Introduction
Assisted reproductive technologies have helped thousands of infertile patients have children. Historically, more than one embryo is transferred into the uterus during the in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) process, which can lead to multiple pregnancy. Multiple pregnancy is not only dangerous for the mother, but also increases the incidence of miscarriage, premature birth and low birth weight (LBW) infants. Single embryo transfer is the most effective way to reduce the likelihood of a multiple pregnancy, and selective single blastocyst transfer can significantly reduce multiple pregnancy without reducing the pregnancy rate at this time (Sundhararaj et al., Reference Sundhararaj, Madne, Biliangady, Gurunath, Swamy and Gopal2017; Kwek et al., Reference Kwek, Saffari, Tan, Chan and Nada2018).
With the development of in vitro culture technology, normal fertilized oocytes can develop into viable blastocysts in vitro by day (D)5 or D6 after insemination. The most viable blastocyst is chosen for transfer to have a healthy live birth, which is beneficial for infertile patients. In the reported literature, there are inconsistencies regarding the influences of D5 and D6 blastocysts on the clinical outcomes of frozen embryo transfer (FET) cycles. A meta-analysis found that D5 and D6 cryopreserved blastocysts at the same developmental stage had similar clinical pregnancy and live birth rates (Sunkara et al., Reference Sunkara, Siozos, Bolton, Khalaf, Braude and El-Toukhy2010). El-Toukhy et al. (Reference El-Toukhy, Wharf, Walavalkar, Singh, Bolton, Khalaf and Braude2011) found that the rates of clinical pregnancy and live birth of high-grade blastocysts were not significantly different between D5 and D6 vitrified–warmed blastocysts. Kaye et al. (Reference Kaye, Will, Bartolucci, Nulsen, Benadiva and Engmann2017) reported that single high-quality D6 blastocyst transfer can obtain a similar clinical pregnancy rate to that of D5 blastocyst transfer in FET cycles.
However, Poulsen et al. (Reference Poulsen, Ingerslev and Kirkegaard2017) reported that the implantation rate of D5 single blastocyst transfer was significantly higher compared with that of D6 single blastocyst transfer from fresh cycles. Several studies have shown that the rates of clinical pregnancy, implantation and live birth of D5 single blastocyst transfer were also significantly higher compared with those of D6 single blastocyst transfer in FET cycles (Ferreux et al., Reference Ferreux, Bourdon, Sallem, Santulli, Barraud-Lange, Le Foll, Maignien, Chapron, de Ziegler, Wolf and Pocate-Cheriet2018; Sciorio et al., Reference Sciorio, Thong and Pickering2018, Reference Sciorio, Thong and Pickering2019; Tubbing et al., Reference Tubbing, Shaw-Jackson, Ameye, Colin, Rozenberg and Autin2018). Two recent meta-analyses reported that the clinical pregnancy, implantation and live birth rates of D5 blastocyst transfer were significantly higher compared with those of D6 blastocyst transfer regardless of whether a fresh transfer or FET cycle was used (Bourdon et al., Reference Bourdon, Pocate-Cheriet, Finet de Bantel, Grzegorczyk-Martin, Amar Hoffet, Arbo, Poulain and Santulli2019; Li et al., Reference Li, Wang, Sun, Lv, Ge, Li and Zhou2020). These results may be related to the fact that the proportion of high-quality and euploid blastocysts on D5 was significantly higher compared with that on D6, which leads to better clinical outcomes (Minasi et al., Reference Minasi, Colasante, Riccio, Ruberti, Casciani, Scarselli, Spinella, Fiorentino, Varricchio and Greco2016; Barash et al., Reference Barash, Ivani, Willman, Rosenbluth, Wachs, Hinckley, Pittenger Reid and Weckstein2017; Zhao et al., Reference Zhao, Yu and Zhang2018). Therefore, the present study analyzed the effects of the day of trophectoderm (TE) biopsy and blastocyst grade on the clinical and neonatal outcomes of preimplantation genetic testing–frozen embryo transfer (PGT–FET) cycles.
