Is the success of human IVF more a matter of genetics and evolution than growing blastocysts?

R.G. Edwards1 and Helen K. Beard

Moor Barns Farmhouse, Madingley Road, Coton, Cambridge CB3 7PG, UK

Quality is clearly deficient in many human embryos growing in vitro. Fewer than 20% implant after transfer to the uterus in the majority of patients, so two or more are replaced which results in high-order multiple pregnancies. Among many approaches to improving quality, co-culture with supporting cells has claimed to be successful (e.g. Bongso et al., 1994Go), although it is now being replaced by the use of sequential culture media (Ménézo et al., 1998Go). Zona drilling was also introduced to enhance implantation rates (Cohen et al., 1990Go). In recent studies, co-cultures combined with assisted hatching failed to improve implantation rates above 23–26% per embryo, which were overall similar to controls (Hu et al., 1998Go). Perhaps variations between clinics account for differences in the success of these approaches.

Growth and transfer of blastocysts, an approach widely used in domestic and laboratory animal species, is now being recommended by several authors, e.g. Bavister and Boatman (1997), Gardner and Lane (1998) and Tsirigotis (1998). Potentially, high implantation rates per embryo seem to be possible, allowing the transfer of only one or two embryos thereby avoiding multiple pregnancies. Blastocyst transfer is claimed to be more physiological than pronucleate and 2-cell transfers. This principle led to the first-ever clinical human pregnancy being established by blastocyst transfer (Steptoe and Edwards, 1978Go). Blastocysts have been grown in vitro for many years, with rates approaching 50% of embryos cultured being achieved >10 years ago, although many failed to expand fully (Fishel and Edwards, 1983Go). Expanded and hatching blastocysts were also cryopreserved, ready for a later transfer (Cohen et al., 1985Go, Fehilly et al., 1985Go).

In some clinics, modern media containing various additives such as hyaluronate and vitamins still yield low rates of development of cleaving human embryos to expanding or hatched blastocysts. One detailed study (Jones et al., 1998Go) utilized a cell- and serum-free system of sequential culture media, brief exposures of oocytes to spermatozoa, culture of 2–3 embryos in one microdrop from days 1–3, then transfer to another medium from days 3–5. Gardner's (1994) G1 and G2 media with added albumin were used in this study, in various protocols, and blastocysts required various intervals in culture, i.e. 5.5, 6.5. and 7.5 days post-insemination. Overall 20% of embryos developed into enlarging blastocysts. Desai et al. (1997) reported 45% of supernumary embryos reached blastocysts, with an inner cell mass in 82% of them. Another recent optimistic study on blastocyst culture in which heparin-binding epidermal growth factor (Hb–EGF) was added to embryos cultured from day 2–7 post-insemination, has also been reported (Martin et al., 1998Go). Rates of blastocyst formation in high-grade cleaving embryos improved from 41 to 71% with 1 nM Hb–EGF and to 71% with 100 mM Hb–EGF. Hatching rates also improved from 46 to 71% and to 82%. The stimulatory effect of Hb–EGF was only apparent from the 8-cell stage. Results went in the opposite direction with low-grade cleaving embryos, with rates of blastulation and hatching declining from 36% in controls to 19% with 100 nM Hb–EGF. Blastocysts in the control and treated groups utilized similar amounts of glucose and pyruvate.

