Double vitrification-warming cycles coupled with one blastocyst biopsy do not affect the chance of live birth after an euploid blastocyst transfer: a propensity score matching study.

After screening 1570 frozen-thawed blastocyst transfer cycles of transferred euploid embryos between January 2018 and March 2023, the study included 766 SVSB cycles and 197 TBR cycles (Fig. 1 ). There were statistically significant differences between the tbr group and the svsb group in oocyte retrieval age, basal LH level, infertility duration, infertility type, infertility factors, etc. After PSM, all data from the TBR group were successfully matched, whereas the successfully matched SVSB group included 351 FET cycles. After matching, the baseline characteristics were comparable between the two groups (Table 1 ). Table 1 Baseline characteristics before and after propensity score matching Before propensity score matching After propensity score matching Characters SVSB group（n = 766） TBR group（n = 197） P a value SVSB group（n = 351） TBR group（n = 197） P a value Age of oocyte retrieval（year） 32.00 (30.00, 35.00) 32.00 (28.00, 35.00) 0.014 32.00 (30.00, 35.00) 32.00 (28.00, 35.00) 0.111 Maternal BMI（Kg/m 2 ） 23.10 (21.30, 25.20) 22.58 (20.65, 25.50) 0.300 23.20 (21.40, 25.40) 22.58 (20.65, 25.50) 0.105 Laboratory testing AMH 3.40 (2.13, 5.13) 3.71 (2.04, 6.09) 0.103 3.70 (2.17, 5.84) 3.71 (2.04, 6.09) 0.576 FSH 6.15 (5.11, 7.42) 6.26 (5.23, 7.03) 0.772 6.12 (5.24, 7.41) 6.26 (5.23, 7.03) 0.711 E2 35.96 (25.57, 48.40) 38.43 (28.58, 49.51) 0.143 36.08 (26.12, 49.00) 38.43 (28.58, 49.51) 0.422 LH 4.66 (3.18, 6.81) 5.49 (3.70, 7.30) 0.023 4.74 (3.20, 6.94) 5.49 (3.70, 7.30) 0.119 T 0.27 (0.18, 0.37) 0.27 (0.18, 0.37) 0.736 0.27 (0.19, 0.37) 0.27 (0.18, 0.37) 0.729 Fertility history Duration of infertility（year） 2.00 (1.00, 4.00) 3.00 (2.00, 5.00) 0.000 3.00 (1.00, 5.00) 3.00 (2.00, 5.00) 0.089 Primary infertility 73 (9.5) 7 (3.6) 0.007 27 (7.7) 7 (3.6) 0.054 Previous conception 2.00 (1.00, 4.00) 2.00 (1.00, 3.00) 0.002 2.00 (1.00, 3.00) 2.00 (1.00, 3.00) 0.155 Previous miscarriage 646 (84.3) 171 (86.8) 0.389 298 (84.9) 171 (86.8) 0.543 Previous live birth 184 (24.0) 66 (33.5) 0.007 103 (29.3) 66 (33.5) 0.312 Infertility diagnosis b Tubal disease 269 (35.1) 100 (50.8) 0.000 169 (48.1) 100 (50.8) 0.557 Polycystic ovary syndrome 62 (8.1) 37 (18.8) 0.000 49 (14.0) 37 (18.8) 0.136 Male factor infertility 159 (20.8) 61 (31.0) 0.002 91 (25.9) 61 (31.0) 0.206 Endometriosis 30 (3.9) 10 (5.1) 0.467 17 (4.8) 10 (5.1) 0.904 Others 320 (41.8) 33 (16.8) 0.000 77 (21.9) 33 (16.8) 0.146 FET protocol NC 113 (14.8) 35 (17.8) 0.295 62 (17.7) 35 (17.8) 0.976 HRT 475 (62.0) 114 (57.9) 0.287 202 (58.1) 114 (57.9) 0.954 GnRH-α HRT 172 (22.5) 46 (23.4) 0.789 80 (22.8) 46 (23.4) 0.882 Stimulate 6 (0.8) 2 (1.0) 0.749 5 (1.4) 2 (1.0) 0.682 Quality of blastocysts Good 527 (68.8) 121 (61.4) 0.049 226 (64.4) 121 (61.4) 0.489 Normal 239 (31.2) 76 (38.6) 125 (35.6) 76 (38.6) PGT indication b Chromosome abnormalities c 66 (8.6) 6 (3.0) 0.008 27 (7.7) 6 (3.0) 0.028 Recurrent pregnancy loss 366 (47.8) 30 (15.2) 0.000 137 (39.0) 30 (15.2) 0.000 Repeated implantation Failure 98 (12.8) 32 (16.2) 0.206 54 (15.4) 32 (16.2) 0.791 Embryo chromosome abnormality history 133 (17.4) 130 (66.0) 0.000 56 (16.0) 130 (66.0) 0.000 Advanced maternal age,≥35 y 12 (1.6) 1 (0.5) 0.422 8 (2.3) 1 (0.5) 0.224 Others 175 (22.8) 26 (13.2) 0.003 100 (28.5) 26 (13.1) 0.000 Note: Data are presented as median (quartiles) or number/total number (%).The variables in the propensity score matching included maternal age at oocyte retrieval, BMI, basal endocrine level; fertility history (duration of infertility, type of infertility, previous conception, previous miscarriages, and previous live birth); infertility diagnosis (tubal factors, PCOS, male factor infertility, endometriosis and other factors); type of FET protocol. High-quality blastocysts were defined as those achieving Gardner scale ratings of A or B for both inner cell mass and trophectoderm morphology. SVSB: single vitrification and single biopsy; TBR: thaw-biopsy-refreeze BMI: body mass index; PGT: preimplantation genetic testing; FET: frozen embryo transfer; NC: natural cycle; HRT: hormone replacement treatment cycle a p  < 0.05 was considered statistically significant b The diagnosis of infertility and the indications for PGT are not mutually exclusive c Chromosomal abnormalities in at least one member of the couple Baseline characteristics before and after propensity score matching Note: Data