The impact of long-acting Gonadotropin-releasing hormone agonist pretreatment on the clinical pregnancy outcomes of hormone replacement therapy-frozen embryo transfer in recurrent implantation failure patients with and without polycystic ovary syndrome: a retrospective clinical study.

In this retrospective study, the inclusion criteria encompassed 1602 patients who experienced RIF and underwent FET at the Reproductive Medicine Center of Nanjing Drum Tower Hospital from January 2017 to December 2021 (Fig. 1 ). The diagnostic criteria for RIF comprised of the following: individuals under 40 years of age, a minimum of two prior embryo transfers, and the inability to conceive post the transfer of at least four high-quality cleavage embryos, two high-quality blastocysts, or a combination of two high-quality cleavage embryos and one high-quality blastocyst [ 21 ]. The patients were subjected to either hormone replacement therapy (HRT, Femoston, 2 mg estradiol; 2 mg estradiol plus 10 mg dydrogesterone, Abbott, USA) or a long-acting GnRHa pretreated-HRT protocol. Given the retrospective design of the study, written informed consent was exempted for all patients, which had received approval from the Ethics Committee of Nanjing Drum Tower Hospital. Prior to the embryo transfer cycles, all patients underwent extensive preconception examinations to rule out any contraindications to medication or pregnancy. The overall health status of the patients involved in this study was deemed normal. The study’s exclusion criteria included: (1) usage of other hormone replacement medications; (2) existence of tubal hydrosalpinx, uterine cavity or endometrial abnormalities; (3) moderate to severe endometriosis or adenomyosis; (4) chromosomal abnormalities or patients who underwent preimplantation genetic testing (PGT). Fig. 1 A flow chart of the inclusion and exclusion of patients A flow chart of the inclusion and exclusion of patients Control group (HRT): During the initial phase of menstruation (on day two or three of the menstrual cycle), hormonal serum tests and transvaginal ultrasound examinations are conducted. In the absence of abnormalities, patients commence an oral fixed dose of exogenous estradiol (Femoston, 2 mg estradiol, 2 mg t.i.d.) and continue for 12–14 days. Serum estrogen (E 2 ) and progesterone (P) levels, along with endometrial thickness, are monitored. Patients with decreased endometrial thickness (< 8 mm) receive additional medication (Femoston, 2 mg estradiol, 4 mg b.i.d.). Upon reaching a specific endometrial thickness standard (≥ 8 mm), patients begin oral administration of estradiol combined with dydrogesterone compound tablets (Femoston, 2 mg estradiol and 10 mg dydrogesterone, t.i.d. for 5 or 6 days), and progesterone injections (progesterone injection solution, Zhejiang Xianju Pharmaceutical, China, 60 mg q.d.) to induce endometrial transformation. On the fifth day of endometrial transformation, cleavage-stage embryos are thawed and transferred, or on the sixth day, blastocysts are thawed and transferred [ 22 ]. Patients typically consume Femoston (2 mg estradiol and 10 mg dydrogesterone, t.i.d.) and receive luteal support with progesterone sustained-release vaginal gel (Crinone, Merck Pharmaceuticals, Switzerland, 90 mg q.d.). For the following patients, we may preferentially choose GnRHa pretreatment in clinical treatment: (1) thin endometrium (< 8 mm or less than expected thickness) in the previous transfer cycle; (2) history of endometritis suggested by previous examinations; (3) potential endometriosis; (4) empirical decision by the clinician. Research group (GnRHa-HRT): In the long-acting GnRHa pre-treated HRT cycle, a long-acting GnRHa (Decapeptyl, 3.75 mg) is administered on day two or three of menstruation, followed by the HRT cycle after 4 weeks. Serum β-human chorionic gonadotropin (β-hCG) is tested two weeks post-embryo transfer to ascertain biochemical pregnancy. Four weeks post-embryo transfer, transvaginal ultrasound examination is performed on patients with elevated β-hCG levels to confirm clinical pregnancy