The proof of this pudding is in the proportion of blastocysts which implant after transfer. Implantation rates per blastocyst transferred in early studies did not exceed 30% (Cohen et al., 1985Go; Fehilly et al., 1985Go), comparing favourably with the highest implantation rates with embryos grown in routine media and transferred on days 2–3, rarely exceeding 25% (e.g. Damario et al., 1997Go). Early work with cryopreserved blastocysts reached implantation per blastocyst rates of 10% (Hartshorne et al., 1991), although their morphological markers were unhelpful in predicting implantation. More recently, an overall pregnancy rate of 23% after the transfer of cryopreserved blastocysts could be divided into rates of 39% with early-blastulating embryos versus 6.2% with those blastulating on day 7 (Shoukir et al., 1998Go). When this implantation rate is corrected for the number of embryos failing to reach blastocysts, i.e. 52.7%, the corrected implantation rate per fertilized egg is <20% in the optimal group. Very recently, in a study of fresh embryos, occasional morulae were added when insufficient blastocysts were available to enable transfer of 2–3 embryos. The overall rate of implantation was 22.6% per embryo (Jones et al., 1998Go). This value is greatly reduced if calculated on the original number of fertilized embryos. Likewise, the transfer of four zona-free expanding and hatching-stage blastocysts in a single patient resulted in one singleton fetus (Fong et al., 1997Go). In a study on culturing embryos for 5 days, then transferring blastocysts, implantation rates were 23% with the first treatment cycle, declining to 12% at cycle 4, although it seems that embryo numbers were made up with earlier-stage embryos if blastocysts were not available (Scholtes and Zeilmaker, 1998Go). It is notable and cautionary that in this study 384 out of 929 (41%) patients did not achieve a single blastocyst in vitro, and this was suggested by the authors to be related to the ovarian function of the patient. Numbers of fertilized embryos established per patient were not quoted, but must have been fairly high since patients were stimulated with gonadotrophin-releasing hormone (GnRH) agonist/follicle stimulating hormone (FSH)/human chorionic gonadotrophin (HCG) protocols.

A comparative report from Gardner et al. (1998a) on transfers on day 3 or day 5 indicated implantation rates per embryo of 21% on day 3 (using mostly 7–8 cell embryos after assisted hatching) versus 46% on day 5 (using blastocysts). The blastocyst transfer group was composed of only eight patients, and only 38% of the original fertilized embryos reached blastocyst in one medium, while 66% reached blastocyst in another. A correction of 33% for those embryos failing to develop into blastocysts reduces the implantation rates in the two groups to 31 and 16% respectively in relation to the original number of fertilized embryos. This work was followed up by a prospective trial comparing day 3 and day 5 transfers in 47 and 45 patients respectively (Gardner et al., 1998bGo). Although implantation rates per embryo increased from 30.1% in the day 3 group to 50.5% in the day 5 group (P < 0.01), the clinical pregnancy rates were similar at 66 and 71% respectively. Perhaps the improved implantation rates per embryo following culture to blastocyst can be explained purely in terms of selection of best embryos.

If achieved overall, implantation rates of ~50% per blastocyst transferred (Gardner et al., 1998aGo,bGo), even if many embryos die before day 5, would be a significant step forward, providing the means to reduce multiple birth rates, albeit perhaps in a selected patient group. It could also have scientific benefits in helping to clarify the processes of implantation. They have survived for 5 days in vitro which must be a token of quality. Why do 50% of them then fail to implant, when these are already the survivors from a much higher number of fertilized oocytes? Aneuploidy is a possible cause. Its frequency in human oocytes reportedly ranges from 25–50% of embryos (Boue et al., 1975Go; Plachot et al, 1988Go; Martini et al., 1997Go). Losses in vivo and non-receptive endometrium could be other factors. It seems that similar failure rates occur in vivo. The human fecundability parameter measures events between sexual intercourse at the correct stage of the cycle and the occurrence of implantation. It varies between 0.15 and 0.35 in a normal population, and has a suggested mean value of 0.25; this means that 25% of couples should become pregnant within a month of trying (Basso et al., 1997Go; Spira, 1998Go). In a recent study on the establishment of pregnancy in multiparous women after discontinuation of oral contraception, sperm quality and bleeding pattern and frequency were recorded in a population of 612 couples over six successive menstrual cycles. Pregnancy rates per cycle were consistently <20% in virtually all couples classified on various semen parameters. At the end of the trial, 256 pregnancies had been established after 1661 cycles (Bonde et al., 1998Go). Epidemiological data also suggest a low implantation efficiency in vivo, with 20% or fewer women conceiving per cycle even when couples are fertile (Edwards and Brody, 1995Go). These values compare with an average of 70% in captive baboons (Stevens, 1997Go), a species offering an exemplary model for humans since it has similar steroid profiles and presumably mono-ovulatory cycles (Henson, 1998Go). Success rates in vivo depend, of course, on several factors including fertilization rates and embryo quality as well as the chance of implantation.