are presented as median (quartiles) or number/total number (%).The variables in the propensity score matching included maternal age at oocyte retrieval, BMI, basal endocrine level; fertility history (duration of infertility, type of infertility, previous conception, previous miscarriages, and previous live birth); infertility diagnosis (tubal factors, PCOS, male factor infertility, endometriosis and other factors); type of FET protocol. High-quality blastocysts were defined as those achieving Gardner scale ratings of A or B for both inner cell mass and trophectoderm morphology. SVSB: single vitrification and single biopsy; TBR: thaw-biopsy-refreeze BMI: body mass index; PGT: preimplantation genetic testing; FET: frozen embryo transfer; NC: natural cycle; HRT: hormone replacement treatment cycle a p  < 0.05 was considered statistically significant b The diagnosis of infertility and the indications for PGT are not mutually exclusive c Chromosomal abnormalities in at least one member of the couple As shown in Table 2 , compared with the SVSB group, the number of blastocysts involved in the biopsy was significantly reduced in the TBR group, and the euploidy rate was significantly higher, while the aneuploidy rate was significantly lower. Nevertheless, no statistically significant discrepancy was observed in the PCR amplification failure rate at biopsy between the two groups. Following the PSM, the number of blastocysts undergoing biopsy in the TBR group was still significantly lower than that in the SVSB group, but the proportions of euploid and aneuploid blastocysts in the two groups were not statistically different. The primary causes of the discrepancy between the number of biopsied blastocysts and thawed embryos in the TBR group were failure of resuscitation (0.41%) and failure of cultured blastocysts (0.54%). Table 2 Comparison of preimplantation genetic testing results between SVSB group and DVSB group before and after propensity score matching Outcomes Before propensity score matching After propensity score matching SVSB group（n = 569） TBR group（n = 165） P a value SVSB group（n = 279） TBR group（n = 165） P a value Warming embryos/blastocysts ­ 740 ­ ­ 740 ­ Number of blastocysts biopsied 3410 696 0.000 1702 696 0.000 Reasons for the decrease in the number of biopsies Failed to warm ­ 3 (0.41) ­ ­ 3 (0.41) ­ Failed blastocyst raising ­ 41 (5.54) ­ ­ 41 (0.54) ­ Failure of PCR 30 (0.88) 4 (0.57) 0.418 15 (0.88) 4 (0.57) 0.442 Euploid blastocyst 1562 (46.21) 349 (50.43) 0.043 810 (48.01) 349 (50.43) 0.284 Aneuploid blastocyst 1818 (53.79) 343 (49.57) 877 (51.99) 343 (49.57) Note: Data are presented as number or number/total number (%).SVSB: single vitrification and single biopsy; TBR: thaw-biopsy-refreeze; PCR: polymerase Chain Reaction; Failed blastocyst raising: Abnormal growth rate and abnormal embryo quality assessment in culture a P  < 0.05 was considered statistically significant Comparison of preimplantation genetic testing results between SVSB group and DVSB group before and after propensity score matching Note: Data are presented as number or number/total number (%).SVSB: single vitrification and single biopsy; TBR: thaw-biopsy-refreeze; PCR: polymerase Chain Reaction; Failed blastocyst raising: Abnormal growth rate and abnormal embryo quality assessment in culture a P  < 0.05 was considered statistically significant As shown in Table 3 , the live birth rate was slightly, but not significantly, higher in the TBR group than in the SVSB group after PSM ( OR , 1.31; 95% CI , 0.92, 1.86; p  = 0.138). Meanwhile, there was no significant difference between the two groups in terms of BPR, CPR, ectopic pregnancy rate, multiple pregnancy rate, pregnancy loss rate, total pregnancy loss rate, preterm labor rate, and caesarean section rate. Table 3 Comparison of pregnancy outcomes between SVSB group and DVSB group after propensity score matching Outcomes SVSB group（n = 351） TBR group（n = 197） OR (95% CI) P a value Primary outcome Live birth 141 (40.2) 92 (46.7) 1.31 (0.92, 1.86) 0.138 Secondary outcomes Biochemical pregnancy 212 (60.4) 125 (63.5) 1.14 (0.79, 1.63) 0.481 Clinical pregnancy 180 (51.3) 110 (55.8) 1.20 (0.85,1.71) 0.305 Ectopic pregnancy (per clinical pregnancy) 8 (4.4) 3 (2.7) 0.60 (0.16,2.32) 0.458 Multiple pregnancy (per clinical pregnancy) 3 (1.7) 3 (2.7) 1.65 (0.33, 8.34) 0.538 Pregnancy loss (per clinical pregnancy) 31 (17.2) 15 (13.6) 0.76 (0.39, 1.48) 0.417 Total pregnancy loss (per hCG positive) 71 (33.5) 33 (26.4) 0.71 (0.43, 1.16) 0.173 Premature (per live birth) 24 (17.0) 9 (9.8) 0.53 (0.23, 1.20) 0.121 Cesarean section (per live birth) 121 (85.8) 75 (81.5) 0.73 (0.36, 1.48) 0.381 Note: Data are presented as number or number/total number (%). Univariate logistic regression and linear regression were used to calculate odds ratio (95% CI) and mean difference (95% CI). SVSB: single vitrification and single biopsy; TBR: thaw-biopsy-refreeze; OR: odds ratio; CI: confidence interva. a p  < 0.05 was considered statistically significant Comparison of pregnancy outcomes between SVSB group and DVSB group after propensity score matching Note: Data are presented as number or number/total number (%). Univariate logistic regression and linear regression were used to calculate odds ratio (95% CI) and mean difference (95% CI). SVSB: single vitrification and single biopsy; TBR: thaw-biopsy-refreeze; OR: odds ratio; CI: confidence interva. a p  < 0.05 was considered statistically significant All neonatal outcomes were comparable in both groups. This analysis included all live-born neonates (Table 4 ). We followed the newborns in both groups for at least six months. In the SVSB group, there was only one case of poor lung development in a 30-week preterm baby. The TBR group reported one case of severe anemia and an adverse obstetric outcome of a 29-week preterm baby who died after 15 days of life. Additional congenital anomalies include cardiac malformations, cutaneous hemangiomas, and methylpropionic acidemia. However, these conditions are typically mild and do not significantly impact neonatal mortality. No significant complications were observed with regard to obstetric outcomes. Table 4 Comparison of neonatal outcomes between SVSB group and TBR group after propensity score matching Neonatal outcomes SVSB group (n = 143) TBR group (n = 96) Z /χ2 P a value Gestational age (weeks) 37.78 (37.00, 39.00) 37.72 (37.00, 39.00) 0.406 0.685 Sex (male/female), n (%) 66/77 (46.15/53.85) 52/44 (54.17/45.83) 1.475 0.224 Length (cm) 49.49 (49.00, 51.00) 49.53 (49.00, 51.00) 0.881 0.378 Birthweight (g) 3250.56 (3000.00, 3650.00) 3298.68 (2987.50, 3800.00) 0.824 0.410 Low birthweight, n (%) 12 (8.39) 7 (7.29) 0.095 0.758 fetal macrosomia, n (%) 13 (9.09) 7 (7.29) 0.242 0.622 Birth defect, n (%) 3 (2.10) 4 (4.17%) 0.290 0.590 Note: Data are presented as median (quartiles) or number/total number (%). SVSB: single vitrification and single biopsy; TBR: thaw-biopsy-refreeze. a p  < 0.05 was considered statistically significant Comparison of neonatal outcomes between SVSB group and TBR group after propensity score matching Note: Data are presented as median (quartiles) or number/total number (%). SVSB: single vitrification and single biopsy; TBR: thaw-biopsy-refreeze. a p  < 0.05 was considered statistically significant

In this retrospective cohort study, we collected all PGT-A cycles from January 2018 to March 2023 for single frozen-thawed blastocyst transplants at the Reproductive Medicine Centre of the First Affiliated Hospital of Zhengzhou University. Only data from thawed cycles of single blastocyst transfers were included, and cycles involving oocyte donation, multiple blastocyst transfers, uterine anatomical abnormalities, and missing data were excluded (Fig. 1 ). The cycles were divided into two groups according to the source of the biopsied embryos. One group was the conventional PGT-A group (SVSB group), which underwent only one biopsy and one vitrification-warming procedure. The other group was the TBR group, which underwent one additional vitrification-warming procedure. RPL is defined as the failure of two or more clinically recognized pregnancies before 20–24 weeks of gestation and includes embryonic and fetal losses [ 13 ]. Repeated implantation failure (RIF) was defined as failure to achieve a clinical pregnancy after transfer of at least 4 good-quality embryos in at least 3 fresh or frozen cycles in patients under 40 years of age undergoing IVF-ET [ 14 ]. High-quality blastocysts were defined as those achieving Gardner scale ratings of A or B for both inner cell mass and trophectoderm morphology. Fig. 1 Flowchart detailing the screening of two study group cycles Flowchart detailing the screening of two study group cycles All embryos in this study were selected for vitrification. Freezing solution, thawing solution, and pipettes were provided by Kitazato (Tokyo). All vitrification and warming procedures were performed at room temperature. After the blastocysts reached the expansion stage (grade 4 or higher according to Gardner