Patient-specific predictors of successful frozen embryo transfer using the freeze-all protocol: a retrospective observational study
Article information
Abstract
Background
The aim of this study was to examine various patient factors affecting first programmed embryo transfer (ET) outcomes under the freeze-all policy at a single tertiary university infertility center.
Methods
This retrospective observational study reviewed the medical records of 243 couples who underwent their first ET using blastocysts collected under the freeze-all antagonist-controlled ovarian stimulation (COS) protocol from 2015 to 2023. Patients were grouped into pregnant and nonpregnant groups, and their data, including demographics, COS and ET outcomes, and embryo storage duration, were analyzed.
Results
Patient body mass index, cause of infertility, follicle-to-oocyte index, distribution of blastocyst grades, number of transferred embryos, and embryo storage duration were not significantly different between the groups. In a simple comparative analysis, patients with successful clinical pregnancy tended to have significantly lower female and male age (33.83±3.67 and 35.32±4.54 years vs. 37.07±4.15 and 39.33±5.60 years, respectively), higher anti-Müllerian hormone levels (6.27±5.32 ng/mL vs. 4.14±3.82 ng/mL) and antral follicle counts (14.20±8.26 vs. 10.04±5.75), and higher numbers of retrieved oocytes and metaphase II oocytes (13.74±6.92 and 9.64±6.19 vs. 11.21±6.04 and 7.53±5.56, respectively). Multivariate logistic regression analysis of these variables revealed that only male age was a significant factor for successful clinical pregnancy (odds ratio, 4.768; 95% confidence interval, 1.252–18.162; p=0.022).
Conclusion
During the first programmed ET using blastocysts collected under the freeze-all antagonist COS protocol, male age and correspondingly the quality of gametes for fertilization were crucial for successful pregnancy, having more importance than calculated female ovarian reserve and embryo storage duration.
Introduction
Treatment of infertility, a persistent global health concern resulting in profound emotional, social, and economic consequences for reproductive-aged individuals and couples, has fortunately witnessed significant advancements in assisted reproductive technology (ART), with the “freeze-all” policy emerging as a promising strategy to optimize in vitro fertilization (IVF) outcomes [1,2]. The freeze-all approach involves cryopreserving all embryos generated during an IVF cycle, allowing for subsequent transfer in a more receptive uterine environment. Its major advantages include potentially improving implantation and live birth rates while mitigating complications associated with ovarian hyperstimulation syndrome (OHSS) [3,4]. The freeze-all policy also facilitates the use of programmed endometrial preparations for frozen embryo transfer (FET), which allows for greater control over the timing of embryo transfer and potentially enhances implantation rates [4-6]. In addition, the first FET cycle following a freeze-all approach holds particular significance as it represents a unique opportunity to assess the impact of patient factors and embryo quality on treatment outcomes and is more likely to be independent of potential confounding effects from prior controlled ovarian stimulation (COS) cycles or fresh embryo transfer [7-9]. However, the freeze-all policy also has a few disadvantages, such as the additional cost and time required for cryopreservation, and the potential risks associated with embryo freezing and thawing. Despite these considerations, the use of freeze-all cycles has been steadily increasing, with recent studies indicating that they are currently employed in a significant proportion of ART cycles worldwide. Thus, evaluating the predictors of success in the initial FET cycle, particularly regarding various unique patient factors, is crucial for optimizing treatment strategies and patient counseling [10].
Previous studies have investigated various factors influencing FET success, including patient demographics, ovarian stimulation protocols, and embryo quality. Clinically, a multitude of female and male factors influence reproductive outcomes in infertility setting; female factors may include ovarian reserve, underlying gynecological diseases such as endometriosis and/or adenomyosis, tubal patency, and other uterine abnormalities, whereas male factors encompass sperm parameters, genetic abnormalities, and lifestyle factors [11,12]. Moreover, although advanced maternal age is a well-established risk factor for decreased fertility, increased miscarriage rates, and adverse perinatal outcomes, the effect of paternal age on reproductive outcomes, especially in ART cycles, remains a subject of ongoing debate [13,14]. Several studies have suggested a potential decline in sperm quality with advancing male age, and a possible association with congenital anomalies and neurodevelopmental disorders [15,16]. Such findings highlight the importance of considering both maternal and paternal age as well as other patient-specific factors in the context of infertility treatment, particularly given the global trend towards delayed childbearing and increasing utilization of ART in women of advanced maternal age [17]. Beyond chronological age, the complex interplay of these patient factors underscores the need for comprehensive assessment of both partners to optimize treatment strategies and patient counseling.
