A quantitative analysis of the impact of cryopreservation on the implantation potential of human early cleavage stage embryos

D.H. Edgar1,2,3, H. Bourne1, A.L. Speirs1,2 and J.C. McBain1,2

1 Reproductive Biology Unit, Royal Women's Hospital, 132 Grattan Street, Carlton, Victoria 3053, and 2 Melbourne IVF, 320 Victoria Parade, East Melbourne, Victoria 3002, Australia


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The impact of cryopreservation on the implantation potential of early cleavage stage (day 2) embryos was assessed by analysing the outcome from > 5000 thawed embryos in relation to the outcome from a similar number of fresh embryos. Analysis of procedures in which all transferred embryos fulfilled equivalent defined criteria revealed no significant difference in the implantation rates (fetal hearts/100 embryos transferred) of fresh 4-cell embryos (16.6%) and fully intact thawed 4-cell embryos (16.9%). Although 2-cell embryos implanted at significantly lower rates, there was again no significant difference between fresh (6.5%) and fully intact thawed (7.2%) embryos. Similar analysis of all embryos (irrespective of cell number on day 2) demonstrated that the implantation potential of partially intact thawed embryos was related to the extent of blastomere loss with the implantation rate of embryos with 50% cell survival (5.4%) being approximately half the rate of fully intact embryos (11.3%). Combining the values obtained from `pure' data for the implantation rates of embryos with defined levels of survival with their relative prevalence in the total population of thawed embryos gave a predicted number of implantations (441) which was similar to the observed outcome (463). This number was ~30% less than the number expected had the same embryos been transferred fresh (635). The results suggest that intact thawed embryos have the same implantation potential as equivalent fresh embryos and that the impact of cryopreservation is limited to blastomere loss which is directly related to loss of implantation potential. The observed frequency of blastomere loss results in a reduction of ~30% in the implantation potential of a population of embryos following cryopreservation.

Key words: blastomere loss/cryopreservation/human embryos/implantation rate


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Cryopreservation of human embryos has been a routine component of clinical in-vitro fertilization (IVF) programmes for well over a decade (see review by Mandelbaum et al., 1998). In treatment cycles where ovarian stimulation in conjunction with IVF results in a large number of fertilized oocytes, it offers the opportunity to reduce the number of embryos transferred per procedure, and thereby restrict the risk of multiple pregnancy, while optimizing the clinical use of available material. In cases where the response to ovarian stimulation may place the patient at risk of ovarian hyperstimulation syndrome or where embryo transfer may be considered inadvisable for other reasons, it offers the option of freezing all the available embryos for thawing and transfer in subsequent, more favourable cycles. Decisions on how many embryos to transfer and how many to commit to cryopreservation, however, must also include consideration of any potential damage which may occur as a consequence of the freeze/thaw procedure and which may result in reduced embryonic viability.

In order to assess the impact of cryopreservation on human early cleavage stage embryos it is necessary to compare the relative outcome from fresh and frozen–thawed embryos. Such comparisons are often complicated by, or misinterpreted due to, difficulties in controlling for differences in the populations of embryos being assessed (Ludwig et al., 1998Go; Speirs, 1998Go; Testart, 1998Go).