classification), they were placed in Shrinkage Solution (VT525-4). It was then quickly placed in Equilibration Solution (300 μL) for 15 min to rest and then transferred to Vitrification Solution (VS) to ensure dehydration. After aspirating excess VS, the cryotop was plunged into liquid nitrogen within 1 second for vitrification. The carrier cannula was removed, and the blastocyst ends were immediately immersed in thawing solution (TS) for 1 min at 37 °C. Then, the blastocyst ends were transferred to dilution solution (DS) and then to washing solutions 1 and 2. Finally, cells were transferred to blastocyst medium (Vitrolife, Sweden) and placed in an incubator containing 6% C02 at 37 °C pending transfer [ 15 ]. At the blastocyst stage, zona pellucida perforation is performed using a laser, followed by the use of a biopsy pipette to aspirate 3–5 trophoblast ectodermal cells from the blastocyst and perform genetic testing. NGS technology is usually chosen in our center for sequencing of the obtained trophoblast cells. After biopsy, the obtained cells were lysed in 0.2N KOH. Whole genome amplification (WGA) was then performed using the REPL-g Single Cell Kit (QIAGEN, 150345) according to the instructions. Upon completion of library preparation and quality control, library sequencing was performed on a HiSeq 2500 sequencer (Illumina, USA) using rapid single-end 50-cycle sequencing mode at our center. Raw sequencing data were demultiplexed and converted to FASTQ format using CASAVA software (v1.8.4; Illumina, USA). Quality filtering was subsequently conducted with Trimmomatic, yielding approximately 2 million filtered reads per sample with an average sequencing depth of ~0.03×. The resulting high-quality reads were then aligned to the hg19 reference genome using the Burrows-Wheeler Aligner package. Unique mapped reads were extracted from the alignment reads (bam file). The reference genome was then divided into non-overlapping observation windows (bins) with a size of 1000Kb. The number of reads and GC-content were calculated in each bin and GC bias correction was performed for every 1% GC-content [ 16 ]. Finally, the R programming language was employed to generate plots of the final relative read counts for each genomic window to visualize copy number variations. Throughout the analysis pipeline, autosomes were assigned a default copy number of 2, while sample gender and sex chromosome copy numbers were determined using an initial calling algorithm [ 17 ]. Our center employs the following diagnostic criteria for copy number variation analysis: euploidy is defined by copy number values of 1.80–2.20; complete aneuploidy by values  2.80; and mosaic aneuploidy by values falling within the 1.20–1.80 or 2.20–2.80 ranges [ 18 ]. The primary pregnancy outcome was the live birth rate (LBR), while the secondary outcomes were the biochemical pregnancy rate (BPR), clinical pregnancy rate (CPR), pregnancy loss rate, multiple pregnancy rate, preterm birth rate, caesarean section rate, and ectopic pregnancy rate. Neonatal outcomes include gestational age, length, birthweight, and birth defects. The biopsy outcomes include the number of embryos biopsied, the aneuploidy rate, and the polymerase chain reaction (PCR) amplification failure rate. Live birth was defined as the live-born delivery of a neonate at or beyond 24 weeks’ gestation. Pregnancy loss was defined as spontaneous abortion of a pregnancy before 20 weeks of gestation, and the total pregnancy loss rate included both pregnancy loss and biochemical pregnancy loss. All data for this study were obtained from the Clinical Reproductive Medicine Management System/Electronic Medical Record Cohort Database (CCRM/EMRCD) of our centre. IBM SPSS 26.0 was used for statistical analysis. Continuous variables in this study did not satisfy normality and were expressed as median (P25,P75). The Mann-Whitney U test was employed for the comparison of these variables. Categorical variables were expressed as frequencies and percentages and were compared using the chi-square test or Fisher’s exact test, as appropriate. The limited sample size of the TBR group and the inclusion of a population with a different baseline level than that of the SVSB group may result in a significant population bias, which could hinder the dissemination of our findings to a broader audience. We therefore performed PSM on them based on age at follicles retrieval, BMI, basal endocrine levels, pregnancy history, infertility diagnosis, and frozen embryo transfer (FET) protocol, and selected the PGT-A patients most similar to the TBR group. In order to minimise data loss, a ratio of 1:4 was chosen for the matching of the two groups. Subsequently, the cycle outcomes were compared following the matching process. p  < 0.05 was considered statistically different. The study was approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University for Research and Publication (2024-KY-2018–002).