and the number of implanted embryos. Clinical pregnancy is characterized by the presence of a gestational sac observed on ultrasound. Luteal support for pregnant patients persists for two months post-embryo transfer. Patients are followed up to detect any pregnancy abnormalities. Early miscarriage is defined as spontaneous miscarriage occurring before 12 weeks of gestation. Live birth is defined as the delivery of a viable infant post 28 weeks of gestation, and the live birth rate is calculated as the ratio of the number of live births to the number of embryo transfer cycles. Most of our data is directly extracted from the patients’ medical record systems, while a small portion is registered by dedicated data entry personnel and verified by the responsible clinical doctors. The method of data collection is standardized, and the data source is genuine and reliable. Primary outcomes investigated in this study was clinical pregnancy rate, while live birth rate and early miscarriage rate were deemed as secondary outcomes. All cycles were categorized into HRT Group and GnRHa-HRT Group, contingent on the usage of long-acting GnRHa pretreatment. The association between GnRHa pretreatment and clinical pregnancy rate in RIF patients during FET cycles was evaluated. Univariate analysis was employed to initially assess potential confounding factors that may influence the clinical pregnancy rate, and these were included as adjusting variables in subsequent analyses. Moreover, a multivariate logistic regression model was utilized to examine the impact of GnRHa pretreatment on the clinical pregnancy rate in RIF patients. The diagnosis of PCOS was according to the Rotterdam criteria [ 23 ], and the population was bifurcated into non-PCOS and PCOS groups. The effect of long-acting GnRHa pretreatment on clinical pregnancy outcomes was reevaluated in both subgroups. Furthermore, an interaction analysis was conducted to assess the influence of PCOS on the relationship between long-acting GnRHa pretreatment and clinical pregnancy rate. We used the Kolmogorov-Smirnov normality test to detect the normal distribution of the variables. T-test was employed for the normally distributed variables and Mann Whitney-U test was employed for the non-normally distributed variables. For the statistical analysis for categorical variables, the variables were tested by chi-squared test (meeting the requirements of chi-square test: theoretical frequency (T) > 5 and sample number (n) > 40). The parameters distributed with normally distribution were explained as Mean ± Standard Deviation (SD) and the parameters distributed with non-normally distribution was explained as Median (25th-75th percentiles). All analyses were performed using R software ( http://www.R-project.org ) and EmpowerStats software ( www.empowerstats.com , X&Y Solutions, Inc. Boston MA). A p -value < 0.05 was considered to indicate statistical significance.

The baseline and clinical outcome variables of the HRT group ( n  = 1283) and the GnRHa-HRT group ( n  = 319) are presented in Table 1 . No significant differences were observed in female age, male age, body mass index (BMI), duration of infertility, type of infertility, baseline follicle-stimulating hormone (FSH), baseline luteinizing hormone (LH), fertilization method in the retrieval cycle, number and quality of transferred embryos, and type of transferred embryos between the two groups. A notable difference was found in anti-müllerian hormone (AMH) levels, with the GnRHa-HRT group showing significantly lower levels compared to the HRT group (3.00 ± 2.11 ng/mL vs. 3.74 ± 2.51 ng/mL, P  = 0.023). However, there were no significant differences in baseline FSH levels and AFC between the two groups, indicating that there is no significant deviation in ovarian reserve function between the two groups. In addition, the endometrial thickness in the GnRHa-HRT group () was significantly greater than that in the HRT group (9.98 ± 1.73 mm vs. 9.42 ± 1.37 mm, P  < 0.001). The number of embryos implanted, and the clinical pregnancy rate differed significantly