Morphological Scoring of Human Embryos
and its Relevance to Blastocyst Transfer
Michael J Tucker PhD FIBiol HCLD
Georgia Reproductive Specialists, Atlanta, GA 30342
The human blastocyst
Until the advent of the use of in vitro fertilization and embryo transfer (IVF-ET) for the treatment of infertility, information about the human blastocyst was virtually non-existent except for the odd serendipitous report of embryos recovered following hysterectomy. Consequently, much more was known about other mammalian blastocysts in terms of their morphology and rate of development, such as the mouse. It has been generally assumed though never fully established that in vivo produced human embryo develop more rapidly than those produced in vitro (Buster et al, 1985), leading to the presumption that extended culture of the human embryo is in some way detrimental. This led in turn to the common practice of transferring embryos on day-two or three of development in IVF-ET therapy. More recently confidence has grown with the development of stage specific sequential media to enable what seem to be consistent rates of blastocyst formation with in vitro-produced human embryos for use in IVF-ET. This has prompted some clinics to move all embryo transfers to the blastocyst stage (Marek et al, 1999), although some centers still report no major benefits to be gained with extended culture (Huisman et al, 2000).
Regardless of the relative environmental merits of in vivo or vitro development, it is understood that approximately 120 hours (day-five) into development the healthy human embryo should be at the blastocyst stage. At this time it should be comprised of some 80 to 160 cells, of which about 30% make up the inner cell mass (ICM), the remainder making up the trophectoderm (TE) (see Figure 1). Loss of the zona pellucida (ZP) occurs by about the sixth day, and thus the blastocyst, if viable, is ready to undergo implantation into the uterine mucosa. Any assessment of blastocyst quality, therefore must accommodate a measure of the three chief elements of the embryo at this stage: ICM, TE and the ZP.
Morphological grading of the blastocyst
The chief virtue of extended in vitro culture of the human embryo is the improved discrimination of potential viability based on morphology, when compared to earlier morphological scoring (see Appendix 2, and Scott et al. 2000). This is very clear when making the change from routine day-three transfers to the blastocyst stage (Graham et al., 2000). Here some significant uncoupling of which embryos are graded as good quality on day-three occurs in terms of their ultimate blastocyst formation and quality on day-five/six, when embryos are tracked individually. If any improvements are seen in terms of pregnancy rates with a shift to blastocyst transfer, it is solely through improved selection of embryos, not through intrinsic improvement of embryonic quality through extended culture. At this time the main drive to move to blastocyst transfer is to reduce the number of embryos transferred routinely, in an attempt to reduce multiple implantation after IVF-ET therapy. It is the case that a blastocyst's quality is very much determined by the quality of the gametes from which it came (Janny & Menezo, 1994; 1996), tempered by the in vitro environment in which it was cultured. It is doubtful that we can currently actually improve embryo quality over that which might be provided in vivo, consequently the only potential way for embryo quality to go is down! Thus, we as embryologists have a very real responsibility to "do no harm" while nurturing the preimplantation human embryo; conversely extended culture is a real privilege that allows us access to study the complete preimplantation period of development in the human.
That said, where do we stand with morphological assessment of blastocysts as a measure of quality? We are very limited with respect to non-invasive tests of potential embryo viability. Cell biopsy is possible, but remains a demanding though exciting area of embryo viability assessment, which is distinctly invasive. Morphological observation needs to be made sufficiently objective to be useful for inter-center comparison, and to this end several systems have been developed to grade blastocyst morphology. The simple description of the morphology would be rather "one-dimensional" without factoring in rate of development. So for grading to be effective it must take this into consideration, as the passage of time will clearly have a profound impact on cell number relative to the number of cell cycles that the embryo has had time to undergo. Dokras et al (1993) proposed a fairly simple approach to blastocyst grading which incorporated both morphology and rate of development. Its simplicity to some extent, however, compromises its usefulness, as it does not allow sufficient flexibility of description. Gardner & Lane (2000), on the other hand, have suggested an approach to grading that is more comprehensive, plus it incorporates assessment of rate of development and independent comment on the ICM and TE. Not just to be difficult, but to accommodate a tradition of embryo quality scoring that has graded the best embryos with a low number, the IVF-ET programs at Shady Grove Reproductive Science Center, Rockville, and at Georgia Reproductive Specialists, Atlanta operate with a grading system presented in Appendix 1. Ultimately though, by far the best morphological judgements are based on photographic images. While numeric and alphabetic scores can be ascribed, a permanent record of the actual image of a blastocyst linked with a photo-library for embryo grading leaves little to the imagination (see, for example, Menezo et al., 1999).
