Evaluating the Impact of Female Factors on Specific Stages of Early Embryo Development: A Retrospective Analysis of 1372 IVF Cycles

Introduction

Infertility affects 15% of reproductive-aged couples worldwide.¹ Within these cases, 50% exclusively stem from female factors, including conditions such as diminished ovarian reserve (DOR), polycystic ovary syndrome (PCOS), endometriosis, and tubal or endometrial pathologies. In contrast, male factors solely account for 20%–30%, with the remaining 20%–30% attributed to a combination of both male and female factors.² Regardless of the underlying causes, assisted reproduction technologies (ART) are considered the ultimate intervention in the treatment of infertility. Advancements in ART have catalyzed a global increase in its utilization over the last decade, despite persistent regional disparities in live birth rates.³ Diverse impacts on the clinical pregnancy after in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) treatment are conditioned by eleven critical predictors, including female age, ethnicity, duration of infertility, body mass index (BMI), antral follicle count (AFC), previous pregnancy, cause of infertility, number of oocytes collected, morphology or quality of embryo transferred, day of embryo transfer (ET), and sperm parameters, and the majority of these variables are primarily attributed to female factors.⁴

The prevailing consensus recognizes that the application of ICSI facilitates the selection of competent spermatozoa with higher fertilization potential, thereby substantially reducing the impact of male factors on embryo quality to a certain degree.⁵ Our unpublished data also indicated that female factors exert a more robust impact on embryo quality in ART cycles compared to male factors. Pertaining to male factors, only male age, sperm concentration, and progressive motility were found to have correlations with a limited number of embryological parameters, which include the cleavage rate and Day 3 (D3) high-quality embryo rate.

Early embryo development during ART treatments is a dynamic and precisely fine-tuned process. It fundamentally encompasses sequential transitions through various stages, including metaphase II (MII) oocytes, zygotes, cleavage stage, morulae, and ultimately reaching the blastocyst stage. Despite a general consensus acknowledging that female-associated factors can primarily influence ART outcomes by altering embryo quality, the specific stages of embryo development that these factors act upon remain uncertain. A possible reason for this uncertainty is that research often focuses on the effects on final clinical pregnancy outcomes,⁶ yet there has been lesser emphasis on the combined effect of various female factors at each specific stage of embryo development.

Cluster analysis, a well-established statistical methodology, is employed to categorize entities with analogous traits. In the study, we employed cluster analysis to categorize ART cycles based on the parameter characteristics of early embryo development, and delved deeper into identifying specific female factors that impact these differentiated stages of embryonic development by one-way analysis of variance (ANOVA). In an attempt to minimize the interference of male factors, we confined our attention to those IVF cycles that included male semen parameters within the standard boundaries, thereby excluding all ICSI cycles. The selection of high-quality embryos is pivotal in determining the success rate of IVF.^7,8 Therefore, within the cluster analysis of this study, we classified the included IVF cycles based on the parameters of high-quality embryos across various developmental stages. The objective of this study is to elucidate the combined effect of female parameters on specific stages of embryo development, with the expectation of providing clinical reference to enhance the acquisition of high-quality embryos during IVF treatments.

