This retrospective study is the first to compare the pregnancy outcomes between poor-quality D5 blastocysts and high-quality D6 blastocysts and to explore the transfer strategies of D6 high-quality blastocysts stratified by 35 years. We hope to provide more appropriate strategies for existing transfer strategies. Our results showed that the live birth rate of single poor-quality D5 blastocysts was higher than that of single high-quality D6 blastocysts, but the difference did not reach statistical significance for patients aged < 35 years (35.48% vs. 31.13%, p > 0.05), and the same trend was observed for ≥ 35-year-old patients (29.09% vs. 21.28%, p > 0.05). Additionally, the live birth and multiple pregnancy rates of D6-SBT were significantly lower than those of D6 double blastocyst transfer (DBT).
SBT is an effective method for reducing the risk of multiple pregnancies. To evaluate whether patients are suitable for SBT, many factors that affect the pregnancy outcomes of blastocyst transfer cycles must be comprehensively considered, including age, blastocyst developmental speed, and morphological grading. Numerous studies have investigated the effect of blastocyst developmental speed on pregnancy outcomes and showed that the rates of ongoing pregnancy and live birth of D5 blastocysts were significantly higher than those of D6 blastocysts (18, 19), which suggested that the timing of blastulation was considered as a stable predictor of live birth for frozen-thawed SBT (6). A meta-analysis also showed that D5 blastocysts were superior to D6 blastocysts in both fresh and frozen transfer cycles in clinical practice (10, 11). When transferring the same quality blastocysts, the live birth rate of D5 blastocysts in FET cycles was still significantly higher than that of D6 blastocysts (9, 20). Additionally, Yerushalmi et al. found that D5 blastocysts in the oocyte donation program resulted in significantly better clinical outcomes and shorter time to delivery than D6 blastocysts, regardless of embryo quality (21). One potential mechanism is that blastocyst developmental speed is associated with the incidence of abnormal spindle morphology. Hashimoto et al. showed that the incidence of abnormal spindles in the D6 blastocysts was significantly higher than that in D5 blastocysts (47% vs. 30%, p < 0.05), and the implantation competence was decreased in D6 blastocysts compared with D5 blastocysts when a total of 533 spindles were analyzed on days 5 and 6 vitrified-warmed blastocysts (22).
Another possible reason is that earlier blastocyst development is independently associated with a higher likelihood of embryonic euploidy, which supports the preferential selection of normally developing D5 blastocysts for transfer to improve pregnancy outcomes compared to growth-retarded D6 blastocysts among unbiopsied embryos (23, 24). However, Kaye et al. found that D6 blastocysts resulted in similar clinical and ongoing pregnancy rates compared to those of D5 blastocysts after vitrification (25). Yang et al. showed that for unbiopsied embryos, the clinical pregnancy rate of high-quality D5 blastocysts did not differ significantly from that of high-quality D6 blastocysts (26). Similarly, for biopsied embryos, D5 euploid blastocysts showed no significant difference in implantation potential and early miscarriage rate compared with similarly graded day 6 euploid blastocysts (12). Nevertheless, a study by Irani et al. showed that D5 blastocysts yielded a significantly higher live birth rate than similarly graded euploid D6 blastocysts (27). Therefore, there is no consensus on the effect of blastocyst developmental speed on pregnancy outcomes, which may be caused by variant populations included and different quality of blastocysts transferred, as well as different blastocyst culture strategies and methods in different centers.
In addition to aneuploidy testing and blastocyst development rate, blastocyst morphological grading is critical for selecting optimal embryos with a good development potential for implantation (27). For unbiopsied blastocyst transfer cycles, high-quality blastocysts were associated with a significantly higher live birth rate than low-quality blastocysts (20). Our previous study is consistent with this conclusion and showed that the live birth rate of good-quality D5-SBT was significantly higher than that of low-quality D5-SBT (15). These conclusions may be explained by the higher euploidy rates of high-quality blastocysts than low-quality blastocysts (28). In cases of euploid blastocyst transfer, the live birth rates of high-quality, average-quality, and poor-quality blastocysts were 67.8%, 53.4%, and 29.5%, respectively, with a statistically significant difference (27). Additionally, the clinical pregnancy rate of high-quality blastocysts was significantly higher than that of poor-quality blastocysts (54.5% vs. 25.0%, p = 0.034), and similar trend also exists with blastocysts at the same developmental speed (12). Therefore, these studies suggest that, whether embryo biopsy is performed or not, blastocysts with high morphological grading should be preferentially selected for transfer to obtain better pregnancy outcomes. However, how can SBT be performed for better pregnancy outcomes when faced with D5 poor-quality and D6 high-quality blastocysts? Our results showed that the live birth rate of D5 poor-quality blastocysts was higher than that of D6 high-quality blastocysts for patients aged < 35 years (35.48% vs. 31.13%, p > 0.05), and the same trend was observed for patients aged ≥ 35 years (29.09% vs. 21.28%, p > 0.05), suggesting that D5 poor-quality blastocysts may be preferred, especially for advanced age patients, although there was no statistical difference.
