Clinical and laboratory characteristics of patients with embryonic arrest in the early embryonic period of in vitro fertilization programs.

Oocytes obtained in in vitro fertilization (IVF) programs exhibit varying fertilization potential, and embryos develop differently during the cleavage period. Consequently, only 5% of oocytes produce embryos, and transfer of these embryos results in live births [1]. The relationship between oocytes and embryo quality has been confirmed in several studies, demonstrating that early embryo development failures are predominantly due to maternal factors [2–5]. Numerous defects in embryo maturation arise even before fertilization within the corresponding oocyte and largely depend on intracellular mechanisms of epigenetic modifications of the genome [6–8]. The process of acquiring oocyte competence remains poorly understood and develops during oogenesis [9]. Of interest is the study of the mechanisms leading to embryonic arrest, where an embryo suitable for transfer stops developing. Depending on the stage at which embryonic arrest occurs, we may observe the inability of the zygote to initiate the cleavage process, cessation of cleavage on the 2nd or 3rd day (at the two- to eight-cell embryo stage), failure to form a morula, and subsequently a blastocyst. The presented publications analyzed possible reasons for the arrest of embryo development at different cleavage stages [10–12].

Several studies have summarized the molecular factors that contribute to embryonic arrest at various stages. Embryonic arrest is a multifactorial phenomenon that is influenced by both internal and external factors. The investigation of cause-and-effect relationships determining embryonic arrest revealed maternal and/or paternal factors, embryonic metabolism activity, disruptions in crucial molecular pathways halting the developmental process, genetic abnormalities, and laboratory conditions as contributing factors [13–15].

Fetal factors associated with embryonic arrest include gene variations, mitochondrial DNA copy number, methylation patterns, chromosomal abnormalities, metabolic profiles, and morphological features. Parental factors include gene variations, protein expression levels, and etiology of infertility [16–18]. Despite an extensive analysis of the literature by the authors, no specific causes of embryonic arrest have been established. All of these factors potentially participate in the process, but how they do so remain unclear, given the diversity and heterogeneity of the causes of embryonic arrest during early embryogenesis, as well as their relationships [19].

However, in clinical practice, it is crucial to establish criteria that allow the prediction of the stage of embryonic development during patient examinations with varying degrees of reliability.

This study aimed to investigate the clinical and hormonal parameters and outcomes of in vitro fertilization (IVF) programs in patients with impaired embryo cleavage compared to those in patients with embryos suitable for transfer.

Materials and methods

This observational case-control clinical study included 770 patients who underwent IVF infertility treatment.

To achieve the goal of the study, all patients who met the inclusion and non-inclusion criteria were divided into two groups: Group 1 (study group) consisted of 287 patients whose embryos arrested in developmentat different stages of cleavage during IVF, and Group 2 (control group) included 483 patients with a blastocyst suitable for transfer on the 5th or 6th day of culture.

A total of 770 IVF cycles were performed at the F. Paulsen Research and Educational Center for ART in the Clinical Department of V.I. Kulakov National Medical Research Center for Obstetrics, Gynecology and Perinatology, Ministry of Health of the Russian Federation in 2022 in patients of the study groups.

Additionally, representative groups of patients were identified (groups 1a and 2a), which were comparable in age, state of ovarian reserve, and number of oocytes and zygotes obtained. Each group comprised 37 patients. In Group 1a, all embryos arrested in development, and patients in Group 2a received embryos suitable for transfer.

The criteria for inclusion in the study groups were absence of pregnancy for one year or more, normal karyotype of both spouses, and indications for achieving pregnancy using IVF.

The criteria for non-inclusion in the study groups were the use of donor gametes in the studied cycle of assisted reproductive technologies (ART), cryopreserved embryo transfer programs, all conditions that are contraindications to ART, and achieving pregnancy in accordance with the order of the Ministry of Health of Russia dated July 31, 2020, No. 803n.