Given the gradual increase in the use of TBR procedures in clinical practice, further investigation into their impact on biopsy outcomes and clinical outcomes could provide a theoretical basis for clinical diagnosis and treatment. Our findings indicate that despite a 6.5% absolute reduction in the live birth rate following a second vitrification-warming procedure, the difference was not statistically significant (40.2 vs. 46.7, p  = 0.138). Furthermore, the safety of the TBR procedure in clinical use was supported by the findings regarding other pregnancy outcomes and neonatal outcomes. Surprisingly, the LBR, BPR, and CPL in the TBR group were even slightly higher than those in the SVSB group, while the total pregnancy loss rate was lower than that in the SVSB group. This discrepancy may be attributed to the inherent differences between the two study populations. Despite the use of PSM, the overall age at egg retrieval and the BMI of the mothers remained lower in the TBR group compared to the SVSB group. The different medical histories resulted in statistically significant differences in the basic characteristics of the two groups. In order to control for the two groups, this study adjusted for potential confounding factors that may affect pregnancy outcomes through the use of PSM [ 19 – 23 ]. Following PSM, although the proportion of high-quality blastocysts transferred remained higher in the SVSB group compared to the TBR group, this difference did not reach statistical significance. The primary indications for PGT-A treatment in both groups were RPL, RIF, and a history of chromosomal abnormalities in embryos. In the SVSB cohort, the primary indication was RPL (pre-PSM: 47.8%; post-PSM: 39.0%), whereas in the TBR cohort, it was a history of chromosomal abnormalities in the aborted tissue (66.0%). Scott et al. and Kokkali G et al. demonstrated that blastocyst biopsy is less detrimental on the developmental potential of the embryo when compared to zygote stage biopsy [ 24 , 25 ]. So if we thaw frozen cleavage-stage embryos, we culture them to blastocyst stage before biopsy. In contrast to the SVSB cohort, where ICSI was the standard procedure, the majority of the TRB group employed cIVF embryos. The results of the prospective studies by Zhang et al. and Dong et al. both indicate that the genetic material of spermatozoa is not amplified in PGT-A [ 26 , 27 ]. Although there may be some maternal contamination, the probability is extremely low (0.67% and 0.83%, respectively). Our study lend further support to this perspective. Prior research has demonstrated that the survival rate of thawed blastocysts following two vitrification cycles remains above 90% even after undergoing one embryo biopsy [ 10 , 11 , 28 ]. Of the 322 blastocysts in the TBR cohort, only one did not survive after thawing, which is consistent with the findings of previous studies. The timing of the initial vitrification-warming cycle in the TBR cohort varied, encompassing both the oocyte and blastocyst stages. A number of studies have indicated that the clinical outcome of vitrification is not affected by whether the procedure is performed at two different stages of development or twice at the blastocyst stage [ 29 – 31 ]. Unexpectedly, the TBR group exhibited a substantially greater proportion of euploid embryos relative to the SVSB group. Post-matching analysis revealed that while the intergroup difference no longer reached statistical significance, the directional trend remained consistent. After controlling for blastocyst quality parameters, maternal age and body mass index at the time of oocyte retrieval were identified as potential determinants underlying this phenomenon. We identified six studies that attempted to examine whether the TBR procedure affects clinical outcomes. Four of these studies were similar to our study and included single blastocyst transfers. Findings by Bradley et al. demonstrated that an additional vitrification-warming, although resulting in a slight reduction in LBRs, biochemical pregnancies, and CPRs, was not statistically different [ 11 ]. However, the