The effect of late-follicular phase progesterone elevation on embryo ploidy and cumulative live birth rates

Research question: Does late-follicular phase progesterone elevation have a deleterious effect on embryo euploidy, blastocyst formation rate and cumulative live birth rates (CLBR)? Design: A multicentre retrospective cross-sectional study including infertile patients aged 18-40 years who underwent ovarian stimulation in a gonadotrophin-releasing hormone antagonist protocol and preimplantation genetic testing for aneuploidies (PGT-A) followed by a freeze-all strategy and euploid embryo transfer between August 2017 and December 2019. The sample was stratified according to the progesterone concentrations on the day of trigger: normal (1.50 ng/ml). Moreover, sensitivity analyses were performed to determine whether different conclusions would have been drawn if different cut-offs had been adopted. The primary outcome was the embryo euploidy rate. Secondary outcomes were the blastocyst formation rate, the number of euploid blastocysts and CLBR. Results: Overall 1495 intracytoplasmic sperm injection PGT-A cycles were analysed. Late-follicular phase progesterone elevation was associated with significantly higher late-follicular oestradiol concentrations (2847.56 +/- 1091.10 versus 2240.94 +/- 996.37 pg/ml, P < 0.001) and significantly more oocytes retrieved (17.67 +/- 8.86 versus 12.70 +/- 7.00, P < 0.001). The number of euploid embryos was significantly higher in the progesterone elevation group (2.32 +/- 1.74 versus 1.86 +/- 1.42, P = 0.001), whereas the blastocyst formation rate (47.1% [43.7-50.5%] versus 51.0% [49.7-52.4%]), the embryo euploidy rate (48.3% [44.9-51.7%] versus 49.1% [47.7-50.6%], the live birth rate in the first frozen embryo transfer (34.1% versus 31.1%, P = 0.427) and CLBR (38.9% versus 37.0%, P = 0.637) were not significantly different between the two groups. Conclusions: Euploidy rate and CLBR do not significantly differ among PGT-A cycles with and without late-follicular progesterone elevation in a freeze-all approach