All married couples underwent full clinical and laboratory examinations in accordance with Order No. 803n of the Russian Ministry of Health, dated July 31, 2020.

Stimulation of ovarian function was performed with gonadotropins (recombinant follicle-stimulating hormone (FSH) and/or human menopausal gonadotropin) according to standard protocols with gonadotropin-releasing hormone agonists or antagonists. Urinary or recombinant forms of human chorionic gonadotropin were used as ovulation triggers. Ultrasound monitoring was performed three to five times during ovarian stimulation. Transvaginal puncture of the follicles was performed 36–37 h after ovulation-trigger administration. Fertilization of aspirated oocytes was performed using IVF or intracytoplasmic sperm injection (ICSI). All embryos were cultured in individual drops of one-step culture medium (G-TL, Vitrolife) in multigas incubators. The quality of the resulting embryos was assessed on the 2nd, 3rd, 4th, 5th and 6th days of culture based on morphological criteria in accordance with Gardner’s grading system (ESHRE 2011 modified classification by D. Gardner) [20].

The study was reviewed and approved by the Research Ethics Committee of V.I. Kulakov National Medical Research Center for Obstetrics, Gynecology and Perinatology, Ministry of Health of the Russian Federation

Statistical analysis

Statistical analysis and graphing were performed using GraphPad Prism v6 (GraphPad Software Inc., USA). The statistical programming language R was used to construct the logistic regression. For qualitative data, counts and percentages were used (N (%)), and Fisher's exact test was used. The distribution of continuous variables was tested for normality using the Shapiro-Wilk test. Numerical variables that were not normally distributed were summarized as medians and interquartile range (Me (Q1; Q3)), and the Mann–Whitney test was used. In order to assess the quality of binary classification, ROC analysis was performed. When performing the ROC analysis, the cutoff threshold, area under the ROC curve (AUC), sensitivity, and specificity were calculated. Regression analysis was performed to determine the relationship between the dependent variables and one or more independent variables. The statistical programming language R was used to build the logistic regression. Multicollinearity was assessed using the Spearman’s correlation test. Differences were considered statistically significant at p<0.05.

Results

Group 1 (study group) included 287 women with a mean age of 38 (34; 41) years, from whom the IVF program obtained three or more (3; 6) mature oocytes and three or more (3; 4) zygotes, but all embryos were arrested in development.

Group 2 (control group) consisted of 483 patients with a mean age of 35 (32; 39) years; eight (5; 13) mature oocytes, six (4; 10) zygotes, and three (2; 6) blastocysts were obtained in the IVF program.

When comparing the clinical characteristics of patients in the study groups, there were no differences in the frequency of primary/secondary infertility, duration of infertility, or number of previous ART attempts (Table 1).

At the same time, analysis of the reproductive function of patients with a history of pregnancy showed that in group 2, 88/234 (37.6%) women had childbirth, 107/234 (45.7%) pregnancies were terminated at the request of the woman, and only 39/234 (16.0%) had spontaneous miscarriages. Among Group 1 patients, only 19/151 (12.5%, p<0.001) had a live birth, 41/151 (27.2%, p<0.001) had an abortion at the patient's request, and 90/151 (59.6%, p<0.001) had at least one missed miscarriage.

A thorough analysis of the causes of infertility in the study groups showed that the tubo-peritoneal factor of infertility occurred in 256/483 (53.0%) patients in group 2 and in 65/287 (22.7%) patients in group 1; infertility of unknown origin was diagnosed in 115/287 (40.1%) patients in group 1 and 58/483 (12. Infertility associated with endometriosis was diagnosed in 46/287 (16.0%) patients in Group 1 and 43/483 (8.9%) women in Group 2. The incidence of the combined forms of infertility and infertility associated with male factors did not differ between the groups (Table 2).