In the present study we have compared the implantation rates (IR) of fresh early cleavage stage embryos with defined characteristics, in terms of growth rate and embryo quality (degree of fragmentation), to the IR of equivalent cryopreserved embryos in which no blastomere loss had occurred following thawing. Drawing on data from >5000 fresh and >5000 thawed embryos, we have circumvented problems associated with concomitant transfer of embryos with different characteristics by deriving our IR from those subsets of procedures in which only one embryo or only embryos with equivalent characteristics were transferred (`pure' data). In this way we were able to assess whether cryopreservation has an impact on early embryos which is independent of blastomere loss. We then used the same approach to assess the impact of defined levels of blastomere loss on implantation potential. The prevalence of fully intact, different categories of partially intact, and lysed embryos was then combined with the IR of each group (derived from `pure' data) to estimate the expected number of implantations from the entire population of thawed embryos and the value compared to the actual outcome. Having established a model for estimating the effect of cryopreservation on early cleavage stage embryos, we then applied it to our total population of embryos to estimate the impact in terms of `lost' implantations.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Embryos
The embryos included in this study were transferred in procedures over the period January 1996 to June 1998. Uniform methodology was employed in both our clinics over the period of the study. Ovarian stimulation, oocyte collection, and embryo culture were as described previously (Bourne et al., 1995Go) except that serum was replaced in the culture medium by human serum albumin (HSA; Albumex®-20, CSL Limited, Victoria, Australia) at a final concentration of 4 mg/ml. Embryos generated by both intracytoplasmic sperm injection (ICSI) and conventional insemination were included in the study. In agreement with previous findings (Kowalik et al., 1998Go), our unpublished observations demonstrate no significant difference in the survival or implantation rate of embryos generated by the two methods. Only embryos which were transferred or cryopreserved 2 days after oocyte collection and insemination (day 2) were included in the analysis. Prior to transfer or cryopreservation, all embryos were assessed for the number of blastomeres present and the degree of cytoplasmic fragmentation present. Embryos with <10% fragmentation were classified as grade A, 10–30% fragmentation as grade B and >30% fragmentation as grade C. During the course of the study 12 399 fresh embryos were generated in our laboratory of which 7339 (59.2%) were classified as grade A, 3641 (29.4%) were classified as grade B and only 1419 (11.4%) were classified as grade C. Only embryos classified as grades A or B were cryopreserved. In all, 5034 fresh embryos were included in 2839 embryo transfer procedures (1.77 per embryo transfer); 5572 embryos were thawed, of which 4720 were transferred in 2867 procedures (1.65 per embryo transfer).

Cryopreservation
Embryos were frozen using 1,2-propanediol and sucrose as the cryoprotectants (Lassalle et al., 1985Go). Freezing and thawing solutions consisted of cryprotectants in phosphate-buffered saline (PBS) supplemented with HSA (10 mg/ml in freezing solutions and 4 mg/ml in thawing solutions). Embryos were equilibrated in 1.5 mol/l propanediol for 10 min at room temperature before being transferred to 1.5 mol/l propanediol/0.1 mol/l sucrose and loaded individually into plastic straws. Cooling was carried out in programmable freezers (Kryo 10 Series; Planer Products, Sunbury-on Thames, UK) at a rate of –2°C/min to –8°C, at which point seeding was induced manually. Cooling was then continued at rates of –0.3°C/min to –30°C and –50°C/min to –150°C before plunging and storage in liquid nitrogen.

Thawing and transfer
Embryos were thawed rapidly by removing straws from storage, exposure to air for 30 s and immersion in a water bath at 30°C for 45 s. Propanediol was removed in three steps in the presence of 0.2 mol/l sucrose at room temperature. Rehydration was completed by incubation in sucrose-free PBS and embryos were transferred to culture medium at 37°C before being assessed for numbers of remaining blastomeres. Thawed embryos were transferred in cycles as previously described (Bourne et al., 1995Go).

Analysis of results
Implantation rates (IR) were calculated as the number of fetal heart beats detected by ultrasonic examination per 100 embryos transferred. Proportions of embryos which implanted were compared using Fisher's Exact Test.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The IR for the total population of embryos transferred fresh was 10.9% (547 out of 5034) which was not significantly different from the rate for all thawed/transferred embryos (9.8%; 463 out of 4720). This comparison, however, did not allow for differences in the populations of transferred embryos. For example, the denominator used to calculate the fresh rate included some poor quality (grade C) embryos which were transferred when no other embryos were available but which were not represented in the population of thawed embryos. It was, therefore, more valid to compare the IR for fresh embryos using only data from embryo transfers in which no grade C embryos were included (11.4%: 510 out of 4493). This rate was significantly (P < 0.05) higher than the rate for all thawed/transferred embryos (9.8%, see above). However even this, albeit more valid, comparison did not take into account other potential differences such as cryopreservation-associated blastomere loss, growth rates and extent of fragmentation. In order to correct for the above variables and also to determine whether cryopreservation exerted an effect on implantation potential independent of blastomere loss, we analysed the data from embryo transfers in which only fully intact embryos which fulfilled specific criteria were included.