between the two groups. The clinical pregnancy rate was significantly higher in the GnRHa-HRT group compared to the HRT group (59.25% vs. 47.70%, P  < 0.001), while no significant difference was observed in the early miscarriage rate. Table 1 Characteristics of FET cycles according to different endometrial preparation protocols Group HRT GnRHa-HRT P -value Number 1283 319 Baseline characteristics  Female age (years) 32.00 (22.00–40.00) 31.00 (22.00–40.00) 0.907  Male age (years) 33.00 (23.00–59.00) 32.00 (23.00–58.00) 0.528  BMI (kg/m 2 ) 22.30 (15.90–37.30) 22.10 (16.00–34.80) 0.751  Duration of infertility (years) 3.00 (1.00–17.00) 3.00 (1.00–18.00) 0.954 Infertility type 0.454  Primary infertility 690 (53.78%) 179 (56.11%)  Secondary infertility 593 (46.22%) 140 (43.89%) AMH (ng/mL) 3.13 (0.09–14.10) 2.58 (0.44–14.70) 0.023 Basic FSH (IU/L) 7.08 (0.69–32.33) 7.09 (3.05–26.81) 0.448 Basic LH (IU/L) 4.26 (0.29–39.49) 4.20 (0.05–22.13) 0.903 AFC (n) 16.7 ± 7.9 17.1 ± 7.5 0.864 Fertilization mode 0.814  IVF 924 (72.07%) 223 (69.91%)  ICSI 262 (20.44%) 73 (22.88%)  RICSI 35 (2.73%) 8 (2.51%)  TESA/PESA-ICSI 61 (4.76%) 15 (4.70%) FET characteristics Endometrium thickness (mm) 9.00 (7.00–16.00) 9.50 (7.00–16.50) < 0.001 Number of transferred embryos 1.63 ± 0.48 1.64 ± 0.48 0.093 Embryo quality 0.077  0 646 (50.35%) 143 (44.83%)  1 637 (49.65%) 176 (55.17%) Type of transferred embryos 0.233  Cleavage-stage embryo 804 (62.67%) 184 (57.68%)  Blastocyst 465 (36.24%) 130 (40.75%)  Sequential transfer 14 (1.09%) 5 (1.57%) Implanted embryo number (n) 0.59 ± 0.69 0.74 ± 0.70 < 0.001 Clinical pregnancy rate 612 (47.70%) 189 (59.25%) < 0.001 Early miscarriage rate 78 (12.75%) 27 (14.29%) 0.583 Embryo quality: 1 for at least one high-quality cleavage-stage embryo or blastocyst FET frozen-thawed embryo transfer, HRT hormone replacement therapy, GnRHa Gonadotropin-Releasing Hormone agonist, BMI body mass index, AMH anti-müllerian hormone, FSH follicle-stimulating hormone, LH luteinizing hormone, AFC antral follicle count, IVF in vitro fertilization, ICSI intracytoplasmic sperm injection, TESA testicular epididymal sperm aspiration, PESA percutaneous epididymal sperm aspiration Characteristics of FET cycles according to different endometrial preparation protocols Embryo quality: 1 for at least one high-quality cleavage-stage embryo or blastocyst FET frozen-thawed embryo transfer, HRT hormone replacement therapy, GnRHa Gonadotropin-Releasing Hormone agonist, BMI body mass index, AMH anti-müllerian hormone, FSH follicle-stimulating hormone, LH luteinizing hormone, AFC antral follicle count, IVF in vitro fertilization, ICSI intracytoplasmic sperm injection, TESA testicular epididymal sperm aspiration, PESA percutaneous epididymal sperm aspiration We conducted preliminary univariate analysis (Table S1), which showed that both female age and male age were significantly negatively correlated with clinical pregnancy rate. GnRHa pretreatment, AMH level, endometrial thickness, number and quality of transferred embryos, as well as embryo type, were all significantly positively correlated with clinical pregnancy rate. These potential confounding factors were included as adjustment variables in subsequent regression analysis to eliminate their influence. Multiple regression analysis (Table 2 ) revealed that after considering the above confounding factors, GnRHa pretreatment still had a positive impact on clinical pregnancy rate. The clinical pregnancy rate in the GnRHa pretreatment group was 1.51 times higher than that in the control group. Table 2 Multivariate analysis for GnRHa pretreatment in FET cycles involved in the clinical pregnancy rate GnRHa pretreatment Adjusted OR 95% CI p value No 1.00 1.00 Yes 1.51 (1.15, 1.99) 0.003 Adjust for: female age, male age, AMH, endometrial thickness, number of transferred embryos, embryo quality and type of transferred embryos GnRHa Gonadotropin-Releasing Hormone agonist, FET frozen-thawed embryo transfer, OR odds ratio, CI Confidence Interval, AMH anti-müllerian hormone Multivariate analysis for GnRHa pretreatment in