Despite the growing body of literature on infertility and ART, the specific predictors of success in the first programmed FET cycle remain unclear. Thus, we conducted a comprehensive analysis of female and male patient characteristics to identify patient-specific factors associated with successful first programmed FET outcomes under the freeze-all policy. By unraveling these sophisticated relationships, this study could lead to personalized infertility treatments, enhanced patient counseling, and ultimately improved success rates for couples pursuing parenthood.
Methods
Ethics statement: The study protocol was approved by the Institutional Review Board (IRB) of Pusan National University Hospital (IRB No: 2204-003-113). The IRB waived the patient informed consent due to the retrospective observational design and anonymized data; however, all patients routinely provided written informed consent for the potential research use of their data during their initial visit to our tertiary care infertility center, according to the patient information provision and processing regulations of the associated hospital.
1. Study design and patient characteristics
The retrospective observational study analyzed electronic medical records of 243 couples who underwent their first programmed FET using blastocysts exclusively, following a gonadotropin-releasing hormone (GnRH) antagonist ovarian stimulation protocol and a freeze-all strategy at our center between 2015 and 2023. The inclusion criteria were as follows: (1) female aged 20 to 45 years, (2) male aged 20 to 55 years, (3) first FET cycle, and (4) availability of complete medical records. The exclusion criteria were as follows: (1) known chromosomal abnormalities; (2) uterine anomalies, including synechiae; (3) a history of recurrent pregnancy loss, including abnormal natural killer cell levels/activity and/or thrombophilia; (4) severe male factor infertility, including azoospermia or severe oligozoospermia; (5) smoking/alcohol abuse; and (6) a history of unilateral oophorectomy/chemotherapy. These exclusion criteria were applied to minimize the potential confounding effects of these factors on FET outcomes.
The included patients were categorized into pregnant and nonpregnant groups based on FET outcomes. Clinical, demographic, COS cycle, and FET outcome data were extracted from the electronic health records. The duration of embryo storage until transfer was also assessed. Couples were excluded if the man had severe male factor infertility or if the woman utilized donor oocytes, changed age groups during the study, had abnormal natural killer cell levels/activity and/or thrombophilia, had uterine anomalies/synechiae, reported smoking and/or alcohol use, or had undergone unilateral oophorectomy/chemotherapy.
2. Controlled ovarian stimulation oocyte retrieval and embryo transfer
All patients underwent COS using a flexible GnRH antagonist protocol followed by IVF or intracytoplasmic sperm injection, including medical injections of recombinant follicle-stimulating hormone (rFSH), human menopausal gonadotropin (HMG), and GnRH antagonists, as previously described [18]. Initial rFSH (Gonal-F; Merck Serono, Geneva, Switzerland) doses ranged from 75 to 300 IU, whereas those of HMG (Menopur; Ferring Pharmaceuticals, Parsippany, NJ, USA) ranged from 75 IU to 150 IU. Stimulation lasted for 10 to 16 days with serial transvaginal ultrasound monitoring of follicular growth and endometrial development to guide dose adjustments. Human chorionic gonadotropin and/or GnRH agonist triggered final oocyte maturation when the dominant follicle reached ≥17 mm. Oocyte retrieval was performed 35 hours post-triggering via ultrasound-guided aspiration [19].
A maximum of two blastocyst embryos were transferred in all cases, which were carried out as programmed FET procedures with appropriate endometrial preparation and embryo transfer timings, as previously described [20].
3. Statistical analysis
Categorical variables were compared between the pregnant and nonpregnant groups using the chi-square test. Continuous variables are expressed as mean±standard deviation or median and interquartile range (IQR), as appropriate; the normality of continuous variables was assessed, and group comparisons were performed using independent t-tests for normally distributed data with equal variances, or the Wilcoxon rank-sum test for non-normally distributed data. The effects of various patient factors on clinical pregnancy were further evaluated using univariate and multivariate logistic regression analyses. The results of these analyses are presented as odds ratios (ORs) with 95% confidence intervals (CIs). Statistical analyses were conducted using R (version 4.2.1, https://cran.r-project.org; R Foundation for Statistical Computing, Vienna, Austria), with p<0.05 considered statistically significant. The study design and statistical approach were reviewed and validated by the Department of Biostatistics of the Biomedical Research Institute at Pusan National University Hospital.
