The impact of completely shrunken blastocysts on clinical outcomes in vitrified-warmed single blastocyst transfer cycles.

Between January 2015 and December 2023, 2276 single blastocyst transfers were performed through frozen-thawed embryo transfer (FET). These were categorized into 104 completely CSBT cycles for analysis and 2172 REBT cycles as controls, based on blastocyst re-expansion status within 2–4 h post-warming. As shown in Table  1 , no significant differences were observed between the two groups in maternal age, body mass index (BMI), duration of infertility, basal FSH levels, anti-Müllerian hormone (AMH) levels, or endometrial thickness. However, the CSBT group demonstrated significantly lower rates of excellent-quality blastocysts (0% vs. 3.9%; P  = 0.031), good-quality blastocysts (6.7% vs. 18.7%; P  = 0.002), and average-quality blastocysts (50.0% vs. 62.5%; P  = 0.010), along with a higher proportion of poor-quality blastocysts (43.3% vs. 14.9%; P  < 0.001) and day 5 blastocysts (58.7% vs. 84.2%; P  < 0.001) compared to the REBT group. Statistically significant differences were also observed between the two groups in primary causes of infertility, proportion of blastocysts with expansion degree 4, and preimplantation genetic testing (PGT) treatment ( P  < 0.05). To control for potential confounding effects of these differences on clinical outcomes, propensity score matching was performed to balance baseline characteristics between the two groups. Not all the key characteristics were similar between groups ( P  > 0.05) after propensity score matching, as the primary causes of infertility and the proportion of blastocyst expansion degree of 4 remained significantly different. Table 1 Baseline characteristics of patients before and after the propensity score matching. Characteristics CSBT group Before matching After matching REBT group P REBT group P Cycles (n) 104 2172 104 Maternal age (years) 31.5 ± 5.7 31.1 ± 4.8 0.447 30.8 ± 5.0 0.390 BMI (kg/m 2 ) 23.8 ± 3.5 23.3 ± 3.6 0.116 23.3 ± 3.6 0.293 Duration of infertility (years) 4.5 ± 3.1 4.4 ± 3.3 0.796 4.9 ± 3.8 0.385 Basal FSH (mIU/ml) 7.7 ± 3.3 7.7 ± 3.0 0.853 8.2 ± 4.2 0.331 AMH (ng/ml) 4.4 ± 3.3 5.1 ± 3.7 0.067 4.5 ± 3.3 0.822 Endometrial thickness (mm) 8.9 ± 1.8 9.0 ± 1.7 0.700 8.9 ± 1.7 0.907 Primary infertility (%) 40.4 (42) 43.3 (940) 0.561 52.9 (55) 0.071 Primary reasons for infertility (%) Tubal factor 67.3 (70) 68.0 (1478) 0.874 63.5 (66) 0.560 Polycystic ovary syndrome 13.5 (14) 10.1 (220) 0.274 9.6 (10) 0.385 Endometriosis 5.8 (6) 1.38 (30) < 0.001 0.9 (1) 0.119 Male factor 4.8 (5) 5.5 (119) 0.768 2.9 (3) 0.471 Others 8.7 (9) 15.0 (325) 0.076 23.1 (24) 0.040 Natural cycles (%) 28.8 (30) 26.2 (569) 0.549 33.7 (35) 0.454 Blastocyst quality: excellent 0 (0) 3.9 (85) 0.031 3.8 (4) 0.121 Good 6.7 (7) 18.7 (406) 0.002 14.4 (15) 0.071 Average 50.0 (52) 62.5 (1358) 0.010 43.3 (45) 0.331 Poor 43.3(45) 14.9 (323) < 0.001 38.5 (40) 0.481 Blastocyst expansion degree of 4 (%) 98.1 (102) 87.0 (1889) 0.001 90.4 (94) 0.017 Day 5 blastocyst (%) 58.7 (61) 84.2 (1829) < 0.001 58.7 (61) 1.000 PGT euploid transfer (%) 13.5 (14) 4.8 (105) 0.001 5.8 (6) 0.092 CSBT, completely shrunken blastocyst transfer; REBT, re-expanded blastocyst transfer; BMI, body mass index; AMH, anti-Müllerian hormone; FSH, follicle-stimulating