis (crispatus, gasseri, iners, and jensenii) and the most common bacterial species implicated in dysbiosis. Distribution of LDM and NLDM profiles were comparable between Samples A and B. No significant difference between any of the potential confounding factors (age, BMI,baseline P) between the lactobacillus CST groups (CST-I,-II,-III,-V & lactobacillus) and the dysbiotic CST group (CST-IV) in Samples A and B were observed. Serum P levels on the ET day were significantly higher in LDM profiles both in Sample A (15.2 vs. 12.9ng/mL in NLDM;p=0.009) and B (15.0 vs. 12.8ng/mL;p=0.014). A positive significant correlation was found between serum P levels and Lactobacillus abundance in Samples A (r = 0.28; p = 0.008) and B (r = 0.30; p = 0.004). The relative abundance of Gardnerella vaginalis in Sample A was negatively related to serum P levels on the ET day (r=-0.24;p=0.02). Vaginal pH was significantly lower in LDM profiles both in Sample A (4.9 vs. 5.6 in NLDM; p = 0.002) and B (4.7 vs. 5.1;p<0.001). pH levels were negatively correlated with Lactobacillus abundance in Samples A (r=-0.34;p=0.0012) and B (r=-0.41;p<0.001). Additionally, serum P levels were negatively correlated with pH levels in Sample A (r=-0.14;p=0.191) and B (r=-0.27; p = 0.010). Indeed, when patients are grouped based on a cut-off of P = 8.8ng/mL, patients with a pH ≤ 4.8 have significantly higher P levels than patients with a pH > 4.8. This is a pilot study, thus further trials with larger sample sizes should be performed in order to confirm these results. Additionally, vaginal microbiome alone does not explain the total variability in serum P levels measured in artificial cycles when using MVP. We might have found one of the causes of the high heterogeneity on vaginal absorption of MVP in artificial cycles. This finding may help us with patient management when programming an ET in this type of cycles, improving personalised patient care and progressing on luteal phase support individualisation. 2110-VLC-095-EL

OBJECTIVE:To find a useful tool for estimating the minimum number of metaphase II (MII) oocytes needed to obtain at least one euploid blastocyst according to female age. DESIGN:Retrospective analysis of in vitro fertilization (IVF) treatment cycles with preimplantational genetic testing for aneuploidies (PGT-A) performed over 5 years in IVIRMA Valencia (Spain), January 2017-March 2022. Approval from the Institutional Review Board of IVI Valencia (2204-VLC-040-CR). SETTING:Private infertility clinic in Spain. PATIENTS:Eligible patients were undergoing their first IVF-PGT-A treatment cycle, in which at least one MII oocyte was obtained, regardless of oocyte and semen origin. Oocyte donation cycles were included in the donor group (≤34 years old). Treatment cycles from women with their own oocytes were selected only when the oocytes were aged ≥35 years (patient group). Only trophoectoderm biopsies performed on days 5 or 6 of development and analyzed using next-generation sequencing were included. Preimplantational genetic testing for aneuploidy cycles because of a known abnormal karyotype were excluded. INTERVENTION:Not applicable. MAIN OUTCOME MEASURES:Number of MII oocytes needed to obtain one euploid blastocyst according to female age. RESULTS:A total of 2,660 IVF-PGT-A treatment cycles were performed in the study period in the eligible population (patients group = 2,462; donors group =198). The mean number of MII oocytes needed to obtain one euploid blastocyst increased with age, as did the number of treatment cycles that did not get at least one euploid blastocyst. An adjusted multivariate binary regression model was designed using 80% of the patient group sample (n = 2,462; training set). A calculator for the probability of obtaining at least one euploid blastocyst was created using this model. The validation of this model in the remaining 20% of the patient group sample (n = 493; validation set) showed that it could estimate the event of having at least one euploid blastocyst with an accuracy of 72.0%. CONCLUSIONS:Our results show a preliminary model capable of predicting the number of MII oocytes needed to obtain at least one euploid blastocyst according to female age, calculated with the largest database of IVF-PGT-A treatment cycles ever used for this purpose, including only treatment cycles using next-generation sequencing on trophoectoderm biopsies. Once this model has been properly validated, it could help with decision-making for both clinicians and patients coming to an infertility clinic.