FET cycles involved in the clinical pregnancy rate Adjust for: female age, male age, AMH, endometrial thickness, number of transferred embryos, embryo quality and type of transferred embryos GnRHa Gonadotropin-Releasing Hormone agonist, FET frozen-thawed embryo transfer, OR odds ratio, CI Confidence Interval, AMH anti-müllerian hormone We conducted further stratified analysis by dividing the RIF population into two groups: non-PCOS patients and PCOS patients. A comparison of baseline and clinical outcome variables between the two groups revealed that the clinical pregnancy rates in the GnRHa-HRT group were higher than those in the HRT group for both groups of patients (Table S2 and S3). To further elucidate the correlation, multivariate regression analysis was performed on the two groups of patients, considering confounding factors such as female age, male age, AMH, endometrial thickness, number of transferred embryos, embryo quality, and type of transferred embryos. For non-PCOS patients, the clinical pregnancy rate after GnRHa pretreatment was 1.59 times higher than that in the control group ( P  = 0.004, Table 3 ). However, for PCOS patients, although the clinical pregnancy rate was higher in those who received GnRHa pretreatment compared to those who did not (aOR = 1.51), it was not statistically significant. This suggests that GnRHa pretreatment can significantly enhance the clinical pregnancy rate in non-PCOS patients, but its impact on PCOS patients is not significant. Table 3 Multivariate analysis for GnRHa pretreatment in FET cycles of patients with PCOS or not involved in the clinical pregnancy rate Non-PCOS PCOS GnRHa pretreatment Adjusted OR 95% CI p -value Adjusted OR 95% CI p -value No 1 1 Yes 1.59 (1.16, 2.18) 0.004 1.51 (0.81, 2.82) 0.195 Adjust for: female age, male age, AMH, endometrial thickness, number of transferred embryos, embryo quality and type of transferred embryos GnRHa Gonadotropin-Releasing Hormone agonist, FET frozen-thawed embryo transfer, PCOS polycystic ovary syndrome, OR odds ratio, CI Confidence Interval, AMH anti-müllerian hormone Multivariate analysis for GnRHa pretreatment in FET cycles of patients with PCOS or not involved in the clinical pregnancy rate Adjust for: female age, male age, AMH, endometrial thickness, number of transferred embryos, embryo quality and type of transferred embryos GnRHa Gonadotropin-Releasing Hormone agonist, FET frozen-thawed embryo transfer, PCOS polycystic ovary syndrome, OR odds ratio, CI Confidence Interval, AMH anti-müllerian hormone To further investigate whether PCOS modifies the impact of GnRHa pretreatment on the clinical pregnancy rate of RIF patients, we conducted an interaction analysis (Table 4 ). After adjusting for confounding factors, we found that for non-PCOS patients, the clinical pregnancy rate of females who underwent GnRHa pretreatment was 1.60 times higher than that of untreated females ( P  = 0.003). However, for PCOS patients, there was no significant difference in the clinical pregnancy rate between females who received GnRHa pretreatment and those who did not (aOR = 0.89, P  = 0.936). Moreover, the influence of GnRHa pretreatment on the clinical pregnancy rate was not affected by PCOS (interaction P -value = 0.818). Table 4 Effect modification of GnRHa pretreatment on clinical pregnancy rate according to whether PCOS or not GnRHa pretreatment PCOS N Clinical pregnancy rate Adjusted OR (95%CI) P -value No No 874 874 (45.88%) 1.00 Yes No 256 256 (57.81%) 1.60 (1.17, 2.19) 0.003 No Yes 409 409 (51.59%) 0.61 (0.04, 9.37) 0.719 Yes Yes 63 63 (65.08%) 0.89 (0.05, 14.54) 0.936 P interaction 0.818 Adjust for: female age, male age, AMH, endometrial thickness, number of transferred embryos, embryo quality and type of transferred embryos GnRHa Gonadotropin-Releasing Hormone agonist, FET frozen-thawed embryo transfer, PCOS polycystic ovary syndrome, OR odds ratio, CI Confidence Interval, AMH anti-müllerian hormone Effect modification of GnRHa pretreatment on clinical pregnancy rate according to whether PCOS or not Adjust for: female age, male age, AMH, endometrial thickness, number of transferred embryos, embryo quality and type of transferred embryos GnRHa Gonadotropin-Releasing Hormone agonist, FET frozen-thawed embryo transfer, PCOS polycystic ovary syndrome, OR odds ratio, CI Confidence Interval, AMH anti-müllerian hormone