Materials and Methods
Patients
All infertile couples signed informed consent for PGT–FET treatment from January 2017 to December 2019 before participating in the present study. Our retrospective study consisted of the D5 TE biopsy group (D5) and the D6 TE biopsy group (D6). Patients with D5 or D6 blastocyst transfers were included in this study. Patients with PGT of vitrified–warmed oocytes, embryos or blastocysts, and twin-pregnancy cycles were excluded from this study (Fig. 1). Our retrospective study included 146 newborns from 237 transfer cycles in the D5 group and 91 newborns from 232 transfer cycles in the D6 group.
Flow diagram of the study
Insemination and embryo culture
Controlled ovarian hyperstimulation and oocyte pick-up were performed in accordance with the routine operation process of our IVF centre. Oocyte denudation was performed 3 h after oocyte pick-up. MII oocytes were inseminated by ICSI 1 h after oocyte denudation. The injected oocytes were cultured in separated G1 microdroplets in a humidified incubator. Oocyte fertilization was checked 16–18 h post-insemination. The fertilized oocytes containing two pronuclei continued to be cultured in G1 microdroplets until D3 post-insemination. A small hole with a diameter of 12 µm was drilled into the zona pellucida of the cleavage stage embryos by means of a laser on the D3 morning post-insemination so that the TE cells could herniate out of the hole for biopsy.
Blastocyst grading, TE biopsy, and biopsied blastocyst vitrification
The Gardner scoring system was used for blastocyst grading (Gardner and Schoolcraft, Reference Gardner and Schoolcraft1999). In our retrospective study, blastocysts with a score > BB, BB and < BB were considered good, fair and poor quality, respectively. TE biopsy was performed on D5 or D6 post-insemination. In total, 5–10 TE cells were cut using a laser and transferred into 200-µl PCR tubes for genetic analysis. A Vitrification Kit (KITAZATO) was used for vitrification of the biopsied blastocysts.
PGT procedure
The PGT procedure was performed with next-generation sequencing (NGS) on a MiSeqDx system (Illumina). Whole genome amplification (WGA) (SurePlex DNA Amplification System, Illumina), library construction (TG DNA Library Prep Kit, Illumina), sequencing (MiSeqTM DX Reagent Kit v3, Illumina), and sequencing data analysis were performed in strict accordance with the manufacturer’s instructions. The PGT cycles from monogenic disorders were diagnosed using NGS-based haplotyping, which was described in detail in our previously published literature (Chen et al., Reference Chen, Diao, Xu, Zhou, Wang, Li, Yan and Sun2016, Reference Chen, Diao, Xu, Zhou, Yan and Sun2017, Reference Chen, Diao, Xu, Zhou, Yan and Sun2019).
FET treatment
Preparation of the endometrium in FET cycles was performed by hormone replacement therapy (HRT), mild stimulation and gonadotropin (Gn) stimulation cycles. Warming of vitrified blastocysts was performed using a Thawing Kit (KITAZATO) on the morning of D6 progesterone administration. The vitrified–thawed blastocysts were transferred into the uterus 2 h after warming.
Definition
Clinical pregnancy, miscarriage, live birth, gestational age, preterm birth, and LBW were defined according to the reported literature (Zegers-Hochschild et al., Reference Zegers-Hochschild, Adamson, Dyer, Racowsky, de Mouzon, Sokol, Rienzi, Sunde, Schmidt, Cooke, Simpson and van der Poel2017).
Follow-up
A gestational sac with a fetal heartbeat scanned by ultrasound on D28 after blastocyst transfer indicated a clinical pregnancy. Data on neonatal outcomes, including date of birth, sex, birth weight and live birth, were obtained after birth.