hormone; PGT, preimplantation genetic testing. Baseline characteristics of patients before and after the propensity score matching. CSBT, completely shrunken blastocyst transfer; REBT, re-expanded blastocyst transfer; BMI, body mass index; AMH, anti-Müllerian hormone; FSH, follicle-stimulating hormone; PGT, preimplantation genetic testing. We compared the clinical outcomes of patients receiving CSBT to those receiving REBT. Table  2 shows that the CSBT group had a lower clinical pregnancy rate (28.8% versus 59.3%, P  < 0.001; 28.8% versus 61.5%, P  < 0.001), ongoing pregnancy rate (22.1% versus 50.4%, P  < 0.001; 22.1% versus 52.9%, P  < 0.001), and live birth rate (20.2% versus 48.5%, P  < 0.001; 20.2% versus 50.0%, P  < 0.001) than the REBT group before and after matching. The other variables studied, such as miscarriage rate, gestational age at birth, and percentage of pre-term and post-term deliveries, showed no statistically significant differences ( P  > 0.05). Table 2 Clinical outcomes of patients before and after the propensity score matching. Characteristics CSBT group Before matching After matching REBT group P REBT group P Cycles (n) 104 2172 104 Clinical pregnancy rate (%) 28.8 (30/104) 59.3 (1289/2172) < 0.001 61.5 (64/104) < 0.001 Ongoing pregnancy rate (%) 22.1 (23/104) 50.4 (1095/2172) < 0.001 52.9 (55/104) < 0.001 Miscarriage rate (%) 30 (9/30) 18.2 (235/1289) 0.101 18.8 (12/64) 0.222 Live birth rate (%) 20.2 (21/104) 48.5 (1054/2172) < 0.001 50.0 (52/104) < 0.001 Gestational age (GA) at birth (days) 275.4 ± 8.2 270.1 ± 14.3 0.088 271.9 ± 9.3 0.137 Pre-term deliveries ( 41 weeks GA) (%) 4.8 (1/21) 0.6 (6/1054) 0.129 0 (0/52) 0.288 Clinical outcomes of patients before and after the propensity score matching. To investigate the effect of patients’ baseline and cycle characteristics on pregnancy in CSBT cycles, all CSBT cycles were selected (Tables  3 and 4 ) and compare the differences between their fresh and FET cycles. In the pregnancy group, we found that maternal age (29.4 ± 4.5 versus 32.4 ± 6.0, P =  0.007) and basal FSH (6.8 ± 2.0 versus 8.0 ± 3.7, P  = 0.029) were significantly lower than in the non-pregnancy group. There was no significant difference in the BMI, duration of infertility, AMH, endometrial thickness, and primary reasons for infertility ( P  > 0.05) (Table  3 ). Table  4 shows the characteristics of fresh and FET cycles of thawed blastocysts. Data from fresh cycles showed that blastocysts in the pregnancy group were formed from more good-quality embryos by day 3 than in the non-pregnancy group (63.3% versus 32.4%, P =  0.004). The FET data showed that the pregnancy group had a higher percentage of day 5 blastocysts than the non-pregnancy group (80.0% versus 50.0%, P =  0.005). The COS strategy (agonist/antagonist), insemination method (IVF/ICSI), mean fertilization rate, mean blastulation rate, and the quality of blastocyst transferred, etc., were not significantly different between the two groups ( P  > 0.05). Table 3 Baseline characteristics of patients were compared between pregnancy and non-pregnancy groups in completely shrunken blastocyst transfer (CSBT) cycles. Pregnancy Non-pregnancy P Cycle (n) 30 74 Maternal age (years) 29.4 ± 4.5 32.4 ± 6.0 0.007 BMI (kg/m 2 ) 23.3 ± 3.6 24.1 ± 3.4 0.306 Duration of infertility (years) 3.8 ± 2.2 4.8 ± 3.4 0.145 Basal FSH (mIU/ml) 6.8 ± 2.0 8.0 ± 3.7 0.029 AMH (ng/ml) 4.1 ± 2.5 4.6 ± 3.5 0.518 Endometrial thickness (mm) 8.7 ± 1.4 9.0 ± 2.0 0.371 Primary infertility (%) 46.7 (14) 37.8 (28) 0.406 Primary reasons for infertility (%) Tubal factor 56.7 (17) 71.6 (53) 0.141 Polycystic ovary syndrome 20.0 (6) 10.8 (8) 0.214 Endometriosis 3.3 (1) 6.8 (5) 0.670 Male factor 10.0 (3) 2.7 (2) 0.143 Others 10.0 (3) 8.1 (6) 0.715 BMI, body mass index; AMH, anti-Müllerian hormone; FSH, follicle-stimulating hormone. Baseline characteristics of patients were compared between pregnancy and non-pregnancy groups in completely shrunken blastocyst transfer (CSBT) cycles. BMI, body mass index; AMH, anti-Müllerian hormone; FSH, follicle-stimulating hormone. Table 4 Cycle characteristics of patients were compared between pregnancy and non-pregnancy groups in CSBT cycles. Pregnancy Non-pregnancy P Cycle (n) 30 74 From fresh cycle Agonist / antagonist ratio 0.5 (10/20) 0.4 (22/52) 0.718 IVF / ICSI ratio 1.5 (18/12) 1.4 (43/31) 0.859 Mean fertilization rate 80.5 ± 15.0 73.7 ± 20.7 0.108 Mean blastulation rate 56.3 ± 26.2 55.8 ± 26.7 0.936 Blastocyst formed from 2PN embryo (%) 80.0 (24) 90.5 (67) 0.141 Blastocysts derived from good-quality day 3 embryos (%) 63.3 (19) 32.4 (24) 0.004 PGT euploid transfer (%) 20.0 (6) 10.8 (8) 0.214 From FET cycle Percentage of natural cycles (%) 33.3 (10) 27.0 (20) 0.520 Blastocyst quality: excellent 0 (0) 0 (0) N/A Good 13.3 (4) 4.1 (3) 0.104 Average 56.7 (17) 47.3 (35) 0.387 Poor 30.0 (9) 48.6 (36) 0.063 Percentage of day 5 blastocyst (%) 80.0 (24) 50.0 (37) 0.005 IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; PN, pronucleus; PGT, preimplantation genetic testing; FET, frozen-thawed embryo transfer, N/A, not applicable. Cycle characteristics of patients were compared between pregnancy and non-pregnancy groups in CSBT cycles. IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; PN, pronucleus; PGT, preimplantation genetic testing; FET, frozen-thawed embryo transfer, N/A, not applicable. To account for potential confounding factors, we performed logistic regression analyses to evaluate pregnancy-related factors in CSBT cycles (Table  5 ). Only day 5 blastocysts demonstrated a significant positive association with pregnancy outcomes (adjusted OR 3.062, 95% CI 1.077–8.704, P  = 0.036) when compared to day 6 blastocysts. In contrast, other variables, including maternal age, basal FSH levels, and day 3 embryo quality, showed no significant associations in the logistic regression analyses. Table 5 Logistic regression analysis was performed to assess the likelihood of clinical pregnancy in CSBT cycles. Univariate OR (95% CI) Multivariate OR (95% CI) Adjusted P Age (years): ≥ 35 Refer 0.086 < 35 3.957 (1.249–12.530) 2.908 (0.861–9.818) Basal FSH (mIU/ml): ≥ 10 Refer 0.512 < 10 2.100 (0.557–7.917) 1.607 (0.389–6.639) Embryo quality on day 3: AP Refer 0.176 Good 1.895 (0.782–4.593) 1.912 (0.748–4.887) Days to reach expanded blastocyst: day 6 Refer 0.036 day 5 4.000 (1.466–10.916) 3.062 (1.077–8.704) FSH, follicle-stimulating hormone, AP, average quality combined with poor quality. Logistic regression analysis was performed to assess the likelihood of clinical pregnancy in CSBT cycles. FSH, follicle-stimulating hormone, AP, average quality combined with poor quality.