Materials and Methods

Subjects

The data were obtained from IVF cycles conducted between January 2021 and December 2022 at the Center of Reproductive Medicine of Zhuhai Maternal and Child Health Care Hospital in Guangdong, China. All participating couples were required to meet the following inclusion criteria: 1. an infertility diagnosis, including primary or secondary infertility, PCOS, endometriosis, tubal factors, or idiopathic causes, is confirmed by the inability to achieve pregnancy for at least one year prior to initiating IVF treatments, excluding ICSI treatment cycles; 2. the male partner’s age ≤40 years, accompanied by a semen analysis indicating no detected abnormalities. To mitigate the risk of statistical bias due to scarce quantity of certain controlled ovarian stimulation (COS) protocols, the study only incorporated cycles from the four most commonly adopted protocols at our center. These encompass the long-acting gonadotropin-releasing hormone (GnRH) agonist protocol, GnRH antagonist protocol, mild stimulation protocol, and progestin-primed ovarian stimulation (PPOS) protocol. The choice of protocol was based on patient-specific factors, including age, AMH levels, AFC, and previous ovarian response. The indications for ovarian stimulation protocols were stratified as follows: The GnRH agonist protocol was indicated for patients with preserved ovarian reserve; the GnRH antagonist protocol was reserved for OHSS-prone populations, particularly advanced maternal age (≥35 years) or DOR cases; mild stimulation regimens were implemented for women of advanced age, DOR, or documented poor ovarian response (POR); while the PPOS approach was preferentially administered to normo-ovulatory women, DOR, or POR patients, with particular consideration for socioeconomic constraints. Any cycle missing essential data was excluded from this study. Consequently, 1372 IVF cycles were enrolled in the present study (Figure 1). This retrospective study was conducted in accordance with the Declaration of Helsinki. The Ethics Committee of Longgang District Maternity & Child Healthcare Hospital of Shenzhen City reviewed the study protocol and granted an exemption from ethical review (Approval No: LGFYKYXMLLM-2025-1) under Article 32 of the Chinese Ethical Review Measures for Life Sciences and Medical Research Involving Humans. This exemption was granted because the research utilized de-identified clinical data without sensitive personal identifiers (eg, names, contact details), posed no physical or psychological risks to participants, and had no commercial intent.

Figure 1 Flow chart illustrating the details of the enrolled cycles in the study.

Parameters of the Enrolled Couples

Our unpublished findings attempted to establish the correlations of male factors such as age, BMI, DNA fragmentation index (DFI), and conventional sperm parameters with the quality of embryos in ART cycles. However, we discovered that only male age, sperm concentration, and progressive motility exhibited correlations with a limited set of embryological parameters. To minimize male-factor impact on embryo quality, this study involved only IVF cycles inclusive of younger males (≤ 40 years) displaying normal sperm concentration and progressive motility. Semen parameters were assessed using computer-assisted sperm analysis (CASA), complying with World Health Organization (WHO) standards.

Analyzing female subjects, factors under scrutiny included age, infertility duration, BMI, basal follicular stimulating hormone (bFSH), basal luteinizing hormone (bLH), basal estradiol (bE₂), testosterone (T), FSH/LH ratio, LH/FSH ratio, anti-mullerian hormone (AMH), AFC, COS protocol, and the initial gonadotropin (Gn) dose.

Controlled Ovarian Stimulation Protocol

Long-Acting GnRH Agonist Protocol

The women undergoing the long GnRH-a protocol received a dose of 1.25 mg long-acting GnRH-a (Diphereline, Ipsen, France) subcutaneously on day 18–20 of the previous cycle. Fourteen days later, 150–300 IU recombinant follicle-stimulating hormone (rFSH; Merck Serono, Inc., Darmstadt, Germany) and/or human menopausal gonadotropin (HMG; Lizhu Pharmaceutical, Inc., Guangzhou, China) were provided daily until trigger.

GnRH Antagonist Protocol

GnRH-antagonist protocol was initiated on day 2 or 3 of the cycle with 75–300 IU of rFSH 0.25 mg of GnRH antagonist (Cetrorelix, Merck serono, Switzerland) was given subcutaneously daily until trigger when the leading follicles reached a mean diameter of 14 mm.

Mild Stimulation Protocol

The milder stimulation protocol using Clomiphene Citrate (CC) was initiated on day 3 of the cycle, with a commencement dosage of 100 mg per day. Five days subsequent to the initiation, when the size of the follicles still registered diameters of 8–10mm, the protocol was supplemented with 150–225 IU/d of HMG. Thereafter, dependent upon the ovarian response, the daily dosage of HMG was tailored, varying in the range of 150–300 IU.