study did not adjust for the significant discrepancy in data volume between the two data sets, with a difference of nearly 60-fold. This may have affected the statistical validity of the findings, thereby undermining their credibility. Subsequently, Zhang et al. included two cohorts with similar data volumes constituting the case and control groups [ 28 ]. The results showed that the trend of biochemical and CPLs in the TBR group was similar to that of our findings but with higher miscarriage rates, which ultimately led to a slight decrease in the LBRs. The lower age at follicle retrieval and BMI in the TBR group may provide an explanation for this phenomenon. Of course, none of these changes were statistically different. The remaining two studies adjusted for confounding variables in their statistical analyses, which is comparable to our approach for PSM. However, the resulting findings were markedly disparate. Both Aluko et al. and Li et al. demonstrated that even after adjusting for potential confounding variables, the LBR was significantly lower in the TBR group [ 12 , 32 ]. In the study by Aluko et al., the CPL was significantly lower in the TBR group, suggesting impaired implantation potential of the embryos [ 32 ], which is consistent with a study by Wang et al. in 2021 [ 33 ]. Whereas Li et al. found no significant difference in biochemical and clinical pregnancy rates between the two groups, the TBR group had a significantly higher rate of pregnancy loss, predominantly late miscarriages [ 12 ]. Finally, two studies on PGT-A of thawed embryos were conducted in the context of an unrestricted number of blastocysts transferred. Taylor et al. observed that while more than one slow freezing reduced blastocyst thawing survival, the outcomes following embryo transfer were comparable irrespective of the number of freezes and the method of embryo transfer employed [ 10 ]. Despite the absence of live birth data in the Wilding et al. study, the authors observed comparable clinical and ongoing pregnancy rates between the TBR and SVSB groups [ 29 ]. The two studies were subject to the same limitations. Primarily, neither study took into account the baseline data for both groups, nor did they adjust for potential confounding variables when analyzing the study outcomes. Secondly, both studies had fewer than 10 cases in the TBR group at the sustained pregnancy stage. Additionally, only the studies by Bradley, Li, and Zhang et al. reported neonatal outcomes among these six studies. The findings of these studies are in alignment with our own, indicating that the TBR procedure does not impact neonatal outcomes [ 11 , 12 , 26 ]. Our study includes the largest TBR cohort among the relevant published studies, which enhances the statistical power and reliability of our findings. Furthermore, we also considered multiple cohort characteristics relevant to the study questions and matched them using PSM, thus enhancing the homogeneity of our research cohort. Nevertheless, the limited number of ectopic and multiple pregnancy cases in our cohort may reduce statistical power, warranting validation through future multi-center studies with larger sample sizes. The present study exclusively encompassed a particular cohort of individuals who elected to pursue PGT-A due to suboptimal transfer outcomes and who often had several high-quality embryos remaining to satisfy biopsy needs. It would be unwise to extend the results arbitrarily to a larger population. Although there was no statistically significant difference in neonatal outcomes between the groups, it is important to note that the clinical significance of the single infant death in the TBR group must be considered in the context of the study’s limited sample size. In addition, this study only considered the clinical and laboratory outcomes without further cost-effectiveness analysis. A comprehensive retrospective study indicated that for patients with more than one embryo, PGT-A resulted in reduced healthcare costs, a shorter treatment period, and a decreased risk of embryo transfer failure and clinical miscarriage [ 34 ]. Furthermore, a theoretical cost-effectiveness study has indicated that PGT-A is a cost-effective option in specific clinical settings and populations [ 35 ]. However, Scriven examined the cost-effectiveness profile of PGT-A using intricate and complex modeling, but failed to show any benefit [ 36 ]. There is a lack of high-quality randomized controlled trials to support clinical provision of diagnosis and treatment.