Results
The current study evaluated possible predictors of successful first programmed FET outcomes in infertile couples undergoing a freeze-all protocol. As described in Table 1, the included patients had a female age of 34.78±4.30 years, male age of 38.50±5.22 years, body mass index (BMI) of 22.89±3.75 kg/m2, and embryo storage duration of 106 days (IQR, 60.50–228.50 days). The patient BMI, cause of infertility, follicle-to-oocyte index, number of transferred embryos, and embryo storage duration were not significantly different between the pregnant and nonpregnant groups. As described in Table 2, in a simple comparative analysis, the patients with successful clinical pregnancy tended to have significantly lower female and male age (33.83±3.67 and 35.32±4.54 years vs. 37.07±4.15 years and 39.33±5.60 years, respectively), higher anti-Müllerian hormone (AMH) levels (6.27±5.32 ng/mL vs. 4.14±3.82 ng/mL), a higher antral follicle count (AFC; 14.20±8.26 vs. 10.04±5.75), and higher numbers of retrieved oocytes and metaphase II (MII) oocytes (13.74±6.92 and 9.64±6.19 vs. 11.21±6.04 and 7.53±5.56, respectively). Multivariate logistic regression identified male age as the only statistically significant predictor of successful clinical pregnancy (OR, 4.768; 95% CI, 1.252–18.162; p=0.022).
Overall patient characteristics
Logistic regression analysis of significant patient factor candidates regarding assisted reproductive technology outcomes
Discussion
The current study revealed that while several factors, including female and male age, AMH, AFC, retrieved oocyte number, and MII oocytes, initially appeared to influence clinical pregnancy rates, only male age remained a statistically significant predictor in multivariate analysis. The mean female and male ages in the pregnant group were significantly lower than those in the nonpregnant group, suggesting that both maternal and paternal ages play a critical role in FET success. However, multivariate analysis indicated that only male age independently predicted clinical pregnancy, highlighting its potential importance in the context of the freeze-all strategy.
To explore the possible patient factors affecting FET outcomes in the freeze-all strategy, the current study focused on the first FET cycle in the protocol, as it allowed for a clearer assessment of the intrinsic patient and embryo factors that contribute to successful implantation and pregnancy, independent of the potential confounding effects of prior ovarian stimulation or fresh embryo transfer [7-9]. As previously discussed, FET offers distinct advantages over fresh transfer, particularly in mitigating the potential negative effects of OHSS and creating a more receptive endometrial environment, especially in women with refractory and/or thin endometria [21-23]. Furthermore, the timing flexibility of FET allows personalized treatment plans and optimization of endometrial receptivity [24]. Although the debate on fresh vs. frozen transfer is ongoing, recent studies have demonstrated comparable, if not superior, success rates of FET, especially in specific patient populations such as women with polycystic ovary syndrome [20,25,26].
However, the freeze-all approach inherently involves a period of embryo cryopreservation, the duration of which can influence embryo transfer outcomes. While the current literature presents a mixed picture, recent studies suggest that cryopreservation itself may have a minimal impact on embryo survival and implantation rates [27-29]. In the present study, blastocyst storage duration, 311.78±442.37 days for the pregnant group and 218.20±333.50 days for the nonpregnant group, did not significantly differ between the study groups; such a finding aligns with those of Cui et al. [27] and Canosa et al. [29], who reported no detrimental effects of prolonged cryopreservation on pregnancy outcomes, especially within 5 years. However, the effects of prolonged cryopreservation on the viability and implantation potential of untested blastocysts, which may harbor chromosomal abnormalities, remain an area of ongoing research. In a study by Cui et al. [27], the duration of untested embryo cryopreservation did not significantly affect pregnancy outcomes within 5 years, but both clinical pregnancy and live birth rates decreased significantly when the cryopreservation duration exceeded 5 years. According to Canosa et al. [29], although the effect of prolonged cryopreservation for more than 1 year on the live birth rate did not significantly differ among euploid blastocysts, untested blastocysts, and even untested cleavage-stage embryos, the cryo-survival rate per warmed embryo in untested blastocysts significantly decreased when the storage duration exceeded 12 months. Moreover, Zhang et al. [28] conducted further stratification analyses based on various cryopreservation durations for untested embryos, from 3 weeks to more than 52 weeks, and their influence on embryo transfer outcomes, suggesting that storage duration was significantly and inversely associated with pregnancy rates, especially in women who are older. Furthermore, Zheng et al. [30] suggested that prolonged cryopreservation of blastocysts for up to 24 months might negatively affect pregnancy outcomes, which progresses as the storage duration exceeds 5 years. However, these studies in analyzing the ultimate pregnancy outcomes varied in specific patient factors, including the age of the couple, infertility duration and etiology distributions, insemination or fertilization method, endometrial preparation method, quality and ploidy confirmation of embryo transferred, and endometrial thickness; therefore, the analysis of the degree of contribution of these factors to the embryo transfer outcome needs to be evaluated in depth.
