OBJECTIVE: To explore the optimal models for predicting the formation of high-quality embryos in Poor Ovarian Response (POR) Patients with Progestin-Primed Ovarian Stimulation (PPOS) using machine learning algorithms.
METHODS: A retrospective analysis was conducted on the clinical data of 4,216 POR cycles who underwent in vitro fertilization (IVF) / intracytoplasmic sperm injection (ICSI) at Sichuan Jinxin Xinan Women and Children\'s Hospital from January 2015 to December 2021. Based on the presence of high-quality cleavage embryos 72 h post-fertilization, the samples were divided into the high-quality cleavage embryo group (N = 1950) and the non-high-quality cleavage embryo group (N = 2266). Additionally, based on whether high-quality blastocysts were observed following full blastocyst culture, the samples were categorized into the high-quality blastocyst group (N = 124) and the non-high-quality blastocyst group (N = 1800). The factors influencing the formation of high-quality embryos were analyzed using logistic regression. The predictive models based on machine learning methods were constructed and evaluated accordingly.
RESULTS: Differential analysis revealed that there are statistically significant differences in 14 factors between high-quality and non-high-quality cleavage embryos. Logistic regression analysis identified 14 factors as influential in forming high-quality cleavage embryos. In models excluding three variables (retrieved oocytes, MII oocytes, and 2PN fertilized oocytes), the XGBoost model performed slightly better (AUC = 0.672, 95% CI = 0.636-0.708). Conversely, in models including these three variables, the Random Forest model exhibited the best performance (AUC = 0.788, 95% CI = 0.759-0.818). In the analysis of high-quality blastocysts, significant differences were found in 17 factors. Logistic regression analysis indicated that 13 factors influence the formation of high-quality blastocysts. Including these variables in the predictive model, the XGBoost model showed the highest performance (AUC = 0.813, 95% CI = 0.741-0.884).
CONCLUSIONS: We developed a predictive model for the formation of high-quality embryos using machine learning methods for patients with POR undergoing treatment with the PPOS protocol. This model can help infertility patients better understand the likelihood of forming high-quality embryos following treatment and help clinicians better understand and predict treatment outcomes, thus facilitating more targeted and effective interventions.

UNASSIGNED: To determine whether endometrial thickness (EMT) differs between i) clomiphene citrate (CC) and gonadotropin (Gn) utilizing patients as their own controls, and ii) patients who conceived with CC and those who did not. Furthermore, to investigate the association between late-follicular EMT and pregnancy outcomes, in CC and Gn cycles.
UNASSIGNED: Retrospective study. Three sets of analyses were conducted separately for the purpose of this study. In analysis 1, we included all cycles from women who initially underwent CC/IUI (CC1, n=1252), followed by Gn/IUI (Gn1, n=1307), to compare EMT differences between CC/IUI and Gn/IUI, utilizing women as their own controls. In analysis 2, we included all CC/IUI cycles (CC2, n=686) from women who eventually conceived with CC during the same study period, to evaluate EMT differences between patients who conceived with CC (CC2) and those who did not (CC1). In analysis 3, pregnancy outcomes among different EMT quartiles were evaluated in CC/IUI and Gn/IUI cycles, separately, to investigate the potential association between EMT and pregnancy outcomes.
UNASSIGNED: In analysis 1, when CC1 was compared to Gn1 cycles, EMT was noted to be significantly thinner [Median (IQR): 6.8 (5.5-8.0) vs. 8.3 (7.0-10.0) mm, p<0.001]. Within-patient, CC1 compared to Gn1 EMT was on average 1.7mm thinner. Generalized linear mixed models, adjusted for confounders, revealed similar results (coefficient: 1.69, 95% CI: 1.52-1.85, CC1 as ref.). In analysis 2, CC1 was compared to CC2 EMT, the former being thinner both before [Median (IQR): 6.8 (5.5-8.0) vs. 7.2 (6.0-8.9) mm, p<0.001] and after adjustment (coefficient: 0.59, 95%CI: 0.34-0.85, CC1 as ref.). In analysis 3, clinical pregnancy rates (CPRs) and ongoing pregnancy rates (OPRs) improved as EMT quartiles increased (Q1 to Q4) among CC cycles (p<0.001, p<0.001, respectively), while no such trend was observed among Gn cycles (p=0.94, p=0.68, respectively). Generalized estimating equations models, adjusted for confounders, suggested that EMT was positively associated with CPR and OPR in CC cycles, but not in Gn cycles.
UNASSIGNED: Within-patient, CC generally resulted in thinner EMT compared to Gn. Thinner endometrium was associated with decreased OPR in CC cycles, while no such association was detected in Gn cycles.