Value of morphological analysis
If morphological assessment is of value, how well does it perform as a means to judge embryonic viability on day-three, the traditional cleavage stage transfer day? By determining how the quality of blastocysts formed on day-five/six during extended culture is related to their morphology as previously assessed on day-three, it is possible to use blastocyst formation as a measure of potential viability. The answer is that day-three morphology can only predict at an approximately 50% rate those embryos that will eventually form blastocysts suitable for use on day-five/six (Rijnders & Jansen, 1998; Graham et al., 2000). In this second study, where blastocyst utilization was considered for both fresh transfer and cryopreservation, a retrospective analysis of 101 IVF cycles was undertaken. Embryos for potential clinical use were individually cultured in sequential media for an extended period in vitro, to enable use of blastocysts for fresh transfer or cryopreservation. Comparison of embryo quality for traditional Embryo Transfer (ET) or cryopreservation on day-three was made, with actual ET and freezing on day-five/six. From 1263 cleaving embryos, 559 were judged to have been of a standard suitable for use on day-three: 355 for ET (average per ET 3.5) and 204 would have been frozen (equivalent to 44% utilization). In actuality, 471 blastocysts were ultimately used on day-five/six, with 234 being transferred (average per ET 2.3), and 237 being frozen (equivalent to 37% utilization). Moreover, only 48% (256/1263) that would have been chosen for both ET and/or cryopreservation on day-three, were eventually used in a comparable fashion as blastocysts. In this case then, extended culture of human embryos seemed to increase discrimination of potential embryonic viability, with morphological criteria for embryo selection on day-three appearing inadequate. It is true that in the U.S. day-three transfers have usually been undertaken with three or more embryos to achieve what are considered good pregnancy rates, but the price that has been paid has been the high incidence of multiple implantation. This in itself is a concession to the poor abilities to judge consistently embryo quality on day-three. It is hoped that we come closer to consistency by selection at the blastocyst stage, whenever embryo quality and/or number allow.
Observation of "surplus" donated embryos and the
relevance of innate blastocyst hatching potential
By following the individual fate of each and every embryo that does not get used during IVF-ET therapy, it may be possible to glean some useful information with reference to the influence of certain general factors on blastocyst formation, its timing and also blastocyst hatching potential. In this regard, we observed growth of embryos for up to nine days, and saw a distinct influence of maternal age on blastocyst formation. This concurs with the trend toward decreased formation seen by Janny & Menezo (1996). Blastocyst hatching ability also seemed to decrease with increased maternal age. These trends were repeated when the basal maternal FSH levels were correlated with both blastocyst formation and hatching. What is interesting is the correlation of formation and hatching with the origin of the embryos relative to whether the embryos were generated by conventional insemination or ICSI. Average formation rates were similar, which seems contrary to Janny & Menezo (1994), although not all ICSI cases were straightforward male factor cases. However intriguingly, the hatching potential was markedly skewed, such that "spare" ICSI-produced embryos hatched much less frequently than their conventional counterparts (p<0.0001). This is surprising as it runs counter to the study of Tucker et al. (1996), where day-three assisted hatching did not seem to improve the implantation potential of ICSI embryos as the above might suggest. This may be for a variety of reasons including: insufficient numbers in the original randomized study; a beneficial impact of the uterus on hatching ability of ICSI embryos when transferred on day-three versus at the blastocyst stage. Whatever the reason, these recent data, if confirmed, prompt further study of the use of assisted hatching at the blastocyst stage in certain patient groups.
Simple observation of blastocyst formation or hatching does not provide the full picture that may be drawn from study of "spare embryos", because the extra dimension of rate of development may improve understanding of the impact of embryo origin on outcome. In progress in our labs is the correlation of day of blastocyst formation with ultimate potential to hatch. In an attempt to ascertain why assisted hatching might be beneficial for "slow" embryos, that do not form as blastocysts until day-six of development (Tucker, 1999), observations on rate of both formation and hatching might give some answers. It might be hypothesized, for example, that "spare embryos", that form blastocysts by day-five, may have a higher rate of hatching in vitro when compared to those that may take six days to form. Likewise this trend might be increasingly extreme the longer a blastocyst takes to form, such that day-seven blastocysts may very rarely hatch free from their ZPs. This could be due to innate inadequacies of such late-developing embryos, but might warrant some investigation of whether such embryos could be "rescued" to attempt pregnancy through ZP removal.
To try to summarize what might be considered a few comments on the "real world" use of blastocysts in IVF-ET and embryo morphology in general, the following points are worth pondering:
- Not all blastocysts are "equal" - it is very much true that blastocyst formation rate (say, 45-55%) is not equatable with usable blastocyst rate (closer to 35-40%).
- The ICM/TE cell count and integrity need to be sufficient to indicate possible embryonic viability before use should be considered.
- There is a strong correlation between the rate of embryonic development and potential viability; but slower development is not totally incompatible with viable pregnancy, especially if assisted hatching is employed.
- Earlier stage assessment of embryonic morphology at the cleavage stage or even the pronucleate (1-cell stage) have their place, but can never be superior to later stage assessment assuming culture conditions are optimal.