Alternative and earlier approaches to selecting embryos with potentially high implantation rates depend on fundamental knowledge on the human embryo. We have recently suggested that mammals, including the human, display polarization in oocytes and cytoplasmic rotation at sperm entry, with closely co-ordinated cleavage planes partitioning polarized maternal transcripts specifically into individual blastomeres. This situation would be similar to that in a wide range of animals from amphibians to nematodes (Edwards and Beard, 1997Go). Evidence exists of an anterior–posterior axis passing diametrically from the second polar body (Gardner, 1996Go; 1997), and of an animal/vegetal polarization of leptin and STAT3 in human oocytes and embryos (Antczak and Van Blerkom, 1997Go), thus supporting our hypothesis. Payne et al (1997, 1998) recently published pictures and displayed a video showing superb examples of cytoplasmic rotation of the pronucleus and movement in human embryos. Searches must be made for distinct parameters of polarity, pronuclear rotation or the segregation of polarized proteins into blastomeres to identify the best-quality human oocytes and embryos. Some investigators have actually taken this approach, even if initially unaware that their findings might relate to polarity. Scott and Smith (1998) scored human pronucleate embryos for closely aligned pronuclei, polarity in nucleoli, heterogeneous cytoplasm, clear halos in cortex and around pronuclei, a dark central halo and an early first cleavage division. Points were awarded for these characteristics, varying betweeen 1 and 10. An embryo could gain a maximum of 15 points when pronucleate at 16–18 h post-insemination, and another 10 at 22–26 h post-insemination after an early first cleavage, giving a total maximum of 25 points (Scott and Smith, 1998Go). Embryos with scores >15 produced implantation rates per embryo of 28%, whereas those with <14 produced rates of 2%. Pregnancy rates reached 71% with 3.7 high-scoring embryos transferred (Scott and Smith, 1998Go). These authors could add more parameters such as pronuclear polarity to calculate their score, which was based predominantly on early cleavage and pronuclear morphology. More than a decade ago, similar studies showed how early-dividing human embryos at the first and second cleavage divisions have implantation rates of 30% per embryo, much higher than those cleaving later (Edwards et al., 1984Go). Benefits in early first cleavage have been shown again more recently, since these embryos had implantation rates per embryo of 26% as compared with 3.2% for later-cleaving embryos (Sakkas et al., 1998Go). This benefit seems to persist into blastocysts, since pregnancy rates were 39% with those cryopreserved after blastulation on day 5 or 6 as compared with 6.2% with those blastulating on day 7.

The question arises as to whether blastocyst culture and pronuclear scoring are equally effective in selecting viable embryos. Gardner's (1998b) day 3 embryo and day 5 blastocyst transfers, and Scott and Smith's (1998) results together with their analyses by high or low corrected embryo scores are compared in Table IGo. Implantation rates per embryo transferred are impressively high for blastocysts (55.4%), considerably higher than 37% for 3 day transfers, 28% for high scoring embryos and only 2% for low-scoring embryos (Scott and Smith, 1998Go). However if allowance is made for embryos lost in culture (46.5% for blastocyst development), the implantation rate per cultured embryo is 21%, just below that obtained in Scott and Smith's (1998) high-scoring embryos (i.e. 28%) but an improvement on their overall implantation rate (i.e. 15%). It is also notable that similar rates of pregnancy (71%) were established in the two best performing groups (blastocysts and high scoring pronucleate embryos), indicating that both selection methods target the same embryos. Surprisingly, blastocyst transfer yielded the highest multiple pregnancy rate. It can be concluded from Table IGo, that growth to blastocyst is the best selection for embryos capable of implantation since overall implantation rates per embryo when using blastocysts are higher than with pronuclear/2 cell transfers. This difference may arise through differences in centres, patient selection, or because the uterus is less favourable to cleaving embryos than to blastocysts.