UNASSIGNED: To explore the effects of insulin resistance (IR) on embryo quality and pregnancy outcomes in women with or without polycystic ovary syndrome (PCOS) undergoing in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI).
UNASSIGNED: A retrospective cohort study concerning patients with/without PCOS who received gonadotropin-releasing hormone (GnRH)-antagonist protocol for IVF/ICSI from January 2019 to July 2022 was conducted. All the patients included underwent oral glucose tolerance test plus the assessment of insulin release within 6 months before the controlled ovarian stimulation. The Matsuda Index was calculated to diagnose IR. Two populations (PCOS and non-PCOS) were included and each was divided into IR and non-IR groups and analyzed respectively. The primary outcome was the high-quality day 3 embryo rate.
UNASSIGNED: A total of 895 patients were included (751 with PCOS and 144 without PCOS). For patients with PCOS, the IR group had a lower high-quality day 3 embryo rate (36.8% vs. 39.7%, p=0.005) and available day 3 embryo rate (67.2% vs. 70.6%, p<0.001). For patients without PCOS, there was no significant difference between the IR and non-IR groups in high-quality day 3 embryo rate (p=0.414) and available day 3 embryo rate (p=0.560). There was no significant difference in blastocyst outcomes and pregnancy outcomes for both populations.
UNASSIGNED: Based on the diagnosis by the Matsuda Index, IR may adversely affect the day 3 embryo quality in patients with PCOS but not pregnancy outcomes. In women without PCOS, IR alone seems to have less significant adverse effects on embryo quality than in patients with PCOS. Better-designed studies are still needed to compare the differences statistically between PCOS and non-PCOS populations.

BACKGROUND: The low live birth rate and difficult decision-making of the in vitro fertilization (IVF) treatment regimen bring great trouble to patients and clinicians. Based on the retrospective clinical data of patients undergoing the IVF cycle, this study aims to establish classification models for predicting live birth outcome (LBO) with machine learning methods.
METHODS: The historical data of a total of 1405 patients undergoing IVF cycle were first collected and then analyzed by univariate and multivariate analysis. The statistically significant factors were identified and taken as input to build the artificial neural network (ANN) model and supporting vector machine (SVM) model for predicting the LBO. By comparing the model performance, the one with better results was selected as the final prediction model and applied in real clinical applications.
RESULTS: Univariate and multivariate analysis shows that 7 factors were closely related to the LBO (with P < 0.05): Age, ovarian sensitivity index (OSI), controlled ovarian stimulation (COS) treatment regimen, Gn starting dose, endometrial thickness on human chorionic gonadotrophin (HCG) day, Progesterone (P) value on HCG day, and embryo transfer strategy. By taking the 7 factors as input, the ANN-based and SVM-based LBO models were established, yielding good prediction performance. Compared with the ANN model, the SVM model performs much better and was selected as the final model for the LBO prediction. In real clinical applications, the proposed ANN-based LBO model can predict the LBO with good performance and recommend the embryo transfer strategy of potential good LBO.
CONCLUSIONS: The proposed model involving all essential IVF treatment factors can accurately predict LBO. It can provide objective and scientific assistance to clinicians for customizing the IVF treatment strategy like the embryo transfer strategy.