PPOS Protocol

Medroxyprogesterone acetate (MPA, Shanghai Xinyi Pharmaceutical Co., China) was given at a dose of 10 mg/day orally in combination with HMG at a dose of 150–225 IU/d from day 2 or 3 of the cycle. The HMG dosage was adjusted according to the serum estradiol (E₂) levels and follicle sizes, while the dose of MPA remained constant.

For all protocols, if three follicles reached a mean diameter of 17 mm or two follicles reached a mean diameter of 18 mm, 250 μg of recombinant human chorionic gonadotropin (r-HCG; Merck Serono, Inc., Darmstadt, Germany) was injected subcutaneously. Oocyte retrieval was performed 36 h after r-HCG injection via vaginal access guided by transvaginal ultrasound. Oocytes were cultured in a tri-gas incubator (37°C, 6% CO₂, 5% O₂) using sequential media (Quinn’s Advantage Fertilization Medium and Blastocyst Medium, SAGE Media). Fertilization was performed using conventional IVF, and embryos were monitored using time-lapse imaging (EmbryoScope^®, Vitrolife). Embryo quality was assessed based on the Istanbul consensus criteria for D3 embryos and the Gardner grading system for D5 blastocysts.^9,10

Evaluation of Embryo Development

Time-lapse imaging (EmbryoScope^®, Vitrolife) was used to evaluate embryo quality, with specific morphokinetic parameters recorded, including tPNf (time of pronuclei fading), t2 (time to 2-cell stage), t5 (time to 5-cell stage), t8 (time to 8-cell stage), and tB (time to blastocyst formation). These parameters were correlated with embryo quality and developmental potential. Initially, we compiled all parameters associated with the continuous stages of early embryonic development during IVF treatments. These parameters encompassed a spectrum from the metaphase II (MII) oocytes to the high-quality D5 blastocysts. The factors under examination included the MII oocyte rate, fertilization rate, two-pronuclear (2PN) fertilization rate, cleavage rate, normal cleavage rate, formation rate of D3 embryos, rate of high-quality D3 embryos, D5 blastocysts formation rate, and the rate of high-quality D5 blastocysts. Recognizing that superior embryos imply a greater probability of successful pregnancy outcomes, our research sought to elucidate the process of obtaining such embryos through the enhancement of female parameters. Consequently, we meticulously selected parameters that correlated with elevated developmental potential at each individual stage. The variables selected for this study, correlated with embryo development, comprised rates of MII oocytes, 2PN fertilization, normal cleavage, high-quality D3 embryos, and high-quality D5 blastocysts.

Statistical Analysis

The statistical analysis of the collected data was carried out utilizing SPSS software, version 22.0 (SPSS, Inc., Chicago, IL). Descriptive variables were cataloged as mean ± standard deviation (SD) if they adhered to a normal distribution; otherwise, they were characterized in terms of the median, 25%, and 75%. Categorical variables were conveyed as frequency and percentage.

We executed a k-means cluster analysis to discern discrete subgroups within the enrolled IVF cycles. This was performed by employing five grouping variables in accordance with the stages of embryo development. These variables were: rates of mature oocytes, normal fertilization, normal cleavage, high-quality D3 embryos, and high-quality blastocysts. Prior to clustering, all five grouping variables were standardized using Z-score normalization to ensure equal weighting in the analysis. We evaluated if these profiles were impacted by female factors, including attributes such as age, infertility duration, BMI, bFSH, bLH, bE₂, T, FSH/LH ratio, LH/FSH ratio, AMH, AFC, the COS protocol, and the initial Gn dose.

To investigate variations in means among these subgroups, a one-way analysis of variance (ANOVA) test was utilized and the least-significant differences (LSD) test was employed. Kruskal–Wallis H-tests were used to compare medians across different groups. For cross-comparison of the categorical variables across the diverse groups, we utilized the Chi-squared test and/or Fisher’s exact test and executed a Bonferroni correction. Univariate and multivariate logistic regression analyses were conducted to identify factors significantly linked to embryo quality, with the latter adjusting for all clinically relevant covariates to mitigate potential confounding effects. A value of P<0.05 was considered statistically significant.