Statistical analysis
SPSS Statistics 22.0 software was used for data analysis. Independent samples Mann–Whitney U-test or Student’s t-test was used to analyze female age, female body mass index (BMI), thickness of endometrium, gestational age and birth weight between the two groups. Data on endometrial preparation, categories of PGT, grade of transferred blastocysts, and the rates of implantation, miscarriage, live birth, infant sex, preterm birth and LBW were analyzed using the chi-squared (χ2) test. Logistic regression analysis after adjusting for confounding factors (female age, female BMI, endometrial preparation, thickness of endometrium, categories of PGT) was used to analyze the association between the implantation and live birth rates and the day of TE biopsy and blastocyst grade. A P-value less than 0.05 means a significant difference.
Results
Female age, endometrial preparation, thickness of endometrium, and categories of PGT in the D5 TE biopsy group were not significantly different compared with that in the D6 TE biopsy group. Female BMI in the D5 TE biopsy group was significantly lower compared with that in the D6 TE biopsy group. The grade of transferred blastocysts in the D5 TE biopsy group was significantly higher compared with that in the D6 TE biopsy group, especially for good quality blastocysts (28.3% vs. 10.3%, P < 0.001, Table 1).
Maternal and cycle characteristics according to D5 and D6 trophectoderm biopsy
a Values in parenthesis are expressed in percentage.
b<0.001.
c0.013.
BMI, body mass index; Gn, gonadotropin; HRT, hormone replacement therapy; PGT, preimplantation genetic testing; PGT-A, PGT for aneuploidies; PGT-M, PGT for monogenic disorders; PGT-SR, PGT for chromosome structural rearrangements.
The implantation rate in the D5 TE biopsy group was significantly higher compared with that in the D6 TE biopsy group (67.9% vs. 46.6%, P < 0.001; Table 2). The odds ratio (OR) remained significant after adjusting for the day of TE biopsy (D5 vs. D6), female age, female BMI, endometrial preparation, thickness of endometrium, categories of PGT, and grade of transferred blastocysts [P < 0.001, adjusted OR (aOR) 2.2, 95% confidence interval (CI) 1.5–3.3; Table 3]. The implantation rate of good quality blastocysts was significantly higher than that of fair quality blastocysts (71.4% vs 59.0%, P = 0.045, Table 2; aOR 1.6, 95% CI 0.9–2.7; Table 3), and a similar result was obtained between good quality blastocysts vs. poor quality blastocysts (71.4% vs. 42.0%, P < 0.001, Table 2; aOR 2.6, 95% CI 1.4–4.9, Table 3) and fair quality blastocysts vs. poor quality blastocysts (59.0% vs. 42.0%, P = 0.003, Table 2; aOR 1.9, 95% CI 1.2–3.1, Table 3). Among good quality blastocysts, the implantation rate of the D5 blastocysts was not significantly different compared with that of the D6 blastocysts (73.1% vs. 66.7%, P = 0.602, Table 2; aOR 1.5, 95% CI 0.5–4.5, Table 3). However, among fair quality blastocysts or poor quality blastocysts, the implantation rate of the D5 blastocysts was significantly higher compared with that of the D6 blastocysts (67.2% vs 51.8%, P = 0.013, Table 2, aOR 2.1, 95% CI 1.2–3.5, Table 3; 62.2% vs 28.4%, P < 0.001, Table 2, aOR 3.7, 95% CI 1.6–8.7, Table 3).
Clinical outcomes according to D5 and D6 trophectoderm biopsy
a = 0.003, b = 0.024.
TE, trophectoderm.
Results of logistic regression analysis of implantation rate after adjusting for confounding factors according to maternal and cycle characteristics
aOR, adjusted odds ratio; CI, confidence interval.
