Is There Really a Value in Embryo Selection?

Since IVF cycles are on purpose hyper-stimulated with gonadotropins in most cases, they usually produce multiple embryos. To select among those the best for embryo transfer into the uterus has been a goal of IVF laboratories since the beginning of IVF. This process is called embryo selection.

Embryo_QualityPractically all embryology laboratories select embryos by morphological criteria at cleavage stage (day-3 after fertilization), which used to be the day of embryo transfer in most IVF centers. A more functional embryo selection evolved, however, with the increasing utilization of longer embryo culture to blastocyst stage (days-5/6 after fertilization), with the claim being that longer embryo culture selects out the “best” embryos with the highest pregnancy potential.

Blastocyst-stage embryo culture has become increasingly popular in the last decade. Indeed, increasing numbers of IVF centers have converted their programs completely from cleavage- to blastocyst-stage transfers. This development was further enhanced based on the secondary embryo selection processes that were introduced in recent years, which are co-dependent on blastocyst-stage embryo transfer, such as preimplantation genetic screening (PGS) and closed incubation systems with time-lapse imaging (CIS/TLI).

The concept of embryo selection is largely futile, and for selected patient populations may, indeed, be harmful by reducing their pregnancy and live birth chances in association with IVF.

Our opinion is based on the physiology of human reproduction as well as clinical realities associated with embryo selection. In considering the physiology of human reproduction, it is widely accepted that embryo quality is primarily dependent on egg quality. The ultimately retrieved egg (oocyte) is, of course, the result of months of maturation within a follicle environment that is influenced by a multitude of factors. While individual follicles may mature differently, it is generally accepted that within one follicle cohort, though oocyte maturity may vary, environmental conditions are likely rather similar.

It, therefore, appears unlikely to us that individual follicles, which for months have been maturing together within a given ovarian environment, at the very end of this long cohabitation process may still be individually influenceable to significant degrees by our current clinical interventions. In other words, it appears much more likely that by the time we intervene in the follicle maturation process in the so-called gonadotropin-sensitive period of folliculogenesis (the last 2 weeks of follicle maturation), the ultimate fate of follicle and egg is already largely determined. Attempts at influencing egg quality (and, embryo quality), consequently, would have to take place at much earlier stages of follicle and egg development.

This conclusion is further supported by a recent study from our center, which demonstrated that in good prognosis patients, who on day-3 after fertilization have a large number of good quality embryos, selection of embryos by oocyte quality parameters actually better correlated with pregnancy outcome than selection by standard morphological embryo parameters (Lazzaroni-Tealdi et al., PLoS ONE 2015;10:e0143632).

The strongest arguments against effectiveness of current methods of embryo selection, however, are based on clinical observations. They mostly center on false outcome claims for blastocyst-stage embryo culture and, therefore, also apply to co-dependent secondary embryo selection methods, like PGS and CIS/TLI.

The principal criticism stems from the fact that alleged outcome advantages of blastocyst-stage embryo transfer have been misleading. Most IVF centers that switched their practice from cleavage- to blastocyst-stage embryo transfer did so under the assumption that this would improve the programs clinical pregnancy and live birth rates. This is, however, an incorrect assumption!

As two meta-analyses in the COCHRANE library demonstrate (Cochrane Database Syst Rev 2007;17:CD002118 and 2012;7:CD002118), if one looks at the cumulative pregnancy rate achieved from a single IVF cycle follicle/oocyte cohort, cleavage-stage embryo transfer will always achieve significantly more clinical pregnancies and live births than blastocyst-stage embryos in all patient populations. The reason is simple: Some embryos, which do not survive longer embryo culture to blastocyst-stage, if transferred on day-3, will still result in completely normal live births. Extended embryo culture to blastocyst stage, therefore, results in loss of embryos with normal pregnancy potential.

Patients who are most negatively affected by blastocyst-stage embryo transfers are women with relatively few embryos (i.e., poor-prognosis patients like older women or women with low functional ovarian reserve at younger ages) because they may end up with simply no transferable embryos. Good-prognosis patients, who usually produce larger numbers of embryos, even after losing some, will still have enough for transfer and, indeed, may have selected out their “best” embryos by culturing to blastocyst stage. The two meta-analyses, therefore, suggest that they marginally improve their immediate IVF outcomes.

Good-prognosis patients (representing ca. 20% of women on average), thus, marginally benefit from blastocyst stage transfers, though still lose cumulative pregnancy chances; average-prognosis patients (approximately 60%) experience no benefit; and poor-prognosis patients (also ca. 20%), indeed, have outcome chances from IVF reduced with long-term culture. One, therefore, really has to wonder whether blastocyst-stage culture is worth the effort.

How come this information is not widely known?

The answer is simple: Proponents of blastocyst-stage transfer report their outcomes with reference point embryo transfer. Consequently, patients who do not reach embryo transfer are never included in outcome reports. Even though outcomes reported this way, therefore, only include relatively good-prognosis patients, their outcomes are, however, erroneously applied to all IVF patients, a statistical error, unfortunately committed only too often in recent years in the introduction of new procedures to IVF, including PGS and CIS/TLI.

This is a part of the January 2016 issue of the CHR VOICE. Read this issue in PDF.