Results

IVF Cycles Clustering According to Embryo Developmental Metrics

In this study, a total of 1372 IVF cycles were scrutinized. Phenotypic clustering was carried out based on features representing five continuous stages of early embryonic development: MII oocyte rate, 2PN fertilization rate, normal cleavage rate, along with the rates of high-quality D3 embryos and D5 blastocysts. The results of the clustering led to the categorization of the IVF patients into five distinct phenotypic groups. The F-test results substantiated the existence of significant variances among these five clusters across all parameters considered (P=0.00). This compelling evidence points towards distinctive attributes inherent in each group, thus delineating the heterogeneity of the IVF patient population.

The distinctive clinical phenotypes observed within groups A, B, C, D and E were characterized by a progressive gradation in early embryo developmental metrics (Table 1 and Figure 2). They were delineated as follows: (1) Group A (low performance; 8.4% or 115/1372 IVF cycles): Characterized by disappointing rates across all parameters, this group had an extremely low yield of MII oocytes rate (5.98%±15.68%). In addition, it manifested neither normal 2PN fertilization nor cleavage embryos, while no high-quality D3 embryos and D5 blastocysts were formed. This group demonstrated sub-optimal developmental potential in the embryos; (2) Group B (high MII oocyte retrieval; accounting for 2.3% or 32/1372 IVF cycles): Though typified by a high number of retrieved mature MII oocytes (95.31%±12.03%), this group exhibited sub-par rates for subsequent 2PN fertilization (19.92%±39.13%), with non-existent normal cleavage (0%) and high-quality embryo development (0%); (3) Group C (high fertilization and cleavage rates with low high-quality embryos rates; 23.0% or 316/1372 IVF cycles): This group demonstrated high MII oocytes (79.59%±23.34%), 2PN fertilization (87.55%±18.65%), and normal cleavage rates (99.37%±4.22%), but a drastic decrease in achieving high-quality D3 embryos (8.14%±13.99%) and D5 blastocysts (2.86%±10.26%); (4) Group D (high quality D3 embryos but low D5 blastocysst formation rates; 29.2% or 400/1372 IVF cycles): This group, characterized by respectable MII oocytes (81.5%±19.98%), normal fertilization (87.57%±17.08%), and cleavage rates (99.37%±4.22%), as well as a high-quality D3 embryos rate (77.53%±21.84%), however, yielded a correspondingly low rate of high-quality D5 blastocysts (5.27%±12.48%); (5) Group E (high performance; 37.1% or 509/1372 IVF cycles): This group exhibited laudable results across all assessed metrics. Not only did it display an optimal MII oocyte rate (84.94%±14.76%), a commendable 2PN fertilization rate (85.95%±13.28%), and a normal cleavage rate (99.26%±2.87%), but it also showcased the formation of high-quality D3 embryos (57.3%±24.52%). Notably, it recorded the highest yield of high-quality D5 blastocysts (77.22%±20.05%), thus distinguishing it from the other clusters.

Table 1 Distribution of IVF Cycles by Cluster Based on Embryo Developmental Metrics

Figure 2 Radar chart comparing embryo developmental metrics across five phenotypic clusters. Group E demonstrates superior performance across all parameters compared to Group A.

Female Factors Impacting Embryonic Development Between Different Clusters of IVF Cycles

The corresponding ANOVA verified statistically significant discrepancies in the following parameters across the five cluster groups: female age (F=42.02, P=0.00), duration of infertility (F=6.21, P=0.00), bFSH (F=25.39, P=0.00), LH/FSH ratio (F=6.25, P=0.00), AMH (F=68.20, P=0.00), AFC (F=74.68, P=0.00), COS protocol (F=203.83, P=0.00), and initial Gn dose (F=10.40, P=0.00). Conversely, no significant differences were apparent among the groups in regard to female BMI, bLH, bE₂, T, and FSH/LH ratio (P>0.05) (Table 2).