This retrospective study analyzed medical records of patients who underwent single blastocyst transfer cycles at the Department of Reproductive Medicine Center, 901st Hospital of the Joint Logistics Support Force of PLA, between January 2015 and December 2023. A total of 2324 frozen embryo transfer cycles were included in the study. Among these, 152 blastocysts failed to re-expand, with 104 being transferred and 48 discarded due to degeneration. The analysis comprised 104 completely shrunken blastocyst transfer (CSBT) cycles and 2172 re-expanded blastocyst transfer (REBT) cycles, categorized based on blastocyst re-expansion status within 2–4 h post-thawing. The study flow chart is presented in Fig.  1 . Patients aged 21–49 years were included, with infertility diagnoses including tubal factor, polycystic ovary syndrome, endometriosis, male factor, or other causes. Exclusion criteria comprised conditions potentially affecting embryo implantation, such as untreated hydrosalpinx, intrauterine adhesions, submucosal fibroids, uterine polyps, or chromosomal abnormalities. Fig. 1 The flow chart of the study. The flow chart of the study. Ethical approval was obtained from the 901st Hospital of the Joint Logistics Support Force of the PLA (IRB number: 202409001). All methods are conducted in accordance with the principles of the Declaration of Helsinki and the relevant guidelines and regulations. The Ethics Committees of the 901st Hospital of the Joint Logistics Support Force of PLA waived informed consent for this retrospective study. Previous research have documented our center’s controlled ovarian stimulation (COS) strategy and IVF/ICSI procedures 26 . Treatment protocols included either a gonadotropin-releasing hormone (GnRH) antagonist or a GnRH agonist. On the day of oocyte retrieval, conventional IVF or ICSI was used in accordance with standard insemination protocols. Fertilized oocytes were transferred to G1-plus media (Vitrolife, Sweden) covered with paraffin oil (OvooilTM, Vitrolife) and cultivated embryos to day 3 at 37 °C, 6% CO2, and 5% O2 conditions (Planer BT37, Origio, USA). Afterwards, utilizable embryos were placed in G2-plus media (Vitrolife, Sweden) and cultivated until the blastocyst stage. Fertilization and embryo morphology were assessed using an inverted microscope with Hoffman modulation contrast microscopy (Olympus IX71, Japan) at 250× magnification. At 16–18 h after insemination, oocytes containing two pro-nuclei (2PN) and two polar bodies were regarded as having normal fertilization. On Day 3, embryos received evaluations based on the Istanbul consensus 27 . Grade A embryos were classified as good quality, grade B embryos as average quality, grade C embryos as poor quality, and grade D embryos were eliminated. The Gardner scoring criteria were used to evaluate the morphology of the blastocysts on days 5 and 6 1 . The morphological quality of the inner cell mass (ICM) and TE was graded as follows: ICM (A = many tightly packed cells, B = several loosely packed cells, C = very few cells) and TE (A = many cells forming a cohesive epithelium, B = several cells organized in a loose epithelium, C = very few large cells). Based on these grades, blastocyst morphological quality was classified into four groups: excellent (AA), good (AB or BA), average (AC, CA, or BB), and poor (BC or CB). To minimize inter-observer variation in embryo grading, blastocysts are usually assessed by at least two trained embryologists. The blastocyst vitrification and warming protocol followed the manufacturer’s recommendations (Kitazato, Japan). Prior to vitrification, the blastocyst was artificially shrunk by laser (Octax, MTG, Germany) and then placed in an equilibration solution (ES) containing 7.5% ethylene glycol (EG) and 7.5% dimethyl sulfoxide (DMSO) for 8–10 min before being transferred to a vitrification solution (VS) containing 15% EG, 15% DMSO, and 0.5M sucrose for 1 min. The blastocyst was then loaded onto the Cryotop strip (Kitazato, Japan) and immediately immersed in liquid nitrogen. For the