BACKGROUND: The utilization of a double trigger, involving the co-administration of gonadotropin-releasing hormone agonist (GnRH-a) and human chorionic gonadotropin (hCG) for final oocyte maturation, is emerging as a novel approach in gonadotropin-releasing hormone antagonist (GnRH-ant) protocols during controlled ovarian hyperstimulation (COH). This protocol involves administering GnRH-a and hCG 40 and 34 h prior to ovum pick-up (OPU), respectively. This treatment modality has been implemented in patients with low/poor oocytes yield. This study aimed to determine whether the double trigger could improve the number of top-quality embryos (TQEs) in patients with fewer than three TQEs.
METHODS: The stimulation characteristics of 35 in vitro fertilization (IVF) cycles were analyzed. These cycles were triggered by the combination of hCG and GnRHa (double trigger cycles) and compared to the same patients\' previous IVF attempt, which utilized the hCG trigger (hCG trigger control cycles). The analysis involved cases who were admitted to our reproductive center between January 2018 and December 2022. In the hCG trigger control cycles, all 35 patients had fewer than three TQEs.
RESULTS: Patients who received the double trigger cycles yielded a significantly higher number of 2PN cleavage embryos (3.54 ± 3.37 vs. 2.11 ± 2.15, P = 0.025), TQEs ( 2.23 ± 2.05 vs. 0.89 ± 0.99, P < 0.001), and a simultaneously higher proportion of the number of cleavage stage embryos (53.87% ± 31.38% vs. 39.80% ± 29.60%, P = 0.043), 2PN cleavage stage embryos (43.89% ± 33.01% vs. 27.22% ± 27.13%, P = 0.014), and TQEs (27.05% ± 26.26% vs. 14.19% ± 19.76%, P = 0.019) to the number of oocytes retrieved compared with the hCG trigger control cycles, respectively. The double trigger cycles achieved higher rates of cumulative clinical pregnancy (20.00% vs. 2.86%, P = 0.031), cumulative persistent pregnancy (14.29% vs. 0%, P < 0.001), and cumulative live birth (14.29% vs. 0%, P < 0.001) per stimulation cycle compared with the hCG trigger control cycles.
CONCLUSIONS: Co-administration of GnRH-agonist and hCG for final oocyte maturation, 40 and 34 h prior to OPU, respectively (double trigger) may be suggested as a valuable new regimen for treating patients with low TQE yield in previous hCG trigger IVF/intracytoplasmic sperm injection (ICSI) cycles.

The paper entitled \"Risk factors for poor ovarian response in patients receiving in-vitro fertilization and embryo transfer\" by Chen et al., which was published in Minerva Surgery 2023 June;78(3):303-4, has been retracted by the Publisher upon the authors\' request; they asked for a retraction because the paper contains faulty data.

UNASSIGNED: Studies on the effect of vaccine type and two other vaccines other than inactivated vaccines approved in China on in vitro fertilization (IVF) pregnancy outcomes are rare. To complement and confirm the existing findings, this research aimed to investigate whether there are adverse effects of different vaccine types in females and males on reproductive function and clinical pregnancy.
UNASSIGNED: This retrospective study enrolled 6,455 fresh embryo transfer cycles at the First Affiliated Hospital of Zhengzhou University between May 1, 2021, and October 31, 2022. The primary outcome is the clinical pregnancy rate (CPR). At the same time, the secondary results are the number of oocytes retrieved, two pronuclei (2PN) rate, blastocyst formation rate, high-quality blastocyst rate, and semen parameters (volume, density, sperm count, forward motility rate, total motility rate, immobility rate, and DNA fragment index (DFI) rate).
UNASSIGNED: In the comparison of ovarian stimulation indicators, no statistically significant differences (P > 0.05) were found in Gn days, endometrial thickness, 2PN rate, metaphase 2 (MII) rate, high-quality embryo rate, and blastocyst formation rate. No significant differences (P>0.05) were found in age, body mass index (BMI), education level, and semen parameters (volume, density, sperm count, forward motility rate, total motility rate, immobility rate, and DFI rate) in these four groups. The multivariate regression model showed that neither the types of vaccines nor the vaccination status of both infertile couples significantly affected clinical pregnancy.
UNASSIGNED: The type of vaccine does not appear to have an unfavorable effect on ovarian stimulation, embryo development, semen parameters, and clinical pregnancy.