Table 2 Baseline Characteristics of Couples Undergoing IVF Treatments Across Five Clusters

IVF cycles of group E exhibited robust performance parameters for embryo development. Factors impacting these superior outcomes in the female partners of this group were examined. Notably, the females in this group were the youngest (32.33±3.77 years), exhibited the lowest bFSH levels (7.43±3.04), the highest AMH (4.95±2.96) and AFC (14.98±7.02) (P<0.05). Although not reaching statistical significance, this group also displayed a relatively shorter infertility duration (3.51±2.58 years), a higher LH/FSH ratio (0.92±0.64), and a lower initial Gn dose (218.8±64.70). The COS protocol in this group primarily incorporated the GnRH antagonist protocol (58.2%) and the PPOS protocol (36.1%).

Inversely, group A exhibited embryo development evaluation indicators that were completely opposites to those of Group E. The early embryo development evaluation parameters were the worst among all the five groups. The female patients in this group were the oldest among the five groups (37.19±4.57 years) (P<0.05), displayed a relatively longer infertility duration (4.82±3.85 years), higher bFSH (10.92±5.64), lower LH/FSH ratio (0.63±0.52), AMH (1.34±1.23), and AFC (5.70±6.14). Primarily, the COS protocol for this group involved GnRH agonist protocol (35.7%) and GnRH antagonist protocol (27.0%).

The embryo developmental metrics of groups B, C, and D exhibited a distinct characteristic of progressive gradation, with the female parameters of these three groups falling between those of groups A and E. Notably, there was a declining trend in female age (36.91±5.88; 35.01±4.73; 34.53±4.73), infertility duration (5.17±4.26; 4.15±3.55; 4.09±3.27), and bFSH (11.99±7.52; 8.92±4.61; 8.38±3.91) across groups B, C and D. Conversely, the LH/FSH ratio (0.64±0.50; 0.78±0.60; 0.86±0.70), AMH (1.54±1.20; 3.08±2.49; 3.24±2.43), and AFC (7.25±4.61; 10.11±5.95; 10.13±5.86) presented an increasing trend. There were no significant differences in the initial Gn dose across the three groups (222.66±77.34; 248.10±73.99; 242.09±72.77). The predominant COS protocol of group B was the GnRH agonist protocol (40.6%), while for Groups C and D, the GnRH antagonist protocol was predominant (46.5%; 48.5% respectively).

Male Factors Impacting Embryonic Development Between Different Clusters of IVF Cycles

Among the three incorporated male parameters, only the male age exhibited noticeable differences across the five groups (F=15.62, P=0.00). Particularly, the male partners in group E were the youngest (34.0±4.10 years), while group A had the oldest males (36.93±4.18 years). The male ages in groups B, C, and D did not exhibit significant disparities (35.88±4.40; 35.48±3.94; 35.16±4.10 years, respectively). Further analysis of sperm parameters, including concentration, progressive motility, and morphology, revealed no significant impact on embryo quality. Both sperm concentration and progressive motility remained similar among the five clusters (P>0.05) (Table 2).

Influences of Various Risk Factors on Early Embryo Development Across Stages

To understand the impact of different risk factors on the various stages of early embryo development, we incorporated factors with statistical significance in ANOVA analysis into binary unconditional logistic regression analysis, thereby identifying the independent risk factors for early embryonic development. As illustrated in Table 3 and Figure 3, the comparative analysis between the high-performance (Group E) and the low-performance (Group A) groups demonstrated that factors such as female age (OR=1.18, 95% CI: 1.07–1.30, P=0.00), AMH (OR=0.40, 95% CI: 0.29–0.56, P<0.001), AFC (OR=0.86, 95% CI: 0.79–0.94, P<0.001), and the initial Gn dose (OR=0.99, 95% CI: 0.99–1.00, P=0.04) independently influence the various stages of early embryo development during IVF treatment cycles. According to the OR values, it can be seen that for every additional year of female age, the likelihood of a decline in the full-stage embryonic developmental potential increases by 18%. For each decrement of 1ng/mL in AMH, the chance of a decline in embryonic developmental potential increases by 60%. A reduction of 1 in AFC would provide a 14% rise in the likelihood of a decrease in embryonic developmental potential. Compared to the aforementioned three risk factors, the influence of the initial Gn dose on the quality of the embryos is relatively small.