In summary, To our knowledge, this is the largest and most rigorously matched study to demonstrate that selection of remaining frozen embryos for PGT-A does not compromise pregnancy and neonatal outcomes. The TBR procedure makes optimal use of a patient’s remaining embryos, which may reduce the physical and psychological stress, as well as the time and financial costs, associated with ovulation restimulation. We present further evidence to support clinicians in recommending PGT-A for remaining frozen embryos when patients are well informed of the risks and benefits of the procedure. More high-quality randomized controlled trials are needed in the future to demonstrate the generalisability of our findings and to further explore the cost-effectiveness of the TBR procedure.

Preimplantation genetic testing for aneuploidies (PGT-A) allows biopsy and genetic diagnosis of embryos during in vitro fertilization-embryo transfer (IVF-ET) to select chromosomally normal embryos for transfer. This approach significantly reduces the risk of early pregnancy loss, embryo demise, and even stillbirths compared to morphological assessment alone [ 1 – 3 ]. The American Society for Reproductive Medicine (ASRM) stated that ‘there is insufficient evidence to recommend the routine use of blastocyst biopsy and aneuploidy testing in all infertile patients undergoing IVF treatment’ [ 4 ]. Despite the limitations of the use of PGT-A, it is recommended by many in patients with repeated pregnancy loss (RPL) because more than half of early pregnancy losses are the result of aneuploidy [ 5 , 6 ]. Several studies have demonstrated that the implementation of PGT-A can result in a reduction in the number of transferred embryos required to achieve comparable live births in patients afflicted with recurrent miscarriages [ 7 ]. Indeed, there is even evidence to suggest that it can enhance clinical pregnancy and live birth rates [ 8 , 9 ]. Some patients undergoing conventional IVF (cIVF) treatment present with clinical manifestations including unexplained embryonic arrest, abnormal chromosomal testing of chorionic tissue, and RPL following several consecutive fresh or thawed cycle transfers. In order to circumvent the potential disadvantages associated with aneuploid embryos, clinicians often recommend PGT-A following a comprehensive evaluation of relevant clinical factors. In cases where there are multiple frozen high-quality embryos or blastocysts, couples may elect to culture the remaining frozen embryos and perform a trophoblast biopsy. This approach obviates the need for further ovulation induction, thereby saving time and reducing costs. All blastocysts undergoing PGT-A are vitrified to await biopsy results at our center. Consequently, the remaining frozen embryos selected for biopsy must undergo a second vitrification-warming process, in contrast to the standard PGT-A procedure. To date, only a limited number of studies have examined the effect of double vitrification on the developmental potential of euploid blastocysts that have undergone trophectoderm biopsy [ 10 – 12 ]. However, the findings from these studies are inconsistent. It is worth noting that the clinical characteristics of the two groups differed considerably, a fact that has been overlooked in the majority of studies conducted to date. Considering that thaw-biopsy-refreeze (TBR) makes full use of the remaining embryos and is gradually increasing in clinical practice, a large-scale study with rigorous matching is warranted. Therefore, this retrospective study thoroughly investigates the pregnancy and neonatal outcomes of blastocyst transfers that underwent one biopsy and double vitrifications on the basis of propensity score matching (PSM).