UNASSIGNED: This study aims to determine whether the live birth rates were similar between GnRH antagonist original reference product Cetrotide® and generic Ferpront®, in gonadotropin-releasing hormone (GnRH) antagonist protocol for controlled ovarian stimulation (COS).
UNASSIGNED: This retrospective cohort study investigates COS cycles utilizing GnRH antagonist protocols. The research was conducted at a specialized reproductive medicine center within a tertiary care hospital, spanning the period from October 2019 to October 2021. Within this timeframe, a total of 924 cycles were administered utilizing the GnRH antagonist originator, Cetrotide® (Group A), whereas 1984 cycles were undertaken using the generic, Ferpront® (Group B).
UNASSIGNED: Ovarian reserve markers, including anti-Mullerian hormone, antral follicle number, and basal follicular stimulating hormone, were lower in Group A compared to Group B. Propensity score matching (PSM) was performed to balance these markers between the groups. After PSM, baseline clinical features were similar, except for a slightly longer infertile duration in Group A versus Group B (4.43 ± 2.92 years vs. 4.14 ± 2.84 years, P = 0.029). The duration of GnRH antagonist usage was slightly longer in Group B than in Group A (6.02 ± 1.41 vs. 5.71 ± 1.48 days, P < 0.001). Group B had a slightly lower number of retrieved oocytes compared to Group A (14.17 ± 7.30 vs. 14.96 ± 7.75, P = 0.024). However, comparable numbers of usable embryos on day 3 and good-quality embryos were found between the groups. Reproductive outcomes, including biochemical pregnancy loss, clinical pregnancy, miscarriage, and live birth rate, did not differ significantly between the groups. Multivariate logistic regression analyses suggested that the type of GnRH antagonist did not independently impact the number of oocytes retrieved, usable embryos, good-quality embryos, moderate to severe OHSS rate, clinical pregnancy, miscarriage, or live birth rate.
UNASSIGNED: The retrospective analysis revealed no clinically significant differences in reproductive outcomes between Cetrotide® and Ferpront® when used in women undergoing their first and second COS cycles utilizing the GnRH antagonist protocol.