Table 3 Analysis of Risk Factors for Affecting Embryo Development During Various Stages of IVF Cycles

Figure 3 Forest plots illustrate key risk factors influencing embryo development across distinct stages of the IVF cycle. (A) Comparative analysis of risk factors between Group A and Group E; (B) Comparative analysis of risk factors between Group D and Group E. Each plot presents adjusted odds ratios (ORs, represented by square markers) with corresponding 95% confidence intervals (CIs, depicted as horizontal lines) for key predictors (from top to bottom), including female age, duration of infertility, bFSH, LH/FSH ratio, AMH, AFC, initial Gn dose, and male age. The vertical dashed line at OR=1 indicates the null effect; factors whose 95% CIs do not cross this line are considered to show statistically significant differences between the compared groups.

We further compared Group E (high performance) and Group D (high-quality D3 embryos but low D5 blastocyst formation rates), aiming to identify independent factors affecting the acquisition of high-quality D5 blastocysts. The findings indicated that female age (OR=1.08, 95% CI: 1.03–1.14, P=0.00), AMH (OR=0.89, 95% CI: 0.83–0.97, P=0.01), AFC (OR=0.91, 95% CI: 0.88–0.94, P<0.001), and LH/FSH ratio (OR=1.37, 95% CI: 1.05–1.80, P=0.02) were the independently risk factors in the acquisition of high-quality D5 blastocysts during IVF treatment cycles. The OR values indicated a substantial influence of the LH/FSH ratio on the blastocyte development stage, with each unit increment in the ratio correlating with a 37% increase in the likelihood of obtaining high-quality D5 blastocysts (Figure 4). In contrast, each additional year of female age diminished the possibility of obtaining high-quality D5 blastocysts by 8%; every decrease of 1ng/mL in AMH presented an 11% reduced likelihood, while each 1 unit reduction in AFC decreased the probability by 9%. Comparisons among Groups B, C, and D showed that despite a continuous change trend in numeric values of risk factors, no statistically significant differences were observed (P>0.05, data not shown).

Figure 4 The dose–response curve depicting the relationship between the LH/FSH ratio and D5 blastocyst quality. A significant positive correlation was observed between LH/FSH ratios and the formation rate of high-quality D5 blastocysts (P < 0.05).

Discussion

Despite significant advancements made in the field of assisted reproduction over the past three decades, the cumulative delivery rate per aspiration remains disappointingly low at only 30.4%.¹¹ A considerable proportion of embryo loss is observed to occur during the preimplantation phases, with merely 50% of embryos cultured in vitro attaining the D6 blastocyst.¹² The course of initial embryo development is regulated by maternal transcripts until replaced by embryonic genomic expression.¹³ Based on our unpublished data, along with findings from various studies, it appears that female factors may play a more significant role during early embryonic development as compared to male-specific factors.⁴ This is particularly applicable among infertile couples, where the male partner displays normal sperm parameters. In this study, the progression of early embryo development, from the oocyte maturity to the D5 blastocyst formation during IVF treatments, was examined. We aimed to elucidate the influence of particular female factors throughout the various phases of early embryogenesis.