blastocyst thawing technique, the cryotop was immediately transferred from liquid nitrogen to the 37 °C thawing solution (TS) of 1 M sucrose for approximately 1 min, and then the embryo was transferred to the diluent solution (DS) of 0.5 M sucrose for 3 min at room temperature, and the embryo was washed twice with the wash solution (WS) for 5 min each. The embryos were then transferred to G2-plus culture medium for observation and photography under an inverted microscope. After warming, the embryos were immediately subjected to assisted hatching; irradiation began at one point and continued until a quarter of the zona pellucida was peeled off, with the exception of blastocysts treated with preimplantation genetic testing (PGT) or expansion degree of 5 prior to vitrification. Finally, the embryo was cultured in G2-plus media (it contains 5% human serum albumin) for at least 2 h (2 h for the vast majority, but no more than 4 h) prior to ET, and blastocoel re-expansion score was recorded. In our center, the criteria for blastocyst transfer after warming required the absence of evident signs of necrosis, degeneration, or regression (such as dark granules or extensive dark areas) prior to transfer. In our laboratory, blastocyst re-expansion after thawing is based on two parameters: (1) the degree of expansion within the blastocyst cavity and (2) the extent of the perivitelline space between the zona pellucida and TE cells. Completely shrunken blastocysts were defined as those in which the blastocyst cavity was entirely or nearly invisible before embryo transfer. All enrolled participants underwent hormone replacement therapy (HRT) or natural cycles (NC) to prepare the endometrium for FET 26 . The embryo transfer was done using a Wallace catheter and abdominal ultrasonography guidance. Following transplantation, all patients were given oral dydrogesterone 10 mg twice daily for luteal support. The primary outcome was clinical and ongoing pregnancies, followed by miscarriage and live birth. Serum human chorionic gonadotropin (hCG) levels were ≥ 50 IU/L 14 days following blastocyst transfer, indicating biochemical pregnancy. Thirty days after embryo transfer, clinical pregnancy was verified by a fetal heartbeat or gestational sac. Miscarriage was defined as pregnancy loss following the identification of a fetal heartbeat on ultrasonography; patients with ectopic pregnancy were excluded from estimating the miscarriage rate. Ongoing pregnancy was defined as the presence of a fetal heartbeat after 12 weeks of gestation. A live birth was defined as a newborn born alive after 24 weeks of gestation. All statistical calculations were performed using the Statistical Package for the Social Sciences (SPSS) software, version 26.0 (IBM Corp., Armonk, NY, USA). Quantitative variables were expressed as mean ± standard deviation (SD) and compared by Student’s t-test. Categorical variables were expressed as proportions and percentages (%, n/N) and compared by chi-squared test or Fisher’s exact test. Binary logistic regression was used to identify potential confounding variables associated with clinical pregnancy, and odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. As this was a retrospective study, we wanted to reduce bias in the results due to patient confounding. Therefore, we used propensity score matching to identify the CSBT cycles that were most identical to the REBT cycles. Matching was performed using nearest neighbor 1:1 matching without replacement, with a caliper width of 0.03. Propensity scores were assigned to maternal age, body mass index (BMI), duration of infertility, basal follicle stimulating hormone (FSH), anti-Müllerian hormone (AMH), endometrial thickness, primary infertility, primary reasons for infertility, percentage of natural cycles, good quality blastocyst transfer, percentage of blastocyst expansion grade 4, percentage of day 5 blastocyst, PGT euploid transfer, which could potentially affect the outcome. A two-sided P value of 0.05 or less was considered statistically significant.

