Developmental potential of surplus morulas with delayed and/or incomplete compaction after freezing-thawing procedures

Current sophisticated culture systems and freezing-thawing procedures have enhanced the application of blastocyst culturing and frozen embryo transfer (FET) [1, 2]. Blastocyst culturing can be used to select embryos that have undergone genome activation [3], and it avoids the risk of transferring potentially arrested embryos or high-order embryos. Morulas with delayed growth sometimes coexist with blastocysts, and surplus morulas are often present. Slow-developing embryos are placed in a low priority queue for transfer selection. There is still limited evidence regarding the optimal disposal of surplus morulas; options include fresh transfer, extended culture and transfer in subsequent FET cycles, among others (ex. discarding, which would be controversial) [4, 5].

The vitrification technique has been widely applied in zygotes with 2 pronuclei, 8-cell embryos, and blastocyst embryos, but little attention has been given to morulas in previous literature, especially concerning poor-quality morulas after vitrification. Few studies to date have focused on the developmental potential or freezing-thawing outcomes of embryos derived from the morula stage [3‐10]. The survival rate of human day 4 morulas after freezing and thawing has been examined in a study addressing the morphological alterations and their applications in embryo selection [10]. The freezing-thawing strategy for surplus morulas at our institution was modified after 2015. Previously, after fresh blastocyst transfer, the morulas with delayed growth were cultivated until they reached the blastocyst stage or were cryopreserved. At present, to determine their growth potential, surplus morulas were thawed 1 day before FET. For the infertile couples, cryopreserving surplus morulas at the same time that blastocysts are cryopreserved is plausible, and they did not pay extra fees for that process at our institution.

Fragmentation has been associated with poor blastocyst formation and implantation rates, as well as chromosomal abnormalities [3, 11‐14]. Little is known about the outcomes resulting from the use of pervasive fragmented morulas in fresh and FET cycles. The effects of delayed compaction and fragmentation on the developmental capacity of morulas in fresh cycles have been determined [9]; the finding was consistent with our experiences [15, 16], except for the day 5 heavily fragmented morulas, which had a low blastocyst formation rate in our previous observation. To our knowledge, no study to date has reported the post-thaw BFR of morulas with either delayed compaction or heavy fragmentation.

The purpose of this study was to determine whether these poor-quality and slow-growing morulas are worthy of cryopreservation by exploring their developmental capacity after freezing-thawing procedures.

This is a retrospective, observational, and proof-of-concept study. A total of 1033 day 5/6 surplus morulas were cryopreserved from January 2015 to December 2018. Couples with vitrified surplus morulas who had returned for at least one FET cycle were included. The study included a total of 167 women undergoing 180 FET cycles. All couples completed the standard infertility workup; couples were not excluded based on age, sperm parameters or causes of infertility.

The study demonstrated the feasibility of freezing poor-quality surplus morulas on day 5/day 6. Following thawing of the morulas and an additional day of culture, satisfactory survival and blastocyst formation rates were observed.

To our knowledge, the developmental potential of delayed or incompletely compacted morulas in frozen-thawed cycles has never been systematically described. This pilot study revealed results obtained from surplus morulas, which showed a BFR as high as 66.6%; this rate is comparable to the BFR of normally developed day 4 morulas during fresh cycles (68.5%) [3]. Our study found that over one-half of heavily fragmented morulas (M3) could develop into blastocysts, which showed a higher BFR than what has been previously reported. In 2011, Ivec M et al. found the BFR of day 5 morulas to be 84% in the fresh cycle [9]. In poor-quality day 5 morulas (> 20% fragments), the optimal (5AA at least) blastocyst formation rate was 13.6% [9]. Our surplus poor-quality day 5 morulas (> 10% fragments) had a 16.7% top-quality (3AA at least) blastocyst formation rate after the thawing procedure. The study details from previous publications discussing different conditions in morulas are summarized along with our pilot study results (See Additional file 1: Table S1).

Since the freezing-thawing strategy was modified, we focused on the development of heavily fragmented morulas following this procedure. The degree of embryo fragmentation is closely related to chromosome anomalies [14]. Fragmentation of cleavage stage embryos is an indicator of poor blastocyst development [11]. The presence of significant numbers of fragments, particularly in conjunction with discrepancies in blastomere symmetry, substantially reduces embryo viability and has a negative impact on clinical outcomes. Studies have suggested that if cellular fragmentation leads to apoptosis or limits the rate of blastomere cleavage, the removal of these fragments might improve cell division and implantation [13, 26]. The selection of human embryos that survive cryopreservation and continue to undergo cleavage in vitro has been shown to significantly improve the delivery rate [27]. It can be hypothesized that anucleate cell fragments that exhibit erratic distribution of the cell adhesion protein E-cadherin cannot survive the freezing-thawing procedures [13]. Moreover, laser-assisted hatching may benefit morulas by expelling fragmented debris and may improve cell division and blastocyst formation [7, 28]. We discovered that the fragments can be removed/released more easily with laser-assisted hatching after thawing (Fig. 1). Further study should explore whether this phenomenon can influence outcomes.