UNASSIGNED: The main purpose of this study is to compare the embryo development and clinical outcomes of women in different age groups undergoing in vitro fertilization (IVF) processes using gonadotrophin-releasing hormone (GnRH) antagonist protocol, GnRH agonist long protocol, and early follicular phase protocol. We aim to provide reliable reference for future clinical treatments.
UNASSIGNED: We conducted a detailed analysis of patients who underwent treatment between January 2021 and February 2023. 1) In the overall patient population, we comprehensively compared the basic characteristics, the embryo development, and the clinical outcomes of patients treated with three different ovarian stimulation protocols, including the GnRH antagonist protocol group (n=4173), the agonist long protocol group (n=2410), and the early follicular phase long protocol group (n=341). 2) We divided the overall population into three age groups, one group for patients under 30 years old (n=2576), one for patients aged 30-35 (n=3249), and one for patients older than 35 years old (n=1099). Then, we compared the three stimulation protocols based on the group division. We separately compared the embryo development and clinical outcomes of patients using the three stimulation protocols in the under 30 years old, the 30-35 years old, and the over 35 years old age groups. With this analysis, we aimed to explore the response of different age groups to different stimulation protocols and their impact on the success rate of IVF.
UNASSIGNED: 1) In the overall population, we found that the average number of oocytes retrieved in the GnRH agonist long protocol group was significantly higher than that in the GnRH antagonist protocol group ([13.85±7.162] vs. [13.36±7.862], P=0.0224), as well as the early follicular phase long protocol group ([13.85±7.162] vs. [11.86±6.802], P<0.0001). Patients in the GnRH antagonist protocol group not only had a significantly lower starting dose of gonadotrophin (Gn) compared to the other two groups (P<0.05) but also had a significantly lower number of days of Gn use (P<0.05). The blastocyst formation rate in the GnRH antagonist protocol group was the highest among the three groups, significantly higher compared to the GnRH agonist long protocol group (64.91% vs. 62.35%, P<0.0001) and the early follicular phase long protocol group (64.91% vs. 61.18%, P=0.0001). However, there were no significant differences in the clinical pregnancy rates or the live birth rates among the three groups treated with different ovarian stimulation protocols (P>0.05). 2) In the <30 age group, the blastocyst formation rate in the GnRH antagonist protocol group was the highest among the three groups, significantly higher compared to the GnRH agonist long protocol group (66.12% vs. 63.33%, P<0.0001) and the early follicular phase long protocol group (66.12% vs. 62.13%, P=0.0094). In the 30-35 age group, the blastocyst formation rate in the GnRH antagonist protocol group was the highest among the three groups, significantly higher compared to the GnRH agonist long protocol group (64.88% vs. 62.93%, P=0.000 9) and the early follicular phase long protocol group (64.88% vs. 60.39%, P=0.0011). In the >35 age group, the blastocyst formation rate in the GnRH antagonist protocol group was significantly higher than that in the GnRH agonist long protocol group (59.83% vs. 56.51%, P=0.0093), while there was no significant difference compared to that of the early follicular phase long protocol group (P>0.05). In the three age groups, we found that there were no significant differences in clinical pregnancy rate, live birth rate, and neonatal outcome indicators (fetal weight and Apgar score) among the three stimulation protocols (antagonist protocol, GnRH agonist long protocol, and early follicular phase long protocol) (P>0.05). The findings showed no significant differences between clinical and neonatal outcomes in patients of all ages, regardless of the ovarian stimulation protocol, suggesting that the three ovarian stimulation protocols have similar therapeutic effects in patients of different ages. The results of this study have important implications for the selection of an appropriate ovarian stimulation protocol and the prediction of treatment outcomes.
UNASSIGNED: In the younger than 30 and 30-35 age groups, the GnRH antagonist protocol showed a more significant advantage over the GnRH agonist long protocol and the early follicular phase long protocol. This suggests that for younger and middle-aged patients, the antagonist protocol may lead to better outcomes during ovarian stimulation. In the older than 35 age group, while the antagonist protocol still outperformed the GnRH agonist long protocol, there was no significant difference compared to the early follicular phase long protocol. This may imply that with increasing age, the early follicular phase long protocol may have effects similar to the antagonist protocol to some extent. The advantages of the antagonist protocol lie in its ability to reduce stimulation duration and the dosage of GnRH, while enhancing patient compliance with treatment. This means that patients may find it easier to accept and adhere to this treatment protocol, thereby improving treatment success rates. Particularly for older patients, the use of the antagonist protocol may significantly increase the blastocyst formation rate, which is crucial for improving the success rates. Although there were no significant differences in the clinical outcomes of patients treated with the three protocols in each age group, further research is still needed to validate these findings. Future multicenter studies and increased sample sizes may help comprehensively assess the efficacy of different stimulation protocols. Additionally, prospective studies are needed to further validate these findings and determine the optimal treatment strategies.

UNASSIGNED: To investigate the effect of body mass index (BMI) on progesterone (P) level on trigger day in gonadotropin-releasing hormone antagonist (GnRH-ant) cycles.
UNASSIGNED: This study was a retrospective cohort study. From October 2017 to April 2022, 412 in-vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) patients who were treated with GnRH-ant protocol for controlled ovarian hyperstimulation (COH) in the reproductive center of our hospital were selected as the research objects. Patients were divided into three groups according to BMI level: normal weight group (n = 230):18.5 kg/m2≤BMI < 24 kg/m2; overweight group (n = 122): 24 kg/m2≤BMI < 28 kg/m2; Obesity group (n = 60): BMI ≥ 28 kg/m2. Variables with p < .10 in univariate analysis (BMI, basal FSH, basal P, FSH days, Gn starting dose and E2 level on trigger day) and variables that may affect P level on trigger day (infertility factors, basal LH, total FSH, HMG days and total HMG) were included in the multivariate logistic regression model to analyze the effect of BMI on P level on trigger day of GnRH-ant protocol.
UNASSIGNED: After adjustment for confounding factors, compared with that in normal weight patients, the risk of serum P elevation on trigger day was significantly lower in overweight and obese patients (OR = 0.434 and 0.199, respectively, p < .05).
UNASSIGNED: The risk of P elevation on trigger day in GnRH-ant cycles decreased with the increase of BMI, and BMI could be used as one of the predictors of P level on trigger day in GnRH-ant cycles.