Other than embryo storage duration, such patient factors, especially the age of the infertile couple, have been previously associated with IVF outcomes in FET cycles utilizing a freeze-all policy through various possible mechanisms [31,32]. In the present study, the lack of an independent association between female age and clinical pregnancy in the multivariate analysis may be attributed to the relatively narrow age range of our study population, of which overall mean ages were 34.78±4.30 years for females and 38.50±5.22 years for males, and the potential influence of other factors, such as ovarian reserve. The positive correlation observed between clinical pregnancy and AMH levels and AFC emphasizes the importance of ovarian reserve as a predictor of FET success, even within the context of a freeze-all strategy. This has been observed in other major infertility populations, such as those with endometriosis and poor ovarian response [33,34].
However, the significant association between male age and clinical pregnancy, independent of other factors, challenges the traditional notion that paternal age has a negligible effect on ART outcomes [14-16]. This finding suggests that advancing paternal age, >40 years in autologous oocyte cycles, and >50 years in donor oocyte cycles, influences embryo competency and subsequent implantation potential, even when embryos are cryopreserved and transferred to a presumably optimal uterine environment [35,36]. The mechanisms underlying this association are multifaceted and have been explored in numerous studies [35]. First, increasing paternal age may be correlated with a decline in various sperm parameters, including concentration, motility, volume, and morphology [37]. Second, a positive correlation between paternal age and sperm DNA damage has been suggested, with the extent of damage potentially doubling between the ages of 25 and 55 years [38]. The integrity of sperm DNA is crucial for successful fertilization and healthy embryo development; thus, age-related increases in DNA damage may contribute to the decreased clinical pregnancy rates observed in couples of advanced paternal age [35]. In addition to the direct effects on sperm parameters, other proposed factors include the association between advanced paternal age and sperm epigenomic alterations. The process of epigenetic reprogramming during spermatogenesis and spermiogenesis can be disrupted in men who are older, leading to global DNA hypermethylation or specific methylation loss of paternal imprinted genes [35,39]. These epigenetic modifications may affect gene expression and embryonic development, potentially contributing to a decline in ART success rates with increasing paternal age [35,39]. Furthermore, studies have indicated a negative correlation between paternal age and genomic stability as well as changes in microRNA expression in sperm [40]. These genetic and epigenetic alterations in sperm from men who are older may compromise embryo quality and implantation potential, ultimately leading to reduced pregnancy rates. Lastly, factors such as smoking and metabolic diseases, which are more prevalent in men who are older, can also influence sperm epigenetics and contribute to the observed age-related decline in ART outcomes. The impact of paternal age on reproductive success is likely a complex interplay of both intrinsic and extrinsic factors [40]. Thus, the findings of the present study, which identified male age as an independent predictor of clinical pregnancy in FET cycles, are supported by a growing body of debatable evidence highlighting the detrimental effects of advancing paternal age on various aspects of sperm quality and function. These alterations, encompassing both genetic and epigenetic changes, may compromise embryo competency and implantation potential, ultimately leading to reduced pregnancy rates in couples with advanced paternal age [35,40].
This study has several limitations that warrant consideration. The retrospective design and single-center nature inherently limit the generalizability of the findings; however, the study population, consisting of individuals of homogenous Korean ethnicity residing in a single metropolitan port city, may offer valuable insights into a specific demographic within the larger population. This focused approach could serve as a foundation for future multicenter nationwide studies, enhancing the external validity of the findings. Additionally, while the relatively small sample size may have limited the power to detect subtle associations between certain patient factors and FET outcomes, the focus of the study on the first FET cycle minimized confounding factors associated with prior ovarian stimulation or embryo transfer, thereby enhancing the internal validity and clinical relevance of the observed associations [7,8].
Understanding the interplay of specific patient factors within the freeze-all policy is crucial for optimizing patient counseling, treatment planning, and ultimately achieving the best possible outcomes in a highly personalized manner. During the first programmed embryo transfer among aged infertile couples using blastocysts under the antagonist COS protocol with freeze-all policy, male age, and therefore the quality of gametes for fertilization, seems significantly crucial in successful pregnancy, having more importance than calculated female ovarian reserve and embryo storage duration. Such results could encourage infertility specialists to decide whether to perform additional ovum pick-up or move directly into embryo transfer cycles, elaborating on highly individualized COS and embryo transfer plans to maximize pregnancy rates in aged infertile couples. Finally, future research should focus on prospective studies with larger sample sizes to confirm these findings and further investigate the underlying mechanisms by which male age influences pregnancy outcomes in freeze-all cycles and explore interventions to mitigate the negative impact of advanced male age on overall fertility.
Notes
Conflicts of interest
No potential conflict of interest relevant to this article was reported.
Funding
This work was supported by a clinical research grant from Pusan National University Hospital in 2023.
Author contributions
Conceptualization, Data curation, Methodology: all authors; Formal analysis, Funding acquisition, Investigation: HJL, JJ; Project administration, Supervision, Validation: JJ; Resources: EHY; Writing-original draft: HJL; Writing-review & editing: HJL, JJ.