Abstract Study question Does high serum progesterone (P) level on the day of FET after hormone replacement therapy affect IVF outcomes? Summary answer The supraphysiological serum progesterone levels on the day of FET after HRT do not impair reproductive outcomes. What is known already A proper synchronization between embryo and the endometrium is still required by controlling the timing and the dosage of exposure to exogenous hormones. Previous observational studies have highlighted the negative effects of serum hormone levels at the minimum threshold during FET cycles. However, based on existing research, the relationship between progesterone levels on FET day and pregnancy outcomes as well as the certain dose of progesterone ideally to achieve pregnancy and a live birth are debatable topics Study design, size, duration This is an observational, retrospective and cohort study including 7539 FET cycles that were performed under artificial endometrial preparation with HRT using exogenous estradiol and progesterone (vaginal, subcutaneous and intramuscular) between January 2017 and December 2022. Both euploid FET from autologous oocyte ICSI-PGT-A cycles (n = 1822) or FET from ICSI cycles using donated oocytes (n = 5724) were considered. Participants/materials, setting, methods Women were divided into three groups according to P4 level on the day of FET:P<20 ng/ml (n = 6623),P≥20-40 ng/ml (n = 770) and P > 40 ng/ml (n = 146).All ET were performed at the blastocyst stage.The primary outcome was live birth rate (LBR). Secondary outcomes evaluated were biochemical, clinical, and ongoing pregnancy, and miscarriage rate (MR) calculated per FET.Categorical variables were compared between groups with Fisher’s exact test. Logistic regression models adjusted (AOR) with several confounders. Main results and the role of chance There were no significant differences in baseline characteristics of the study population. The mean of P4 level was 15,66 ng/ml. Female age was P < 20:38.7 ±3.1, P ≥ 20-40: 39.0 ±3.0, and P > 40: 38.1 ±2.9 (p = 0.53) years old. The group with the highest progesterone levels always found lower biochemical pregnancy even no statistically significant [64.6% (P < 20) vs 63.7% (P ≥ 20-<40) vs 61.1% (P > 40); p = 0.62]; in clinical pregnancy [56.3% (P < 20) vs 56.8% (P ≥ 20-<40) vs 51.4% (P > 40); p = 0.5] and in ongoing [47.0% (P < 20) vs 47.0% (P ≥ 20-<40) vs 41.8% (P > 40); p = 0.47] pregnancy rates. The MR was also higher in the group with higher progesterone, although not significantly: 16.2% (P < 20) vs 15.0% (P ≥ 20-<40) vs 18.0% (P > 40) (p = 0.60). No statistically significant differences were reported for live birth rate, 41.8% (P < 20) vs 40.3% (P ≥ 20-<40) vs 34.2% (P > 40) (p = 0.2). The multivariate analysis showed an increased risk of miscarriage when P > 40ng/ml, aOR= 1.14 (0.75-1.74) (p = 0.55) whereas no such association was found with P4 ≥20-40 ng/ml, aOR=0.92 (0.74-1.15) (p = 0.46). Additionally, the probability of live birth is similar when the patient was P > 40 (aOR= 0.77 (0.51-1,15) (p = 0.20)) than when P ≥ 20-<40 (aOR=0.94 (0.79-1,11) (p = 0.47) compared with the references group. Limitations, reasons for caution The retrospective design and different female factors for IVF were included in the study leading to a possible biased population although the adjusted analysis. Different routes of progesterone administration were considered in the study. Currently, we don’t report the timing between the last P4 administration and dosage on the FET. Wider implications of the findings This information is useful for clinicians who monitor progesterone levels prior to FET to avoid reaching high levels that would undermine the chances of achieving a newborn. Trial registration number Not Applicable