The results of this analysis are shown in Table IGo. In order to be included in any category shown in this analysis, it was necessary for all embryos transferred to fulfil equivalent criteria. Therefore, the restricted subsets of embryos with defined cell numbers on day 2 did not include many embryos from transfer procedures containing embryos of mixed characteristics which could be included in the analysis of all embryos. To account for potential differences due to growth rate, data were separated into the most common developmental stages represented in fresh and thawed embryos, i.e. 4-cell and 2-cell. In addition, any impact of embryo quality (assessed by the extent of pre-freeze fragmentation) was corrected for by restricting the analysis to either only grade A embryos or embryos which were either grades A or B.


View this table:
[in this window]
[in a new window]
 
Table I. Pure data for implantation rates (%) of fresh versus fully intact thawed embryos
 
Having established that intact thawed embryos have the same implantation potential as their fresh counterparts (Table IGo), we then analysed the significance of blastomere loss in thawed early cleavage stage embryos. We were unable to detect any significant difference in the survival pattern of grade A compared with grade B embryos or of 4-cell compared with 2-cell embryos (data not shown). Using data from transfer procedures in which only partially intact embryos (irrespective of cell number) were transferred we established that the implantation rate was significantly lower than the rate for fully intact embryos, i.e. 6.2% (51 out of 824) compared with 11.3% (243 out of 2155); P < 0.001. To examine this further, we firstly established the prevalence of different degrees of blastomere loss in the entire population of thawed embryos (n = 5572). It can be seen from this data (Table IIGo) that over half of the thawed embryos survived cryopreservation with all their blastomeres intact and that ~15% of embryos lost all their cells and, therefore, could not be transferred. Between the two extremes were embryos which survived with half or more of their blastomeres intact (22.8%) and a relatively small proportion of embryos which survived with less than half of their cells intact (6.4%).


View this table:
[in this window]
[in a new window]
 
Table II. Blastomere loss in total population of thawed embryos (n = 5572)
 
Restricting analysis to procedures in which all transferred embryos were in equivalent blastomere survival categories allowed us to estimate the impact of defined levels of blastomere loss on implantation potential (Table IIIGo). This approach allowed us to establish that blastomere loss was related to a reduction in the implantation potential of thawed embryos. Thus, the IR of partially intact embryos in which half or more of the blastomeres survived (6.9%) was significantly (P = 0.002) lower than that of fully intact embryos (11.3%) but significantly (P < 0.02) higher than that of embryos with less than half of their blastomeres surviving (1.0%). In addition to the analysis shown in Table IIIGo it was also possible to estimate the IR of embryos in which exactly 50% of the blastomeres survived cryopreservation. Since only embryo transfer procedures in which all thawed embryos had lost exactly half of their original blastomeres could be included in this analysis using our `pure' approach, the numbers of embryos included in this group was relatively small. Interestingly however, the IR (5.4%; eight out of 149) was approximately half the value obtained for fully intact embryos (11.3%; Table IIIGo; P = 0.02).


View this table:
[in this window]
[in a new window]
 
Table III. Pure data for implantation rate versus blastomere survival in thawed embryos
 
In order to assess the accuracy of the values obtained for the IRs of intact and partially intact thawed embryos, we combined the IRs shown in Table IIIGo with the total numbers of embryos in each category as shown in Table IIGo. This allowed us to calculate the predicted number of implantations from the total population (5572) of thawed embryos and compare it to the actual value. Table IVGo shows that the predicted value (441) was very similar to the actual number of implantations (463). This analysis also suggested that ~80% (349 out of 441) of all implantations from thawed embryos were likely to be derived from embryos which survived cryopreservation without blastomere loss.