Therefore, the clinical factors associated with impaired embryo development included late reproductive age 38 years (34.5;41) in group 1 and 35 years (32;39) in group 2; missed miscarriage in 59.6% of women in group 1 and 16% in group 2 among patients with a history of pregnancy, infertility of unknown origin in 40.1% and 12%, respectively; and extragenital endometriosis in 16% of women in group 1 and 8.9% in group 2.

Objective confirmation of a possible impairment of embryo development can be the analysis of previous IVF attempts when an embryo suitable for transfer was not obtained. In the present study, almost half of the patients in the study group had a history of repeated IVF attempts, and none of them produced an embryo suitable for transfer.

Significant differences were observed in the state of ovarian reserve of the study participants (Table 3). Patients with embryo development disorders had diminished ovarian reserve rates compared with in those the control group.

The data obtained confirmed the opinion of many experts that late age is a decisive factor in determining not only the quantity but also the quality of the oocytes obtained [21]. However, it is known that ovarian reserve, in particular the anti-Müllerian hormone (AMH) level, is a predictor of the number of oocytes retrieved but is not associated with their quality and pregnancy rate [22].

The number of oocytes obtained in the study and control groups was analyzed (Table 4).

All results are presented for each IVF attempt. As shown in the table, the probability of embryonic arrest in all embryos was directly related to the number of preovulatory follicles and the number of oocytes retrieved.                     

The fertilization rate (2PN) in the study groups was 68.4 and 76.3%, respectively.

The presented data demonstrate a relationship between the number of oocytes retrieved and embryonic arrest. Thus, in the study group the mean number of oocytes was 3, while in the control group it was 10; 71.8% of attempts in the study group were characterized by a “poor response” of the ovaries, while in the control group there were 24.4% of such patients. The results obtained were consistent with the clinical and hormonal characteristics of the patients; in the study group, there were more older women with diminished ovarian reserve.

Using a control group of patients as an example, blastocyst yield was calculated as a function of the number of mature oocytes obtained. Blastocyst yield was calculated by dividing the number of blastocysts by the number of oocytes obtained.

Considering that the average percentage of blastulation per oocyte was 31.1%, it was possible to calculate the probability of embryonic arrest when obtaining a small number of oocytes, assuming that this process was random.

To clarify the influence of individual predictors, logistic regression was performed, and the results of constructing the dependence of the parameters are presented in Table 5. When checking for multicollinearity (by correlation analysis), only a weak negative correlation was found between age and AMH levels (r=-0.172, p=0.004). Embryonic arrest was used as an independent variable. Age, AMH level, type of infertility, duration of infertility, and infertility factors were used as the dependent variables. To eliminate predictors in a stepwise manner, we used the stepAIC function of the MASS package. The regression equation fits the model, as follows:

Y = exp(z)/(1+exp(z));

z = Intercept+0.433*(Age, years)-0.073*(AMH, ng/ml);

Intercept=2,340.

Thus, only patient age and AMH level were significant factors. To determine the reliability of these signs, a receiver operating characteristic (ROC) curve was constructed, the ROC curve is shown in Figure 1. The area under the curve (AUC) was 0.739, the sensitivity was 83.9%, and the specificity was 48.4% (OR = 5.57; 95% CI 3.88–8.01).

The number of oocytes and zygotes obtained are important indicators of the effectiveness of the embryological phase of IVF programs (Fig. 2).

Without denying the dependence of blastocyst formation on the number of oocytes and zygotes, this fact alone cannot be completely justified.

In the present study, in 28.2% of patients with embryonic arrest, a sufficient number of mature oocytes (six or more) were obtained, comparable to the control group, which dictates the need to search for other factors that primarily determine the quality of gametes, and not only their quantity.

We compared the clinical and laboratory characteristics of selected age-representative patients in the study and control groups (Fig. 3).