- Good quality blastocysts come from good quality eggs (and spermatozoa). Extended culture will not improve embryo quality, but merely reveal it!
Buster JE, Bustillo M, Rodi IA (1985): Biologic and morphologic development of donated human ova recovered by non surgical lavage. Am J Obstet Gynecol 153:211-7.
Dokras A, Sargent I, Barlow DI (1993): Human blastocyst grading: an indicator of developmental potential? Hum Reprod 8:2119-27.
Gardner DK & Lane M (2000): Embryo culture systems. In: Handbook of In Vitro Fertilization. 2nd Edition, Editors: Trounson AO & Gardner DK. Chapter 11, pp205-64; CRC Press, Boca Raton, FL, U.S.A.
Graham J, Han T, Porter R, Levy M, Stillman R, Tucker M (2000): Day-three morphology is a poor predictor of blastocyst quality in extended culture. Fertil Steril: in press.
Hardy K (1993): Development of human blastocysts in vitro. Serono Symposium on Preimplantation Embryo Development, Chapter 14; pp184-99, Newton, MA, USA; August 15-18, 1991.
Huisman GJ, Fauser BCJM, Eijkemans MJC, Pieters MHEC (2000): Implantation rates after in vitro fertilization and transfer of a maximum of two embryos that have undergone three to five days of culture. Fertil Steril 73:117-22.
Janny L & Menezo YJR (1994): Evidence for a strong paternal effect on human preimplantation embryo development and blastocyst formation. Mol Reprod Dev 38:36-42.
Janny L & Menezo YJR (1996): Maternal age effect on early human embryonic development and blastocyst formation. Mol Reprod Dev 45:31-7.
Marek D, Langley M, Gardner DK, Confer N, Doody KM, Doody KJ (1999): Introduction of blastocyst culture and transfer for all patients in an in vitro fertilization program. Fertil Steril 72:1035-40.
Menezo YJR, Kauffman R, Veiga A, Servy E (1999): A mini-atlas of the human blastocyst in vitro. Zygote 7:61-5.
Scott LA, Alvero R, Leondires M, Miller B (2000): Pronuclear morphology. Hum Reprod 15:1- 10.
Tucker MJ, Morton PC, Wright G, Ingargiola PE, Sweitzer CL, Elsner CW, Mitchell-Leef, DE, Massey JB (1996): Enhancement of ICSI outcome: do co-culture or assisted hatching improve implantation rates? Hum Reprod 11:2434-7.
Tucker M (1999): Relevance of assisted hatching with blastocyst transfer. The 1st Congress on Controversies in Obstet, Gynecol & Infertility, Prague, Czech Rep., Oct. 28-31, 1999. Monduzzi, Bologna, Italy, 1999.
BLASTOCYST SCORING - Modification for Shady Grove Fertility Center, Rockville, and Georgia Reproductive Specialists.
- Give each blastocyst an overall numerical score based on rate of development and degree of expansion; i.e.,
||Fully expanded or hatching day-five
||Fully expanded or hatching day-six
Moderate expansion day-five
||Moderate expansion day-six
Early cavitation day-five
||Early cavitation day-six
Morula day-five (?six)
- Add to this two alphabetic scores to grade i.e.,
1st). Inner Cell Mass
||high cell number with good cell/cell adhesion
||lower cell number with poorer cell/cell attachment
||no cells apparent (ICM) sparse granular (degenerate?) cells (trophectoderm)
For example, a good quality well-expanded blastocyst on day-six with good cell count in the ICM (>12 to 15), and good integrity of the trophectoderm would be scored as 2AA
A second example: a fully-expanded blastocyst on day-five with nice trophectoderm but non-existent ICM would be scored as 1CA
DAY-THREE CLEAVAGE STAGE SCORING.
- Quantification of embryo quality requires assessment of three major parameters:
||Rate of development
- Additionally, how even the cell sizes are, and the presence of multinucleate cells (blastomeres) need to be noted on individual embryos. However, for the sake of this scoring scale the above three parameters will be used routinely with the cell size and nucleation being noted where relevant to readjust the score - usually bringing it down one grade.
|Best Prognosis Embryo
||6-cell or greater;
||6-cell or greater;
fragmentation 10 - 25%
||i. 4 -5-cell; fragmentation <10% ii. 6-cell or greater;
||4 - 5-cell;
fragmentation 10 - 45%
|Worst Prognosis Embryo
ii. arrested 4-cell (day-2 to day-3)
iii. any embryo with fragmentation >45%
Pronucleate one-celled embryo the day after egg collection - it is normally fertilized with two pronuclei that are unbalanced in their size and nucleolar content (see Scott et al. 2000).
Grade 3 day-three four-cell embryo with <10% fragmention. (note: this would be an excellent embryo on day-two of development)
Grade 1 day-three eight-cell embryo with <10% fragmentation.
Two expanded blastocysts on day-six of development:
||hatching healthy blastocyst with good morphology
||unhatched blastocyst with poor morphology