View this table:
[in this window]
[in a new window]
 
Table IV. Theoretical outcome compared with actual outcome from 5572 embryos
 
Given that the expected number of implantations from a population of 5572 fresh embryos would be 635 out of 5572 (i.e. 11.4%; Table IGo), it was estimated that the implantation potential of a population of embryos was reduced by ~30% by being subjected to cryopreservation (Table IVGo).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Do frozen embryos have the same developmental potential as fresh embryos? What is the impact of cryopreservation on a population of frozen embryos? The difficulties inherent in providing a simple answer to these questions have been clearly discussed (Ludwig et al., 1998Go; Speirs, 1998Go; Testart, 1998Go). Given the importance of the answer in helping to decide the optimal way to use embryos in a clinical assisted reproduction programme, it is essential that controlled studies attempt to address the confounding issues and provide a quantitative basis for such decisions.

Our results complement and extend a previous analysis (Mandelbaum et al., 1998Go) in which an implantation rate of 8.4% per transferred thawed embryo was reported together with a post-thaw survival rate of 73% from >14 000 thawed embryos in four centres over a 10 year period. They further estimated that embryo cryopreservation produced 8% additional births in women having transfers of fresh and frozen embryos. The present study has also attempted to address three questions. Firstly, what is the impact of cryopreservation on human early cleavage stage embryos if we exclude the most obvious effect, i.e. blastomere loss? Secondly, what is the prevalence of blastomere loss and the relationship between defined levels of blastomere loss and implantation potential? Thirdly, how do the answers to the first two questions interact to give us an indication of the overall impact of cryopreservation on a population of embryos? The crucial aspect of our approach was the estimation of the implantation potential of embryos with clearly defined characteristics by restriction of the analysis to embryo transfers which only included one type of embryo. Sufficient homogeneous data, which are necessary to permit unequivocal correlation of embryos with outcome, were available due to the large populations of embryos (>5000) in both the thawed and the control (fresh) group.

Three main points can be made from the data shown in Table IGo. Firstly, there is no significant difference in the IR of fresh and intact thawed embryos in any of the categories shown, suggesting that an embryo with defined characteristics which survives cryopreservation without loss of blastomeres has the same developmental potential as a similar fresh embryo. This would agree with previous studies (Mandelbaum et al., 1987; and also three of Selick (Selick et al., 1995Go), who randomly allocated donated oocytes to frozen and fresh embryo transfers in an attempt to overcome potential embryo quality bias associated with the results reported by Levran et al. (1990). Secondly, we observed a highly significant (P < 0.001) increase in the implantation rates of faster growing (4-cell) embryos relative to those of equivalent slower growing (2-cell) embryos in all categories examined. The association between growth rate and implantation potential in fresh embryos, which also persisted in frozen/thawed embryos in our study (Table IGo), confirms the previous findings of Ziebe et al. (1997), who concluded that cleavage stage was more important than fragmentation in predicting the developmental potential of fresh embryos. Thirdly, the extent of fragmentation within the limits investigated (0–30%) does not appear to have a significant effect on implantation potential since there was no significant difference in the IRs obtained in any category when data were derived from grade A embryos only. The results of Staessen et al., based on relatively small numbers of embryos, suggest that >20% fragmentation is associated with a reduction in implantation (Staessen et al., 1992Go). Similarly, the results of Ziebe et al. show a reduction in implantation rate with increasing levels of fragmentation, although the data in this study combine outcomes from all embryos with 10–50% fragmentation (Ziebe et al., 1997Go). The large data set in our study together with the above reports suggest that fragmentation may influence outcome only when it exceeds a critical threshold, which is difficult to define accurately given the small numbers in some studies, the ranges of fragmentation used in the studies, and the subjective nature of the quantitative assessment.

Although the exact significance of the role, if any, of oestrogen in implantation (Edgar, 1995Go) and the hormonal control of endometrial receptivity (de Ziegler, 1995Go) are still unclear, it is impossible to exclude completely the possibility that endometrial receptivity is altered in cycles in which thawed embryos are transferred relative to those in which fresh embryos are replaced (de Ziegler and Frydman, 1990Go). Such differences in receptivity may obscure differences in intrinsic embryo viability but it would seem to be unlikely that the balance of the factors would result in the close similarity seen between the IRs of equivalent fresh and fully intact thawed embryos.

