[Preimplantation assessment: experimental data on animals and humans].

Assessing and/or improving the implantation prognostic remain a major goal for research studies as well as for teams doing embryo transfer in many species. Criteria for such a goal were focussed ten years ago and combined theoretically: -sensitivity for applying to one embryo; -clear cut-off for individual decision; -fastness to be suitable for embryo transfer in due time; -no toxicity or invasiveness for embryo; -finally some simple technical approach in order to be applies by a large number of teams. Moreover whatever may be qualitative or quantitative criteria, they should be relied to the final result as alive newborn. The more ancient way to appreciate embryo quality deal with the simple observation of morphological and kinetic criteria about embryo, but such non invasive approach was obviously limited, in spite of the positive influence of regular blastomers, absence of fragmentation and synchronization with time of transfer on implantation rate. The major transcriptional activity of human embryo developing between 6 and 8 cells stage, of course, were unassessed by transfer to day 2. Moreover the apparent quality of the embryo better reflected oocyte quality than embryo quality. Coculture development encompassed only partially such limitation. Using fluorescent probes, it was possible to evaluate some metabolic activity as well as membrane integrity, but such criteria revealed to be both invasive and uneasily reliable with developmental ability of the embryo. Methods dealing with glucidic, protidic or lipidic metabolisms are developed elsewhere, but revealed uneasy to apply, due both to their invasiveness or technical difficulties and their large inter-individual variability. Some hope has raised by the finding of growth factors or cytokines which are expressed by the embryo and/or embryotrophic but a lot of works remain to be down before an easy practical application.