In this study, we categorized IVF cycles into five distinct groups based on the continuous stages of early embryo development during IVF procedures. The findings established that certain characteristics such as female age, duration of infertility, bFSH, LH/FSH ratio, AMH, AFC, COS protocol, initial Gn dose, and male age, exert independent influence over various stages of embryonic development. As we move sequentially from group A (low performance) through group E (high performance), parallel with the progression of early embryonic development and improvement in embryo quality, a sustained decrease was noted in female age, infertility duration, bFSH, initial Gn dose, and male age. Conversely, LH/FSH ratio, AMH, and AFC presented a significant ongoing upward trend. Data from the present study compellingly aligned with clinical practice, suggestive of its authenticity and relevance. For a more comprehensive understanding of the independent variables influencing the different stages of embryo development, we performed a supplementary series of unconditional logistic regression analyses between neighboring groups. Although factors like female age, AMH, AFC, and the initial Gn dose can independently affect the overall early embryo development (Group A vs Group E), a consolidated analysis of other adjacent stages (Group A vs Group B; Group B vs Group C; Group C vs Group D) unveiled only numerical discrepancies between the groups but lacked statistical significance (data not shown).

The Comparison between group A and group E revealed that certain variables, including female age, AMH, AFC, and the initial Gn dose appeared to consistently exert an independent influence on early embryo developmental progression. Specifically, advanced female age, decreased AMH, and AFC levels could imply a decline in the ovarian reserve function and oocyte quality, therefore leading to the production of low-performance embryos. The present data is in agreement with the existing clinical consensus, suggesting that AMH, AFC, and age are more potent predictors of ovarian response during ART.^14,15 Additionally, our study suggested that an excessively increased initial Gn dose might impair the developmental potential of early embryos. Research has shown that daily dosages of FSH in the range of 150–225 IU can induce maximal stimulation in presumedly normal responders under 39 years undergoing IVF cycles, increasing FSH to a dosage of 300 IU will not create more oocytes.¹⁶ The initial Gn dose of Group E, which corresponded to 218.81±64.70 IU, lied within the optimal range. This could potentially be associated with the predominant GnRH antagonist protocol within group E, wherein younger females exhibited an optimal ovarian response and a favorable outcome following Gn stimulation.¹⁷

Groups D and E demonstrated comparable success rates pertaining to MII oocyte, 2PN fertilization, normal cleavage, and high-quality D3 embryo. However, a striking disparity was exclusively observed in the rate of high-quality D5 blastocysts, with group E possessing a significantly higher rate (77.22%±20.05%) in comparison to group D (5.27%±12.48%). The current findings suggested that factors such as female age, AMH, AFC, and LH/FSH ratio could potentially influence the transition of high-quality D3 embryos to high-quality D5 blastocysts. An alternate interpretation could be that these variables might have a sustained impact on early embryo development. Upon reaching a significant cumulative effect, they could significantly enhance the production of high-quality blastocysts.