The prevalence of fully intact thawed embryos in our study (55.5%; Table IIGo) and the proportion of thawed embryos which survived with at least 50% of their blastomeres intact (78.3%; Table IIGo) are at least consistent with the reported results from a large multicentre series (Mandelbaum et al., 1998Go) and other studies (Hartshorne et al., 1990Go; Horne et al., 1997Go; Kowalik et al., 1998Go). However, in contrast to the conclusions of Hartshorne et al. (1990) which were based on smaller numbers of embryos, we found a highly significant association between blastomere loss and reduction in implantation potential (Table IIIGo) even in embryos with >=50% of their blastomeres intact.

The conclusions drawn from `pure' data, that intact thawed embryos are equivalent to their fresh counterparts and that blastomere loss at each defined level is associated with a defined reduction in implantation potential, are strengthened by the similarity in the actual number of implantations derived from the total population of thawed embryos and the number predicted by applying the values from pure data to the total population (Table IVGo).

The overall impact of cryopreservation on our population of embryos is apparent from the difference in the number of post-thaw implantations and the number which would have been expected had it been possible to transfer all embryos fresh. The estimated loss is of the order of 30% of the potential implantations. It is, of course, impractical to consider transferring all embryos fresh in order to avoid these losses, although the results suggest that the impact of potential cryodamage should be considered in conjunction with the risk of multiple implantations when deciding how many embryos to transfer fresh and how many to freeze. Further assessment of post-thaw viability based on resumption of cleavage (Van der Elst et al., 1997Go; Ziebe et al., 1998Go) can also help in deciding how to use thawed embryos. Our own unpublished observations, however, agree with the findings of the latter group that pregnancies and births can result following the transfer of embryos which do not resume cleavage, albeit at a lower frequency than following the transfer of cleaving embryos.

The results further emphasize the need to develop and apply methods such as blastocyst culture and transfer (Gardner et al., 1998Go) and preimplantation genetic diagnosis of aneuploidy (Gianaroli et al., 1997Go) which permit the selection of embryos with enhanced developmental potential and reduce the chance of subjecting them to the potentially deleterious effects of cryopreservation. Although Hardy et al. (1990) have reported the lack of an adverse effect on human preimplantation development in vitro following biopsy at the 8-cell stage, our results from thawed embryos with >=50% of their blastomeres intact suggest that loss of a minority of cells at early cleavage stages may have an impact on subsequent implantation potential. It is, therefore, important to stress that comparisons of alternative strategies for embryo utilization must include a quantitative assessment of how each aspect of the approach, e.g. embryo biopsy, extended culture or cryopreservation at early or later (blastocyst) stages, will be likely to affect the final outcome from a population of fertilized oocytes.


    Notes
 
3 To whom correspondence should be addressed at: Reproductive Biology Unit, Royal Women's Hospital, 132 Grattan Street, Carlton, Victoria 3053, Australia Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Bourne, H., Richings, N., Harari, O. et al. (1995) The use of intracytoplasmic sperm injection for the treatment of severe and extreme male infertility. Reprod. Fertil. Dev., 7, 237–245.[ISI][Medline]

de Ziegler, D. (1995) Hormonal control of endometrial receptivity. Hum. Reprod., 10, 4–7.[ISI][Medline]

de Ziegler, D. and Frydman, R. (1990) Different implantation rates after transfers of cryopreserved embryos originating from donated oocytes or from regular in vitro fertilization. Fertil. Steril., 54, 682–688.[ISI][Medline]

Edgar, D.H. (1995) Oestrogen and human implantation. Hum. Reprod., 10, 2–4.[ISI][Medline]

Gardner, D.K., Schoolcraft, W.B., Wagley, L. et al. (1998) A prospective randomized trial of blastocyst culture and transfer in in-vitro fertilization. Hum. Reprod., 13, 3434–3440.[Abstract]