To the best of our knowledge, this is the first study to evaluate the clinical outcomes and investigate factors influencing clinical pregnancy outcomes in CSBT cycles. The findings of this retrospective cohort study suggest that blastocysts which fail to re-expand within 2–4 h post-thawing should not be considered non-viable; even completely shrunken blastocysts can result in clinical pregnancies and live births, particularly when blastocyst formation is achieved by day 5. However, it is important to emphasize that CSBT is associated with significantly lower success rates compared to REBT (CPR: 28.8% vs. 61.5%; LBR: 20.2% vs. 50.0%; all P  < 0.001). Furthermore, binary logistic regression analysis revealed that maternal age, basal FSH levels, and day 3 embryo quality were not significantly associated with pregnancy outcomes in CSBT cycles. With the widespread adoption of embryo vitrification, several studies have demonstrated that frozen-thawed blastocyst transfers yield superior better clinical outcomes compared to fresh blastocyst transfers 28 – 30 . Moreover, the survival rate of vitrified blastocysts has significantly improved relative to slow freezing techniques 31 . Nevertheless, the systematic review and meta-analysis by Rienzi and colleagues found that 1–5% of vitrified blastocysts fail to survive after warming 32 . Blastocyst survival post-thawing is thought to depend on cellular integrity and the ability of the blastocyst to re-expand before transfer 7 , 24 , 33 . It is more difficult to accurately determine the number of cryo-injured cells in blastocysts than in cleavage embryos because of their different shape, but also because of their larger number of cells 34 , 35 . A current debate among researchers centers on whether blastocyst cavity re-expansion should be used as a criterion for assessing blastocyst survival. Desai et al. reported that no pregnancies were achieved following the transfer of single non-re-expanded blastocysts 19 . While the pregnancy success rate of non-re-expanded blastocysts was 0% in their study, this outcome may be attributing this to significant areas of necrosis observed in nearly all such blastocysts in thawed blastocysts 19 . In our study, we observed that approximately one third of the thawed, non-re-expanded embryos showed degeneration (Fig.  1 ). This suggests that blastocysts likely sustain cryo-injury during the freezing and thawing process, which can adversely affect subsequent embryo development, even when such damage is not visibly apparent 36 . Gardner et al. demonstrated that blastocysts which re-expanded within 5 h post-thawing exhibited significantly higher glucose and pyruvate uptake, as well as increased lactate production, compared to those that failed to re-expand even after 14 h of incubation, indicating a loss of metabolic capacity in damaged cells 37 . Additionally, the shrinkage and re-expansion processes during vitrification, along with the addition of cryoprotectants, can lead to cellular damage 34 , 38 . Our results demonstrate that blastocysts exhibiting re-expansion achieved significantly higher rates of clinical pregnancy, ongoing pregnancy, and live birth compared to those without re-expansion within 2–4 h post-warming, consistent with previous findings 9 , 39 . However, we also observed that while completely shrunken blastocysts had acceptable pregnancy success rates, these rates remained significantly lower than those of re-expanded blastocysts (Table  2 ). We hypothesize that although these blastocysts showed no visible signs of necrosis, a portion of TE cells may have sustained cryo-damage during freezing and thawing. This damage could impair cellular sealing, preventing water influx into the blastocoel cavity through osmotic pressure generated by ion channels post-warming, thereby inhibiting blastocyst re-expansion 16 , 17 , 20 . Additionally, blastocyst hatching is characterized by repeated cycles of contraction and expansion of the blastocyst cavity 40 . Since embryologists typically assess embryos at a single time point, the expansion status of the blastocyst cavity may be misinterpreted if the observation coincides with a contraction phase. Therefore, the addition of an embryo or cancelling the embryo transfer cycle simply because the blastocoel cavity has not been re-expanded is not recommended unless the first embryo has suffered severe cryo-injury. Giunco et al. measured changes in blastocyst cavity diameter after thawing and before transfer (including transitions from contraction to expansion or expansion to contraction) and concluded that there was no significant effect on clinical pregnancy