Recurrent Implantation Failure (RIF) is a prominent challenge within Assisted Reproductive Technology (ART), significantly impacting the fertility success rates in women. Despite the absence of a universally accepted definition within the academic sphere, RIF is typically characterized by the failure to achieve a clinical pregnancy following a minimum of two transfers involving high-quality embryos (inclusive of at least four high-quality cleavage embryos or two high-quality blastocysts) [ 1 – 3 ]. The prevalence of RIF ranges between 10% and 20%, and for patients within this category, the success rate per embryo transfer may fall below 10% [ 4 ]. The etiology of RIF is multifactorial, encompassing embryo-associated factors, uterine cavity environmental factors, maternal factors, and immune factors, amongst others [ 5 ]. Presently, there are no efficacious interventions available to enhance the clinical pregnancy outcomes in patients diagnosed with RIF [ 4 ]. Numerous studies have explored the potential impact of gonadotropin-releasing hormone agonists (GnRHa) pretreatment on the clinical outcomes of frozen embryo transfer (FET) cycles. Several findings indicate that pretreatment with GnRHa does not enhance reproductive outcomes in women undergoing hormone replacement therapy (HRT)-FET [ 6 ]. Some other studies have shown a positive effect of GnRHa pretreatment on patients with a history of multiple failed embryo transfers. Additionally, research has indicated that pretreatment with long-acting GnRHa may substantially enhance the clinical pregnancy rate in patients experiencing recurrent implantation failure (RIF) [ 7 – 9 ]. This improvement is believed to be due to GnRHa’s ability to enhance endometrial receptivity, leading to increased success in embryo implantation [ 10 ]. GnRHa achieves this by suppressing excessive gonadotropin secretion, preventing premature endometrial thickening, improving the uterine environment, and enhancing the success rate of frozen embryo transfer [ 10 – 12 ]. As a result, the use of pretreatment with long-acting GnRHa is becoming increasingly popular in addressing RIF. Several studies have identified an elevated risk of RIF in patients diagnosed with Polycystic Ovary Syndrome (PCOS), which might be associated with the endocrine and metabolic dysfunctions, aberrant endometrium, and chronic inflammatory state observed in these patients [ 13 , 14 ]. PCOS, as an endocrine disease, is characterized by excessive production of luteinizing hormone (LH) and a hyperandrogenic microenvironment, as well as the role of inflammatory factors. Previous studies suggested that pretreatment with GnRHa for PCOS patients may be beneficial for embryo implantation by adjusting the levels of estrogen and LH, which could enhance the clinical pregnancy outcomes in patients diagnosed with PCOS [ 15 – 17 ]. Several investigations have reported an increase in the clinical pregnancy rate and live birth rate in PCOS patients following GnRHa pretreatment [ 18 , 19 ]. Nonetheless, a study has found no significant improvement in the clinical pregnancy rate in PCOS patients post-GnRHa pretreatment [ 20 ]. Consequently, the findings of various studies regarding the efficacy of GnRHa pretreatment in improving clinical pregnancy outcomes in PCOS patients are incongruous. Moreover, research investigating whether long-acting GnRHa exhibits superior clinical efficacy in RIF patients with a history of PCOS is limited. Therefore, we conducted a retrospective study on RIF patients at our reproductive medicine center from 2017 to 2021 to investigate the impact of long-acting GnRHa pretreatment on the clinical pregnancy outcomes of RIF patients. Additionally, we aimed to explore the differential effects of long-acting GnRHa on both PCOS and non-PCOS populations.