Gianaroli, L., Magli, M.C., Ferraretti, A.P. et al. (1997) Preimplantation genetic diagnosis increases the implantation rate in human in vitro fertilization by avoiding the transfer of chromosomally abnormal embryos. Fertil. Steril., 68, 1128–1131.[ISI][Medline]

Hardy, K., Martin, K.L., Leese, H.J. et al. (1990) Human preimplantation development in vitro is not adversely affected by biopsy at the 8-cell stage. Hum. Reprod., 5, 708–714.[Abstract]

Hartshorne, G.M., Wick, K., Elder, K. et al. (1990) Effect of cell number at freezing upon survival and viability of cleaving embryos generated from stimulated IVF cycles. Hum. Reprod., 5, 857–861.[Abstract]

Horne, G., Crichlow, J.D., Newman, M.C. et al. (1997) A prospective evaluation of cryopreservation strategies in a two-embryo transfer programme. Hum. Reprod., 12, 542–547.[ISI][Medline]

Kowalik, A., Palermo, G.D., Barmat, L. et al. (1998) Comparison of clinical outcome after cryopreservation of embryos obtained from intracytoplasmic sperm injection and in-vitro fertilization. Hum. Reprod., 13, 2848–2851.[Abstract/Free Full Text]

Lassalle, B., Testart, J. and Renard, J.-P (1985) Human embryo features that influence the success of cryopreservation with the use of 1,2 propanediol. Fertil. Steril., 44, 645–651.[ISI][Medline]

Levran, D., Dor, J., Rudak, E. et al. (1990) Pregnancy potential of human oocytes – the effect of cryopreservation. N. Engl. J. Med., 323, 1153–1156.[Abstract]

Ludwig, M., Al-Hasani, S., Felberbaum, R. et al. (1998) No impact of cryopreservation on embryo developmental potential – one more example of the problems of retrospective, non-controlled data. Hum. Reprod., 13, 786–787.[ISI][Medline]

Mandelbaum, J., Junka, A.M., Plachot, M. et al. (1987) Human embryo cryopreservation, extrinsic and intrinsic parameters of success. Hum. Reprod., 2, 709–715.[Abstract]

Mandelbaum, J., Belaisch-Allart, J., Junka, A.-M et al. (1998) Cryopreservation in human assisted reproduction is now routine for embryos but remains a research procedure for oocytes. Hum. Reprod., 13 (Suppl. 3), 161–174.[Abstract]

Selick, S.E., Hofmann, G.E., Albano, C. et al. (1995) Embryo quality and pregnancy potential of fresh compared with frozen embryos – is freezing detrimental to high quality embryos? Hum. Reprod., 10, 392–395.[Abstract]

Speirs, A.L. (1998) Aspects of epigenetic factors in artificial procreation. Hum. Reprod., 13, 787–788.[ISI][Medline]

Staessen, C., Camus, M., Bollen, N. et al. (1992) The relationship between embryo quality and the occurrence of multiple pregnancies. Fertil. Steril., 57, 626–630.[ISI][Medline]

Testart, J. (1998) Episcientific aspects of the epigenetic factors in artificial procreation. Hum. Reprod., 13, 783–785.[Free Full Text]

Van der Elst, J., Van den Abbeel, E., Vitrier, S. et al. (1997) Selective transfer of embryos with further cleavage after thawing increases delivery and implantation rates. Hum. Reprod., 12, 1513–1521.[Abstract]

Ziebe, S., Peterson, K., Lindenberg, S. et al. (1997) Embryo morphology or cleavage stage: how to select the best embryos for transfer after in-vitro fertilization. Hum. Reprod., 12, 1545–1549.[Abstract]

Ziebe, S., Bech, B., Petersen, K. et al. (1998) Resumption of mitosis during post-thaw culture: a key parameter in selecting the right embryos for transfer. Hum. Reprod., 13, 178–181.[Abstract]

Submitted on June 28, 1999; accepted on October 5, 1999.