whether the embryo went from contraction to expansion or vice versa at the time of transfer 10 This finding supports our view that relying on a single observation point to assess blastocyst survival, particularly for those that have not re-expanded, is not appropriate. Moreover, time-lapse imaging revealed that after 5 h of incubation, some shrunken embryos formed a blastocoel cavity and grew to the hatching stage 11 . However, Ahlström et al. assessed blastocyst re-expansion over a period of 1–5 h and concluded that the effect of time on the degree of re-expansion was minimal after thawing, with up to 2 h of incubation, as blastocysts incubated for longer periods prior to evaluation did not consistently exhibit greater expansion 41 . Furthermore, AS is an invasive procedure involving puncture of both the zona pellucida and TE. This intervention may potentially impair blastocyst cavity re-expansion post-thawing due to compromised TE integrity 42 . However, more recent studies have demonstrated that AS (via laser-assisted or biopsy-induced methods) prior to vitrification is associated with reduced degeneration risk after warming (OR 0.26, 95% CI = 0.09–0.79). Logistic regression analyses from these studies further indicated that AS does not significantly affect embryo re-expansion potential 43 . The second aim of the study is to determine which factors were most influential in the clinical pregnancy of CSBT. After adjusting for confounders, the logistic regression analysis reveals that the transfer of day 5 blastocyst is a positive factor of pregnancy in CSBT cycles. The mechanisms of influence are still unknown. The lower likelihood of pregnancy in fresh blastocysts with slow growth to blastocyst formation on day 6 44 may be due to the asynchronous uterine environment with poor endometrial receptivity 45 but even in frozen cycles with impaired receptivity and aneuploid embryo avoidance, the outcome of blastocyst pregnancies on days 5 and 6 differed 46 . Desai et al. found that vitrified day 5 blastocyst transfer resulted in three times higher pregnancy and live birth rates than day 6 embryos 47 . The difference in clinical outcomes may be due to the quality of day 5 versus day 6 embryos. It has been reported that day 5 embryos had a greater high-quality blastocyst (HB) rate than day 6 embryos (61.6% vs. 29.4%), resulting in improved clinical pregnancy and implantation rates (57.4% vs. 46.2%, 58.9% vs. 47.3%). If only HB embryos were counted, there was no difference in clinical outcomes between day 5 and day 6 (60% vs. 54.5%, 62% vs. 56.3%) 48 . Sunkara et al. found no difference in CPRs or OPRs / LBRs between studies comparing day 5 and day 6 blastocyst transfers with the same morphological quality after freezing 49 . However, contradictory results have been reported 50 , 51 . Perhaps the variation in clinical outcome is attributable to metabolic or epigenetic differences between day 5 and day 6 embryos 52 or it may be linked to a greater aneuploidy rate in developmentally delayed embryos on day 6 53 . In addition, in animal models, slower-developing blastocysts exhibit distinct RNA expression patterns than faster-developing blastocysts 54 . Although we did not assess quality differences between day 5 and day 6 blastocysts or perform PGT. This may have resulted in an uneven distribution of quality and differences in aneuploidy rates between the groups, which could explain the observed higher success rate of day 5 blastocysts compared to day 6 blastocysts. A recent study indicated that excellent-quality embryos and those biopsied on day 5 demonstrate higher euploidy rates compared to good or poor quality embryos and day 6 biopsied embryos, respectively. Poor quality embryos and day 6 biopsied embryos exhibit significantly reduced survival rates following vitrification 55 . Of note, an important limitation of this study is the absence of time-lapse imaging to monitor the dynamic developmental processes of embryos. Consequently, we were unable to determine whether blastocysts remained consistently shrunken or underwent cycles of expansion and re-shrinkage, potentially introducing bias into our observations. In conclusion, blastocysts that fail to re-expand within 2–4 h post-warming demonstrate significantly lower CPR, OPR, and LBR compared to their re-expanded counterparts. However, when completely shrunken blastocysts show no evident signs of cryo-damage, we strongly recommend against discarding them. Instead, these blastocysts should be considered for clinical use, with appropriate patient counseling regarding expected outcomes and the option for patients to make informed decisions.