View this table:
[in this window]
[in a new window]
 
Table I. Implantation rates per embryo and per cultured embryo in two recent studies on improving the growth of human embryos in vitro
 
It is essential now to combine these approaches, scoring pronucleate embryos and then finding out if these are the embryos which grow to blastocysts in vitro. This would permit extra attention to be given to the best embryos, and media might be devised for several stages of their growth. There is no point in devoting energy to embryos incapable of implanting. They would best be studied in vitro to find out where they go wrong. Early selection criteria up to and including the first cleavage division avoid long-term culture, and may have advantages beyond this. Transcription levels particularly in the male pronucleus (Ram and Schultz, 1993Go; Ao et al., 1994Go) suggest that events at the 1-cell stage are critical for later development. This new outlook implies that more information can be gleaned on embryo quality by correctly assessing the pronucleate/2-cell embryo rather than other preimplantation stages. Transfers of 1- or 2-cell stages are fully acceptable because the human uterus seems to differ to those of other species, in that it is an excellent incubator for such early preimplantation embryos (Bavister and Boatman, 1997Go). It is even possible to establish human pregnancies by transferring unfertilized eggs with zona-attached spermatozoa (Rizk et al., 1990Go). For certain purposes, it remains essential to grow embryos to blastocysts, e.g. for preimplantation diagnosis of inherited disease such as a reciprocal balanced translocation (Ménézo et al. 1997Go).

Yet, overall, the proportion of newly fertilized human embryos capable of implantation still remains <30%, whatever selection procedure is applied. This is the most disappointing statistic of human IVF and is only one of several observations which indicate a low degree of fertility in our species. All forms of IVF, including variations in ovarian stimulation and embryo culture, ultimately indicate a 20–25% rate of implantation per embryo in vivo and in vitro. All we are doing in IVF is selecting the best available embryos, and these are best identified at the blastocyst stage when all weaker embryos have died or arrested or by typing pronucleate eggs for polarity characteristics. We await with interest an explanation of why human semen samples contain so many abnormal spermatozoa, and why oocytes and embryos carry so many chromosomal anomalies and implant in very low numbers, while a woman has only 500 cycles or thereabouts in her whole lifetime and ovulates only a single egg in most of them! Do any other species display such an immense loss of reproductive potential? Perhaps cranial size and prolonged parent–child bonding have modified evolutionary pressures and restricted human reproduction to a few single births.

Notes

To whom correspondence should be addressed Back

References

Antczak, M. and Van Blerkom, J. (1997) Oocyte influences on early development: the regulatory proteins leptin and STAT3 are polarized in mouse and human oocytes and differentially distributed within the cells of the preimplantation stage embryo. Mol. Hum. Reprod., 3, 1067–1086.[Abstract]

Ao, A., Erikson, R.P., Winston, R.M.L. and Handyside, A.H. (1994) Transcription of paternal Y-linked genes in the human zygote as early as the pronucleate stage. Zygote, 2, 281–287[Medline]

Basso, O., Olsen, J., Bisanti, L. et al. (1997) Repeating episodes of low fecundability. A multicentre European study. Hum. Reprod., 12, 1448–1453.[Abstract]

Bavister, B.D. and Boatman, D.E. (1997) The neglected human blastocyst revisited. Hum. Reprod., 12, 1607–1610.[ISI][Medline]

Bonde, J.P.E., Ernst, E, Jensen, T.K. et al. (1998) Relationship between semen quality and fertility, a population based study of 430 first-pregnancy planners. Lancet, 352, 1172–1177.[ISI][Medline]

Bongso, A., Fong, C.Y., Ng, S.C. and Ratnam, S.S. (1994) Blastocyst transfer in human in vitro fertilization: the use of embryo culture. Cell Biol. Int., 18, 1181–1189.[ISI][Medline]

Boue, J., Boue, A. and Lazar, P. (1975) Retrospective and prospective epidemiological studies of 1500 karyotyped spontaneous human abortions. Teratology, 12, 11–26.[ISI][Medline]