The live birth rate of the D5 TE biopsy group was significantly higher compared with that of the D6 TE biopsy group (61.6% vs. 39.2%, P < 0.001, Table 2; aOR 2.4, 95% CI 1.6–3.6, Table 4). Similar to the implantation rate, there was a significant difference in the live birth rate between good quality blastocysts and poor quality blastocysts (59.3% vs. 39.3%, P = 0.005, Table 2; aOR 1.6, 95% CI 0.8–2.9, Table 4) and between fair quality blastocysts and poor quality blastocysts (52.3% vs. 39.3%, P = 0.024, Table 2; aOR 1.6, 95% CI 1.0–2.6, Table 4). The live birth rate of good quality and fair quality blastocysts was not significantly different (59.3% vs. 52.3%, P = 0.273, Table 2; aOR 1.2, 95% CI 0.7–2.0, Table 4). Among good quality blastocysts, the live birth rate of the D5 TE biopsy group was not significantly different compared with that of the D6 TE biopsy group (64.2% vs. 45.8%, P = 0.148, Table 2; aOR 2.1, 95% CI 0.8–5.8, Table 4). Among fair and poor quality blastocysts, the live birth rate of the D5 TE biopsy group was significantly higher compared with that of the D6 TE biopsy group (60.0% vs. 45.4%, P = 0.020, Table 2; aOR 2.0, 95% CI 1.2–3.3, Table 4; 62.2% vs. 23.9%, P < 0.001, Table 2; aOR 4.4, 95% CI 1.9–10.4, Table 4).
Results of logistic regression analysis of live birth rate according to maternal and cycle characteristics
aOR, adjusted odds ratio; CI, confidence interval.
The miscarriage rates of the D5 TE biopsy and D6 TE biopsy groups were not significantly different (9.3% vs. 15.7%, P = 0.126, Table 2). At the same time, the miscarriage rate was not significantly affected by the grade of transferred blastocysts (Table 2). Moreover, the rates of infant sex, preterm birth and LBW, gestational age and birth weight of infants of the D5 TE biopsy group were not significantly different compared with that of the D6 TE biopsy group (Table 5).
Neonatal outcomes according to D5 and D6 trophectoderm biopsy
a Values in parentheses are expressed in percentage.
LBW, low birthweight.
Discussion
The present study showed that the PGT–FET cycles involving better-grade blastocysts obtained higher implantation and live birth rates. The implantation and live birth rates of D5 TE biopsy were superior to those of D6 TE biopsy for similarly graded blastocysts. The blastocyst grade and the day of TE biopsy did not significantly affect the miscarriage rate. The neonatal outcomes, including sex, gestational age, preterm birth, birth weight, and LBW of newborns, were not significantly different between D5 TE biopsy and D6 TE biopsy.
Minasi et al. (Reference Minasi, Colasante, Riccio, Ruberti, Casciani, Scarselli, Spinella, Fiorentino, Varricchio and Greco2016) reported that the euploid rate of blastocysts with top-quality ICM was significantly higher than that of blastocysts with poor quality ICM, and the similar results were obtained for blastocysts with high-quality TE compared with blastocysts with poor quality TE; that is, a higher blastocyst grade will lead to a higher euploid rate for blastocysts. Ozgur et al. (Reference Ozgur, Berkkanoglu, Bulut, Yoruk, Candurmaz and Coetzee2019) found that the live birth rate was not significantly different between single best-scoring and unknown-ploidy blastocyst transfer and single best euploid blastocyst transfer from infertile patients who were no more than 35 years old. Viñals Gonzalez et al. (Reference Viñals Gonzalez, Odia, Naja, Serhal, Saab, Seshadri and Ben-Nagi2019) reported that the euploid rate of blastocysts was associated with blastocyst grade, while the rates of implantation and live birth were not significantly affected by blastocyst morphology in patients of advanced maternal age who had preimplantation genetic testing for aneuploidy cycles. However, our retrospective study found that the rates of implantation and live birth were related to blastocyst grade, and higher quality blastocysts could obtain better implantation and live birth rates, which was consistent with two other studies (Zhao et al., Reference Zhao, Yu and Zhang2018; Irani et al., Reference Irani, O’Neill, Palermo, Xu, Zhang, Qin, Zhan, Clarke, Ye, Zaninovic and Rosenwaks2018).