It may be an option to let the delayed morulas declare their growth capacity in fresh cycles, with increasing literature having demonstrated that day 6/7 blastocysts achieved live births [29‐32]. However, concerning the earlier implantation window in the stimulated cycle, the slow-developing embryos often miss the synchronized endometrium and are cryopreserved [4, 5, 29]. Morula stage embryos have undergone genome activation [3]; referring to the adequate post-thaw BFRs, our policy of cryopreserving surplus morulas enabled the laboratory process and offered a flexible schedule for the following FET for both the couple and the clinicians. The manipulations and medical records of a couple can be completed on the same day in most circumstances, which reduces mistakes. The morulas used in these studies were surplus, which indicated that most of the couples had other embryos frozen. The couples did not pay extra fees for vitrification and storage at our institution. Further randomized controlled studies may be required to compare the BFRs of day 5/6 morulas in fresh and FET cycles and to explore the pregnancy outcomes following extended culture of surplus morulas after undergoing the freezing-thawing procedure (See Additional file 2: Table S2).

Unlike the day 3 cleavage and day 5 blastocyst stages, no consensus has been reached about the embryo morphological scoring criteria for morulas [1, 3, 8, 9, 15, 21, 28, 33]. This may have contributed to the neglect of morulas in research. A consensus has been reached about morphologic criteria at the blastocyst stage, with the transfer of good-quality blastocysts resulting in better implantation potential and better clinical outcomes than what is achieved with other blastocysts [34‐36]. A sophisticated scoring system for morulas had a better predictive value than the traditional scoring system based on the day 3 cleavage stage [3]. The grading system proposed in the present study was based on compaction and fragmentation percentages. The BFR and top-quality BFR were significantly different and had a decreasing trend among M1 to M3 in day 5 morulas (Table 2). The predictive value for blastocyst formation was good, and the scoring principle was easy to remember (Table 3).

A higher aneuploidy rate has been reported in slow-growing embryos [37]. Expanded blastocysts at day 5 had a higher chance of live birth rate than those at day 6 after thawed blastocyst transfer [38]. The embryo expansion day as well as the morphology need to be given full consideration when selecting embryos. During the study period, 18 day 6 morulas were thawed. Only one top-quality blastocyst was formed from these morulas, in which no positive pregnancy outcomes were reported. The aneuploidy rate of day 6 compacted morulas obtained during fresh cycles from women aged > 38 years was 97% [31]. Further studies into the cost-effectiveness of biopsy and/or the vitrification of day 6 morulas are required, and the decision-making process should be shared with the patients.

In our study, we analyzed the post-thaw developmental potential of slow-growing poor-quality morulas. We graded the morulas and found a linear trend of the BFR among M1, M2 and M3. However, this study had several limitations. First, embryo viability was assessed by morphology alone. Additional methods, such as morphokinetics and metabolism, may be incorporated into this strategy. Second, the embryo assessment is subjected to interoperator variability, and until now, there have been no objective methods or commercial tools available for measuring embryo fragments. Third, preimplantation genetic screening (PGS) was not applied to our cohort; therefore, the euploidy rate of these surplus morulas was unclear. Fourth, the cost-effectiveness of this strategy was difficult to define, so further randomized controlled trials may be required (see Additional file 2: Table S2).

This proof-of-concept study indicates that the described strategy should not be overlooked and may be applicable to the schedule of high-volume IVF laboratories (i.e., the surplus embryos can be cryopreserved on the same day with fresh embryo transfer and weekend work can be avoided). In clinical practice, programming is very convenient, both for physicians by facilitating their work schedules and for patients by enabling them to plan appointments. We suggest that the morulas be cryopreserved in separate containers from blastocysts to facilitate single embryo transfer (SET) or further cultivation. Regarding the high top-quality BFR of day 5 M1 morulas, we suggest cryopreserving them individually, which will prevent revitrification. In this study, we did not follow through with SET because the average age of the women was 35.8 years (> 35 years), there was a lack of PGS, and most of the outcomes from the fresh cycles proved to be a failure.

The post-thaw blastocyst formation rate was satisfactory, with approximately one-half of heavily fragmented morulas (M3) developing into blastocysts. Most of the poor-quality morulas were worth to freeze, with the reasonable goal of obtaining pregnancy and live birth. This alternative strategy may be a feasible approach for coping with poor-quality surplus morulas in non-PGS cycles.