It is widely recognized that basal hormonal levels are pivotal in the evaluation of ovarian stimulation, determining the quality of oocytes, and consequently, embryos.¹⁸ With respect to the current study, only bFSH and LH/FSH ratio were found to associate with early embryo development, while no notable variances were observed among bLH, FSH/LH ratio, bE2 and T at different stages of embryo development (Groups A-E). Further comparative analysis between consecutive stages indicated that bFSH may not be an independent influential factor in early embryo development during IVF treatments, and the significant difference of the LH/FSH ratio was only detected between groups D and E. As enhancement in embryo quality, a numerical decline in bFSH across was noted. It is a widely acknowledge that a cut-off value of bFSH below 10 mIU/mL implies normal ovarian reserve and correlates with favourable pregnancy outcomes.^19,20 Given that all patients enrolled in this study demonstrated a moderate level of bFSH, group A exhibited the highest bFSH level, which equated to 10.92±5.64 mIU/mL. This could possibly explain why there was a negligible difference in the bFSH levels across the different groups. Interestingly, the gradual increasing trend of LH/FSH ratio was found across the five groups whilst the significant difference was only present between groups D and E, suggesting that LH/FSH ratio may be associated with the generation of high-quality Day 5 blastocysts. Recent randomized trials have demonstrated that LH supplementation in patients with LH/FSH ratios <1.0 significantly improves blastocyst formation rates.²¹ Our findings align with these observations, suggesting that maintaining an LH/FSH ratio above 1.0 through tailored gonadotropin dosing could be associated with enhanced embryo developmental competence. The exposure of bovine oocytes to the purified LH preparation can improve IVF and blastocyst development.²² Clinical investigations also revealed that a low LH level could be indicative of hypothalamic dysfunction, potentially leading to suboptimal pregnancy outcomes.²³ Hence, supplementing LH during COS could potentially result in improved outcomes.²¹ As an intrinsic characteristic, the implications of an unexceptionally elevated LH/FSH ratio are principally focus on the pregnancy outcomes of infertile women suffering from PCOS and the LH/FSH ratio is thus considered as an independent risk factor affecting live birth.²⁴ Based on current data, the overwhelming majority of our study population comprised non-PCOS women possessing normal ovarian reserve. Our findings suggest that, among infertile women regardless of the underlying etiology, the LH/FSH ratio may also exert a greater influence on early embryo developmental potential during IVF cycles than bFSH, bLH, or the FSH/LH ratio. However, it is imperative to maintain a balanced interaction between the levels of LH and other basal hormones. Within a certain range, a moderate concentration of LH could positively impact pregnancy outcomes during ART treatments. The ideal LH/FSH ratio merits further exploration by means of a larger sample size study and requires validation in multi-center cohorts.

Aiming to eliminate the confounding effects of the male partner, all male participants involved in the study were established to be ≤40 years. However, findings from the ANOVA test still indicated a substantial age progression in males across the five groups corresponding with declining embryo quality. It is suggested that regardless of gender, age may predominantly govern embryo quality in IVF treatments. Interestingly, the subsequent multivariate logistic regression analysis revealed that male age did not seem to independently influence early embryo development. In contrast to male factors, female factors appear to be the crucial determinants affecting early embryo development, which is in accordance with our previously unpublished data.

However, certain limitations in this study must be acknowledged. First, as a single-center investigation, our findings may lack generalizability despite the inclusion of a substantial cohort (n=1372 IVF cycles) derived from an initial pool of 3340 cycles. Notably, this selection included multiple-treatment cycles from the same couple, introducing a potential bias to our results. Second, while our study focused on preimplantation embryo quality, it did not assess downstream clinical outcomes, including implantation rates, miscarriage rates, live birth rates, or neonatal outcomes. Such data would provide a more comprehensive understanding of how female characteristics influence not only early embryogenesis but also long-term reproductive success. That said, high-quality embryo selection has been established as a robust surrogate endpoint for live birth in IVF cycles, particularly in studies examining ovarian response and embryo developmental competence.²⁵ Third, we cannot exclude the possibility that the observed associations may be confounded by unmeasured factors, including ovarian reserve heterogeneity, lifestyle variables, or genetic differences. Consequently, the proposed LH/FSH ratio threshold should be interpreted with caution, given the continuous nature of our analytical approach. Future multi-center studies with extended follow-up periods are needed to validate these findings and explore their implications for longitudinal pregnancy outcomes.

Conclusion

The current study conducted a cluster analysis using five indicators representative of distinct stages of early embryo development to classify IVF cycles based on distinguishable embryonic quality and demonstrated the influence of specific female attributes at various stages of early embryo development. Notable determinants, female age, AMH, AFC, and initial Gn dose, were identified to exert profound influence on the overall early embryo development process. Importantly, they govern the achievement of high-quality embryos during IVF treatments. Furthermore, parameters including female age, AMH, AFC, and LH/FSH ratio appear to play a significant role in steering the transition from high-quality D3 embryos to superior D5 blastocysts. It is reasonable to suggest the LH/FSH ratio as a valuable early hormonal indicator for predicting blastocyst quality and the subsequent pregnancy outcome in ART treatments, thus necessitating further exploration for defining its optimal clinical threshold.