In addition to blastocyst morphology, the speed of blastocyst development significantly affected clinical outcomes. Franasiak et al. (Reference Franasiak, Forman, Patounakis, Hong, Werner, Upham and Scott2018) reported that the sustained implantation rate in slowly blastulating embryos on D5 was significantly lower than that in normally blastulating embryos regardless of age in fresh cycles. At the same time, D6 fresh embryo transfer also had a significantly lower sustained implantation rate in slowly blastulating embryos on D5 than that in normally blastulating embryos, despite the blastocyst morphology grade being equivalent when the embryo transferred. The sustained implantation rate in the FET cycle was not significantly different between slowly blastulating embryos and normally blastulating embryos. The above results suggest that the different sustained implantation rates in the fresh cycle between slowly blastulating embryos and normally blastulating embryos on D5 are due to desynchrony between the embryo and endometrium. However, two new meta-analyses reported that the implantation, clinical and live birth rates of D5 blastocyst transfer were significantly higher compared with those of D6 blastocyst transfer regardless of whether a fresh and frozen transfer cycle was used (Bourdon et al., Reference Bourdon, Pocate-Cheriet, Finet de Bantel, Grzegorczyk-Martin, Amar Hoffet, Arbo, Poulain and Santulli2019; Li et al., Reference Li, Wang, Sun, Lv, Ge, Li and Zhou2020). This may be related to the euploidy rate being significantly higher among D5 blastocysts compared with among D6 blastocysts, which can result in better clinical outcomes. Our retrospective study showed that the implantation and live birth rates of D5 euploid blastocyst transfer were significantly higher compared with those of D6 euploid blastocyst transfer from PGT–FET cycles, which was similar to the reported study (Irani et al., Reference Irani, O’Neill, Palermo, Xu, Zhang, Qin, Zhan, Clarke, Ye, Zaninovic and Rosenwaks2018).
It has been reported that the clinical pregnancy and live birth rates of high-quality and high-grade blastocysts were comparable among D5 and D6 vitrified–warmed blastocysts (El-Toukhy et al., Reference El-Toukhy, Wharf, Walavalkar, Singh, Bolton, Khalaf and Braude2011; Kaye et al., Reference Kaye, Will, Bartolucci, Nulsen, Benadiva and Engmann2017). Similar to the reports presented above, the implantation and live birth rates from good quality blastocysts were not significantly different between D5 and D6 euploid blastocyst transfers in our retrospective study. The present study also found that the implantation and live birth rates of the D5 euploid blastocyst transfers were higher than those of the D6 euploid blastocyst transfers for similarly graded euploid blastocysts, which was consistent with Irani’s study (Irani et al., Reference Irani, O’Neill, Palermo, Xu, Zhang, Qin, Zhan, Clarke, Ye, Zaninovic and Rosenwaks2018). There may also be some embryonic intrinsic factors, such as RNA expression, metabolic differences or epigenetic differences resulting in superior clinical outcomes for D5 euploid blastocysts compared with D6 euploid blastocysts.
Aneuploidy of blastocysts is related to miscarriage. It has been reported that there was no significant difference in the miscarriage rates between D5 and D6 euploid blastocyst transfer cycles (Hernandez-Nieto et al., Reference Hernandez-Nieto, Lee, Slifkin, Sandler, Copperman and Flisser2019). Similar to the literature reported above, the present study showed that blastocyst morphology grade and development speed did not significantly affect the miscarriage rate of either the D5 or D6 euploid blastocyst transfer cycle. A meta-analysis showed that there was no significant difference in perinatal outcomes between D5 and D6 blastocyst transfer cycles, while birth weight was associated with extended in vitro culture (Zeng et al., Reference Zeng, Su Qin, Wen, Xu and Duan2020). Our study showed that there was no significant difference in neonatal outcomes between the two groups.
In conclusion, both blastocyst grade and the day of TE biopsy should be considered simultaneously for euploid blastocyst transfer so that better clinical and neonatal outcomes can be obtained.
Financial support
The present study was funded by the Clinical Medicine Research Special Fund of Chinese Medical Association (17020220691).
Conflicts of interest
The authors declared no potential conflicts of interest.
Ethics statements
This retrospective study was approved by the Ethics Committee of Nanjing Drum Tower Hospital affiliated with Nanjing University Medical School (No. 2020–019).