In the subsequent analysis, we excluded the factors of age and low number of oocytes received, comparing patients under 36 years of age with more than five oocytes and at least three zygotes received, to identify factors associated with embryonic arrest in the group of patients in whom this outcome was least expected (Table 6) and women with similar characteristics but with at least one blastocyst suitable for transfer. Group 1a (study group) consisted of 37 women aged 32 (30.5; 35.5) years, of whom eight (6; 12) mature oocytes and four (3; 6) zygotes were obtained in the IVF program, but all embryos were arrested in development. Group 2a (control) consisted of 37 patients aged 32 (30.5; 35.5) years, from whom 10 (8; 18) mature oocytes, 8 (5; 10) zygotes, and 4 (3; 5) blastocysts were obtained.         

Despite the comparability of patients, the following differences were identified:

1. Patients with embryonic arrest had a longer duration of infertility (5 (3; 8) years versus 3 (2; 4) years, respectively; p=0.035).
2. Primary infertility occurred in 54.1% and 27.0% of patients in groups Ia and IIa, respectively (p=0.032).
3. A severe form of pathozoospermia, in combination with other infertility factors, was present in 17/37 (46.0%) and 6/37 (16.2%) spouses, respectively (p=0.011).
4. Infertility of unknown origin was diagnosed in 24.3% of the cases in group Ia versus 5.4% in group IIa (p=0.046).
5. Extragenital endometriosis was detected in 25.9% and 3.7% of the patients (p=0.050).
6. In group Ia, 88.2% of the patients with secondary infertility had a history of missed miscarriages versus 44.4% in group IIa (p=0.005).
7. Patients in group Ia had at least three previous unsuccessful IVF attempts (three (0; 4), and in group IIa, only one (0; 1)) (p=0.048).

Discussion

Embryos arrested in development obtained during IVF can be classified as a physiological phenomenon, a type of natural selection, where non-viable embryos do not develop. It is believed that the normal blastulation rate is 40–45% of the number of zygotes obtained [23]. However, in some cases, it is not possible to obtain an embryo suitable for transfer, and the development of all resulting embryos stops. Unfortunately, this phenomenon is aggravated by the older age of patients and the small number of oocytes obtained. This study confirms the well-known fact that the clinical factors causing impaired embryo development are the woman’s age and the associated decrease in ovarian reserve, as diagnosed by the level of AMH. This study aimed to determine whether blastocyst formation depends on the number of oocytes and zygotes obtained. The identified patterns can be useful in the practical work of a doctor, allowing assessment of the possible number of oocytes obtained during ovarian stimulation to predict the likelihood of a blastocyst suitable for transfer to the uterine cavity. Moreover, if the prognosis for embryo fragmentation is unfavorable, it is possible to change the patient’s management tactics and carry out an early embryo transfer on the 2nd or 3rd day of development, or perhaps at the zygote stage. Few studies have addressed this topic [24–26]. Situations where at least five mature oocytes and at least three zygotes are obtained, and the embryos stop developing, are difficult to explain and make clinical decisions. In this study, we compared the clinical and anamnestic data of women representative of age, state of ovarian reserve, number of mature oocytes, and zygotes obtained in women with impaired parameters of early embryogenesis and patients who had blastocysts suitable for transfer. The observed differences include a greater number of missed miscarriages in group 1a than in group 2a, the presence of infertility of unknown origin, and a high proportion of detected extragenital endometriosis. These factors may be the possible clinical causes of impaired embryo development, as confirmed by the results of previous IVF attempts in patients, where none of the attempts yielded embryos suitable for transfer. The presented results emphasize the necessity for a thorough analysis of the anamnestic and clinical characteristics of patients and the results of previous IVF programs to determine tactics for achieving pregnancy.

Conclusion

The development of embryos in the early stages of culture and the frequency of blastocyst formation depend on the number of mature oocytes and zygotes obtained, which in turn are determined by the age of the patient and the state of the ovarian reserve. A history of missed miscarriages, infertility of unknown origin, extragenital endometriosis, and impaired embryonic development in previous IVF attempts are clinical and anamnestic characteristics that may be risk factors for embryonic arrest in young women, even if at least five mature oocytes are obtained and at least three zygotes are present.