Prior research has indicated that pre-treatment with long-acting GnRHa may enhance clinical pregnancy outcomes in patients with RIF [ 7 , 24 , 25 ]. Our study corroborated these findings: through multiple regression analysis, it was observed that, when accounting for pertinent confounding variables, the clinical pregnancy rate in the GnRHa pre-treatment cohort was 1.51 times higher compared to the control group. Subsequent stratified analysis revealed that, among non-PCOS patients, the clinical pregnancy rate post-GnRHa pre-treatment was 1.59 times greater than that in the control group. Conversely, for PCOS patients, while the clinical pregnancy rate was elevated in those who underwent GnRHa pre-treatment in contrast to those who did not, the disparity lacked statistical significance. Furthermore, the impact of GnRHa pre-treatment on clinical pregnancy rate remained unaffected by PCOS status. The artificial cycle is a commonly utilized method for endometrial preparation in FET, yet research findings regarding the addition of GnRHa pre-treatment are inconclusive. GnRHa may enhance embryo implantation rates and clinical pregnancy outcomes by influencing endometrial status and early embryos [ 26 ]. Previous studies have indicated that GnRHa can enhance clinical outcomes in fresh embryo transfer, although findings in FET cycles are inconsistent [ 27 , 28 ]. A meta-analysis encompassing a total sample size exceeding 10,000 cases assessed the impact of GnRHa pre-treatment on outcomes in HRT-FET cycle: the embryo implantation rate (OR = 1.31), clinical pregnancy rate (OR = 1.27), and live birth rate (OR = 1.16) in the GnRHa pre-treatment group surpassed those in the non-pretreatment group. Subgroup analysis focused on patients with recurrent implantation failure, revealing that GnRHa pre-treatment significantly enhanced embryo implantation and clinical pregnancy rates in such individuals [ 25 ]. Furthermore, a study explored the effects of long-acting GnRHa pre-treatment on the clinical outcomes of women with varying numbers of prior transplant failures: while reproductive outcomes were comparable for women with no or one prior failure, those with multiple failures exhibited higher pregnancy rates in the GnRHa pre-treatment group, particularly among younger patients aged 35–37 years [ 7 ]. Our study produced similar findings, underscoring the potential of long-acting GnRHa pre-treatment to enhance clinical pregnancy outcomes in patients with Recurrent Implantation Failure. Prior research has indicated that GnRHa may enhance endometrial receptivity and facilitate embryo implantation, potentially serving as a mechanism to enhance clinical outcomes in patients with RIF. The expression of GnRH and its receptors in various peripheral tissues, including the uterus, suggests that GnRH may exert autocrine and paracrine effects, with the uterus serving as a target for GnRHa activity. GnRHa therapy has been shown to modulate the growth and apoptosis processes of diverse cells, as well as the expression of growth factors and cytokines. Moreover, GnRHa treatment influences the expression of lipid receptors in the uterine endometrium and exhibits a notable anti-proliferative impact [ 12 , 29 – 34 ]. RIF may be linked to immune dysregulation, such as an imbalance in Th17/Treg cells. GnRHa can directly modulate immune cell receptors in the uterus and impact Th17/Treg levels, thereby enhancing the uterine environment [ 11 ]. A high proportion of PCOS patients fail to conceive after ovulation induction, and the frequency of requiring in vitro fertilization (IVF) treatment is eight to ten times higher than non-PCOS women. Therefore, in addition to ovulation disorders, other factors such as uterine endometrial function and ovarian function may independently reduce the fertility of PCOS women [ 35 ]. Research findings have indicated that patients with PCOS exhibit reduced fertility rates and elevated miscarriage rates even following IVF treatment [ 14 ]. Experimental and clinical data indicate that the endometrium of women with PCOS exhibits abnormalities. PCOS and its associated conditions may lead to dysregulation of endometrial expression of sex hormone receptors and co-receptors, increased endometrial insulin resistance, impaired glucose transport and utilization, resulting in chronic low-grade inflammation, immune dysfunction, uterine vascular abnormalities, abnormal endometrial gene expression, and cellular abnormalities [ 6 ]. Studies have revealed notable decreases in crucial molecules involved in embryo adhesion within the endometrium, such as homeobox A10 (HOXA10) and leukemia inhibitory factor (LIF), in individuals with PCOS, potentially attributed to the inhibitory effects of elevated androgen levels. Furthermore, factors like insufficient energy supply, heightened oxidative stress, diminished angiogenesis, and abnormal pro-inflammatory conditions within the endometrium of PCOS patients can also impact the process of embryo implantation [ 13 ]. Based on the existing data, there are no recommended treatments or strategies to enhance endometrial receptivity in women with PCOS. This assertion holds true for PCOS patients facing recurrent miscarriages and/or RIF. PCOS is a complex syndrome comprising various disorders with diverse endocrine