Why the hypothesis of embryo selection in IVF/ICSI must finally be reconsidered.

Embryo selection (ES) during IVF is expected to select the 'best' embryo(s) from among a cycle's embryo cohort and has been a core concept of IVF for over 40 years. However, among 36 492 articles on ES in a recent PubMed search, we were unable to locate even a single one questioning the concept that, beyond standard oocyte and embryo morphology, ES has remained an unproven hypothesis. In unselected patient populations, attempts at ES have universally, indeed, failed to improve cumulative pregnancy and live birth rates. The only benefit ES appears to offer is a marginal shortening in time to pregnancy, and even this benefit manifests only in best-prognosis patients with large oocyte and embryo numbers. Excluding in vitro maturation efforts, oocytes, once retrieved, and their resulting embryos have predetermined finite cumulative pregnancy and live birth chances that cannot be further improved. The hypothesis of ES has, however, remained a driving force for research and the introduction of a multitude of 'add-ons' to IVF. Enormous investments over decades in ES, therefore, should be better redirected from post- to pre-retrieval efforts.