To date, only a limited number of studies have explored transfer strategies for high-quality D6 blastocysts. A study by Kang et al. showed that D6-SBT produced significantly lower clinical outcomes compared with D6-DBT (23.4% vs. 51.6%, p = 0.001) when patients were transferred with high-quality blastocysts in FET cycles (29). However, this study did not analyze results stratified by age, a factor that has a significant impact on pregnancy outcomes. Kaing et al. showed that older maternal age (p < 0.0001) is independently associated with a lower euploidy percentage (23). Another study suggested that the quality of blastocysts degraded and the proportion of high-quality blastocysts decreased as age increased (14). When blastocysts developed into the same quality, clinical pregnancy and live birth rates tended to decrease with increasing age (14). Therefore, multiple embryo transfers are considered to improve pregnancy outcomes in elderly patients. Our previous study showed that SBT of D5 blastocysts could be recommended for advanced women, owing to significantly reduced multiple pregnancy rate and acceptable live birth rate compared to DBT, regardless of blastocyst quality (15). For poor-quality D6 blastocysts, DBT is recommended for transfer to improve the live birth rate (17). However, it is unclear whether SBT is adequately recommended for high-quality D6 blastocysts, especially in elderly patients. This study showed that for patients aged < 35 years, SBT of good-quality D6 blastocysts can be suggested for a live birth rate of 31.13%, and the multiple pregnancy rate is significantly lower than that of DBT. While for patients ≥ 35 years of age, a combination of poor-quality D6 blastocysts with high-quality D6 blastocysts could be considered because the live birth rate of SBT of good-quality D6 is 21.28%, but patients should be informed that this transfer strategy was associated with increased risk of multiple pregnancies. The DBT of high-quality D6 blastocysts is not suggested regardless of age due to significantly increased multiple pregnancy rates, 32% for patients aged < 35 years, and 40% for patients ≥ 35 years of age.
There is no consensus regarding the effects of blastocyst developmental speed and morphological grading on neonatal outcomes. Previous studies have shown that D6 blastocyst transfer is associated with increased birth weight after FET compared with that in the D5 group (9, 30), but there was no difference in gestational age between D5 and D6 blastocysts (30). Meanwhile, the conclusion of the meta-analysis was consistent with the above-mentioned study (31). However, Wang et al. showed that birth weight in singleton newborns of the D5 blastocyst group exhibited no significant difference from that of the D6 group (32). When blastocyst quality was considered, single poor-quality blastocyst transfer had significantly lower birth weight and gestation-adjusted birth weight than SBT of high-quality blastocysts during FET cycles (33). However, Oron et al. found that embryo quality was not associated with adverse obstetric and perinatal outcomes, with no significant differences in gestational age and birth weight between single good- or poor-quality embryo (34). Similar to the above-mentioned study, our study showed that the gestational age and birth weight of D5 poor-quality blastocyst transfer were not significantly different from those of D6 high-quality blastocyst, and this non-statistically significant result may be possibly due to the limited population included in this study. Larger-scale population studies are needed to confirm the effects of blastocyst developmental speed and quality on neonatal outcomes.
This study had some limitations. First, it was a retrospective study design; however, the baseline characteristics of the patients among the four groups were not statistically different, indicating that the populations included in this study among these groups were similar. Second, there was a relatively inadequate sample size of live births, resulting from patients aged ≥ 35 years, and the results from women of advanced age need to be verified in further large population studies. Third, patients over 40 years of age were not recruited in this study, which may prevent generalization of the conclusions. The main strength of this study was that the FET cycles were specifically grouped based on a combination of morphological grading and blastocyst developmental speed, and data were analyzed among groups stratified by 35 years of age, which may help improve existing blastocyst selection strategies