traits, and the effectiveness of interventions for patients may differ based on individual hormonal and metabolic profiles [ 36 ]. There are also studies suggesting that changing lifestyle, using metformin, and weight loss may be relatively effective ways to improve endometrial function in women with PCOS [ 6 ]. In order to explore more potential clinical methods, clinical investigations have demonstrated that the utilization of a combination therapy of GnRHa and human menopausal gonadotropin (hMG) can lead to reduced miscarriage rates and increased live birth rates in anovulatory PCOS women compared to gonadotropin monotherapy [ 37 ]. Specifically, in PCOS women undergoing GnRHa pretreatment, there is a significant enhancement in ongoing pregnancy rates during frozen embryo transfer cycles [ 19 ]. Conversely, conflicting study results have been reported regarding the impact of GnRHa pretreatment on live birth rates in PCOS women [ 20 ]. Subgroup analysis was performed to evaluate the potential differential effects of GnRHa pretreatment on treatment outcomes in RIF patients with PCOS. The analysis revealed a notable rise in clinical pregnancy rates among the PCOS subgroup of RIF patients following GnRHa pretreatment. However, correlation and interaction analyses indicated no statistically significant association between GnRHa pretreatment and improvements in clinical pregnancy rates. We speculate that patients with PCOS generally have a higher number of retrieved oocytes compared to non-PCOS patients, as well as a higher number of high-quality embryos overall. Therefore, even after multiple failed transfers, PCOS patients still have a higher probability of having high-quality embryos compared to non-PCOS patients. This may be the reason why their overall pregnancy rates are higher than those of non-PCOS patients. Consequently, we believe that a personalized definition of RIF is particularly crucial, as PCOS patients may require more transfer attempts and the use of more high-quality embryos before being considered as having RIF. The definition of RIF cannot be generalized, which provides new ideas for our future research. Therefore, despite observing an increase in clinical pregnancy rates post-GnRHa pretreatment, statistical significance was not achieved. There are several limitations to our study. Our investigation is constrained to HRT cycles primarily due to the limited number of PCOS patients undergoing controlled ovarian stimulation to prime the endometrium, aiming to prevent excessive follicle development that could result in ovarian hyperstimulation syndrome (OHSS). There is limited research on the selection of endometrial preparation protocols for FET cycles in patients with PCOS. Some studies [ 38 ] suggest that compared to HRT cycles, stimulated cycles have higher live birth rates and lower miscarriage rates. Additionally, the use of letrozole in stimulated cycles can lead to better reproductive outcomes [ 39 ]. Moreover, we did not categorize PCOS patients into specific subtypes, raising the possibility that GnRHa may exert a more pronounced therapeutic effect on certain PCOS subtypes. While the number, type, and quality of embryo transfers were considered as adjusted variables for statistical analyses, detailed embryo grading was not incorporated. Additionally, due to constraints in our data system, accessing comprehensive data on patient follicle retrieval cycles during the pre-transfer preparation phase was challenging, preventing further analysis of potential confounding variables associated with these cycles (such as infertility factors, ovarian stimulation protocol, number of previous transfers and so on). Due to our inability to identify potential patients with endometriosis and obtain corresponding data, we are unable to exclude the potential benefits of using GnRHa treatment in such patients. Immunological abnormalities could be one of the potential causes of RIF, but our patients have not undergone comprehensive examinations, which is also a deficiency of this study. The primary limitation of our study lies in its retrospective nature, underscoring the need for caution when extrapolating these findings to real-world scenarios. To gain deeper insights into the impact of GnRHa pretreatment on patients with RIF and concurrent PCOS, high-quality, large-scale prospective studies are essential. Such studies would facilitate more precise selection of cycle preparation strategies for RIF patients, minimize unnecessary GnRHa pretreatment, reduce time and costs, and uphold superior clinical pregnancy rates.

Our study demonstrates that long-acting GnRHa pretreatment can enhance clinical pregnancy outcomes in patients with RIF. Among RIF patients without PCOS, the clinical pregnancy rate exhibited a significant increase following GnRHa pretreatment compared to the control group. However, in RIF patients with concurrent PCOS, although there was an elevation in the clinical pregnancy rate post-GnRHa pretreatment, the difference did not reach statistical significance. Moreover, the influence of GnRHa pretreatment on the clinical pregnancy rate was found to be consistent regardless of the presence of PCOS.

Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2. Supplementary Material 3. Supplementary Material 3.