Cohen, J., Elsner, C., Kort, H. et al. (1990) Impairment of the hatching process following IVF in the human and improvement of implantation by assisted hatching using micromanipulaion. Hum. Reprod., 5, 7–13.[ISI][Medline]

Cohen, J., Simons, R.F., Edwards, R.G. et al. (1985) Pregnancies following the frozen storage of expanding human blastocysts. J. In Vitro Fertil. Embryo Transfer, 2, 59–64.[Medline]

Damario, M.A., Barmat, L., Liu, H.-C. et al. (1997) Dual suppression with oral contraceptives and gonadotrophin releasing-hormone antagonists improves in-vitro fertilization outcome in high responder patients. Hum. Reprod., 12, 2359–2365.[Abstract]

Desai, N., Kinzer, D., Loeb, A. and Goldfarb, J. (1997) Use of synthetic serum substitute and {alpha}-minimum essential medium for the extended culture of human embryos to the blastocyst stage. Hum. Reprod., 12, 328–335.[Abstract]

Edwards, R.G and Beard, H.K. (1997) Oocyte polarity and cell determination in early mammalian embryos. Mol. Hum. Reprod., 3, 863–906.[Abstract]

Edwards, R.G. and Brody, S.A. (1995) Principles and Practice of Assisted Human Reproduction. W.B.Saunders Co, Philadelphia, USA.

Edwards, R.G., Fishel, S.B., Cohen, J., et al. (1984) Factors influencing the success of in vitro fertilization for alleviating human infertility. J. In Vitro Fertil. Embryo Transfer, 1, 3–23.[Medline]

Fehilly, C.B., Cohen, J., Simons, R.F., et al. (1985) Cryopreservation of cleaving embryos and expanded blastocysts in the human: a comparitive study. Fertil. Steril., 44, 638–644.[ISI][Medline]

Fishel, S. and Edwards, R.G. (1983) In vitro fertilization of human oocytes: Factors associated with embryonic development in vitro, replacement of embryos and pregnancy. In Beier, H.M. and Lindner, H.R. (eds), Fertilization of the Human Egg in vitro. Springer-Verlag, Berlin, Germany, pp 251–269.

Fong, C.Y., Bongso, A., Ng, S.C., et al. (1997) Ongoing pregnancy after transfer of zona-free blastocysts: implications for embryo transfer in the human. Hum. Reprod., 12, 557–560.[ISI][Medline]

Gardner, D.K. and Lane, M. (1998) Culture of viable human blastocysts in defined sequential serum-free media. Hum. Reprod., 13 (Suppl. 3), 148–159.[Medline]

Gardner, D.K., Lane, M., Spitzer, A. and Batt, P.A. (1994) Enhanced rates of cleavage and development of sheep zygotes cultured to the blastocyst stage in vitro in the absence of serum and somatic cells: amino acids, vitamins and culturing embryos in groups stimulate development. Biol. Reprod., 50, 390–400.[Abstract]

Gardner, D.K., Vella, P., Lane, M. et al., (1998a) Culture and transfer of human blastocysts increases implantation rates and reduces the need for multiple embryo transfers. Fertil. Steril., 69, 84–88.[ISI][Medline]

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

Gardner, R.L. (1996) Can developmentally significant spatial patterning of the egg be discounted in mammals? Hum Reprod. Update, 2, 3–27.[Abstract/Free Full Text]

Gardner, R.L. (1997) The early blastocyst is bilaterally symmetrical and its axis of symmetry is aligned with the animal-vegetal axis of the zygote in the mouse. Development, 124, 289–301.[Abstract/Free Full Text]

Harshorne, G.M., Elder, K., Crow, J. et al. (1991) The influence of in-vitro development upon post-thaw survival and implantation of cryopreserved human blastocysts. Hum. Reprod., 6, 136–141.[Abstract]

Henson, M.C. (1998) Pregnancy maintenance and the regulation of placental progesterone biosynthesis in the baboon. Hum. Reprod. Update, 4, 389–405.[Abstract/Free Full Text]