The introduction of embryo culture and transfer at the blastocyst stage marked a significant advancement in in vitro fertilization and embryo transfer (IVF-ET), reducing the number of embryos transferred while improving clinical pregnancy and live birth rates 1 , 2 . Blastocyst culture is a process in which the embryo undergoes self-correction, potentially selecting the most viable embryo(s); indeed, blastocysts exhibit lower aneuploidy rates compared to cleavage-stage embryos 3 . In addition, blastocyst transfer improves the synchronicity between the endometrium and the embryo, resulting in a higher implantation rate 4 , 5 . Consequently, as the number of cultured blastocysts increases, so does the number of blastocysts available for frozen-thawed cycles. As is well known, vitrified blastocyst transfer led to much higher clinical pregnancy and live birth rates than slow frozen blastocyst transfer, with somewhat better perinatal and neonatal outcomes than fresh blastocyst transfer 6 , 7 . The vitrification technique is simple, inexpensive, and efficient, does not require expensive equipment, and has now replaced the slow freezing methods 8 . Although vitrified-warmed blastocysts have been widely used in assisted reproductive technology (ART), the assessment of blastocyst survival after warming is currently controversial 9 – 12 . The survival of a blastocyst after thawing usually depends on two different events. The first is the percentage of cells that survive, and the second is whether the re-expansion of the embryo. Embryo survival depends in part on the ability of the blastocyst and its constituent cells to respond effectively to vitrification and the warming process. Unlike embryos at the cleavage stage, blastocyst survival appears to be difficult to quantify, with no agreement among published studies in this field. In theory, blastocysts in vitro continue to complete mitosis after warming, making it easier to identify ‘living’ embryos. However, the small size and shape, as well as the high density of trophectoderm (TE) cells, make it difficult to notice the resumption of mitosis in the same manner as cleavage stage embryos do 13 , 14 . When vitrifying the blastocyst, the blastocoel cavity is filled with a considerable volume of fluid, which can cause intracellular ice crystal formation and potentially fatal embryo damage during cooling by altering the dehydration shrinkage rate. Studies have shown that artificial shrinkage (AS) of the blastocoel cavity prior to vitrification can minimize ice crystal formation and reduce cryo-damage, improve thawed blastocyst survival rates, and greatly increase pregnancy rates 15 . Blastocyst fluid accumulation during cavitation is controlled by a Na/K-ATPase-generated trans-TE ionic gradient that promotes water accumulation across the epithelium 16 which, when combined with the formation of the tight junction seal that controls paracellular water movement between adjacent TE cells 17 results in the formation of a fluid-filled blastocoel cavity and re-sealing of the TE after warming, allowing the blastocyst re-expansion 16 , 18 . Re-expansion events are similar to those observed during the development of fresh blastocysts, where TE cells are likely to be viable and capable of metabolic activity, after which TE cells actively pump sodium ions, followed by a passive influx of water due to osmotic imbalances 17 . It was hypothesized that delayed or no re-expansion of thawed blastocysts would signal that the TE cells had been damaged during the freezing-warming process and thereby had lost their ability to re-seal 19 , 20 . Cryo-injury occurs as a result of cytoskeletal damage during vitrification-warming, which has a direct impact on mitotic resumption and subsequent implantation 21 , 22 . As a result, many clinics have adopted blastocyst re-expansion as a survival criterion, eliminating blastocysts that fail to re-expand 23 – 25 . However, some researchers argue that blastocyst development after thawing is a dynamic process involving repeated cycles of contraction and expansion. Consequently, assessing blastocyst viability based on a single observation point, especially for those that have not re-expanded, if the observation timing coincides with the blastocyst contraction phase, it may lead to misinterpretation of the embryo’s developmental status 9 , 10 . The recent exponential increase in vitrified blastocyst transfer cycles, combined with the paucity of studies on the non-re-expanded blastocysts post-thawing, highlights the need to identify factors influencing the success of non-expanded blastocyst transfers. Therefore, the aim of this study was to evaluate the clinical outcomes of blastocysts that failed to re-expand within 2–4 h post-thawing and to analyze the relevant factors affecting clinical pregnancy in CSBT cycles.