Hu, H., Maxson, W., Hoffman, D., et al. (1998) Co-culture with assisted hatching of human embryos using Buffalo rat liver cells. Hum. Reprod., 13, 165–168.[Abstract]

Jones, G.M., Trounson, A.O., Gardner, D.K. et al. (1998) Evolution of a culture protocol for successful blastocyst development and pregnancy. Hum. Reprod., 13, 169–177.[Abstract]

Martin, K.L., Barlow, D.H. and Sargent, I.L. (1998) Heparin-binding epidermal growth factor significantly improves human blastocyst development and hatching in serum-free medium. Hum. Reprod., 13, 1645–1652.[Abstract]

Martini, E., Flaherty, S.P., Swann, J., et al. (1997) Analysis of unfertilized oocytes subjected to intracytoplasmic sperm injection using two rounds of fluorescence in-situ hybridization and probes to five chromosomes. Hum. Reprod., 12, 2011–2018.[Abstract]

Ménézo, Y.J.R., Bellec, V., Zaroukian, A. and Benkhalifa, M. (1997) Embryo selection by IVF, co-culture and transfer at the blastocyst stage in case of translocation. Hum. Reprod., 12, 2802–2803.[Abstract]

Ménézo, Y.J.R, Hamamah, S., Hazout, A. and Dale, B. (1998) Time to switch from co-culture to sequential defined media for transfer at the blastocyst stage. Hum. Reprod., 13, 2043–2044.[Free Full Text]

Payne, D., Flaherty, S.P., Barry, M.F. and Matthews, C.D. (1997) Preliminary observations on polar body extrusion and pronuclear formation in human oocytes using time-lapse video cinematography. Hum. Reprod., 12, 532–541.[ISI][Medline]

Payne, D. (1998) Time-lapse video presented at 14th Annual ESHRE Meeting in Gotenberg.

Plachot, M., Veiga, A., Montagut, J., et al. (1988) Are clinical and biological IVF parameters correlated with chromosomal disorders in early life: a multicentric study. Hum. Reprod., 3, 627–635.[Abstract]

Ram, P.T. and Schultz, R.M. (1993) Reporter gene expression in G2 of the 1 cell mouse embryo. Dev. Biol., 156, 552–556.[ISI][Medline]

Rizk, B., Bekir, J.S., Avery, S. et al. (1990) Intrauterine replacement of oocytes with bound sperms: a pilot study. In the Proceedings of the II Joint ESCO–ESHRE Meeting, Milan, 1990. Abstract No. 18, p. 5.

Sakkas, D., Shoukir, Y., Chardonnens, D., et al. (1998) Early cleavage of human embryos to the two-cell stage after intracytoplasmic sperm injection as an indicator of embryo viability. Hum. Reprod., 13, 182.[Abstract]

Scholtes, M.C.W. and Zeilmaker, G.H. (1998) Blastocyst transfer in day-5 embryo transfer depends primarily on the number of oocytes retrieved and not on age. Fertil. Steril., 69, 78–83.[ISI][Medline]

Shoukir, Y., Chardonnens, D., Campana, A., et al. (1998) The rate of development and time of transfer play different roles in influencing the viability of human blastocysts. Hum. Reprod., 13, 676–681.[Abstract]

Spira, A. (1998) The use of fecundability in epidemiological surveys. Hum. Reprod., 13, 1753–1756.[ISI][Medline]

Scott, L.A. and Smith, S. (1998) The successful use of pronuclear embryo transfers the day following oocyte retrieval. Hum. Reprod., 13, 1003–1013[Abstract]

Steptoe, P.C. and Edwards, R.G. (1978) Birth after the reimplantation of a human embryo Lancet, ii, 366.

Stevens (1997) Some reproductive studies in the baboon. Hum. Reprod. Update, 3, 533–540.[Abstract/Free Full Text]

Tsirigotis, M. (1998) Blastocyst stage transfer: pitfalls and benefits. Hum. Reprod., 13, 3285–3289.[Free Full Text]