P-633 Progesterone elevation on the day of trigger in the presence of low ovarian response signifies abnormal follicular environment and poorer oocyte health

Abstract

Abstract Study question Is progesterone elevation (PE) in the presence of low number of follicles (unexpected PE) associated with oocyte or embryo quality? Summary answer PE on day of trigger in cycles with ≤4 follicles is associated with significantly reduced fertilisation rates and fewer top-quality blastocysts What is known already The negative effect of PE on pregnancy rates in fresh IVF cycles has been demonstrated by multiple studies and meta-analyses, and most studies support no negative effect on oocyte and embryo quality. However, as it is thought that PE is the product of cumulative secretion of low levels of progesterone from multiple developing follicles, PE in the presence of a low number of follicles (unexpected PE) is considered a different pathophysiological entity. It has been hypothesized, but never proven, that in these cases, PE is the result of follicular dysregulation and that it might indicate poorer follicular and oocyte health. Study design, size, duration This is a multi-centre retrospective analysis of a large cohort (n = 1,393) of IVF/ICSI cycles with a low response to ovarian stimulation (1-4 follicles ≥14mm) performed from January 2018 to April 2022. Only cycles of ovarian stimulation using gonadotrophins and GnRH analogues with a measurement of serum progesterone on the day of triggering final oocyte maturation were included. Natural cycles and cycles involving the administration of clomiphene were excluded. Participants/materials, setting, methods Unexpected PE was defined as serum progesterone levels greater than 4.77 nmol/L (1.5 ng/mL) on day of trigger. The primary outcome was the number of usable blastocysts. Secondary outcomes included the number of oocytes retrieved, two pronuclei (2PN) oocytes, number of top-quality embryos, fertilisation rate and blastocyst formation rate. Generalised estimating equations were used to account for the clustered nature of data. Multivariable regression analysis was performed to control for the effect of multiple confounders. Main results and the role of chance In the univariate regression analysis, the number of oocytes retrieved was not significantly different between cycles with and without PE (mean 2.3 vs. 2.1 respectively; p = 0.550). The number of 2PN oocytes was significantly (p = 0.021) lower in the PE group (mean 0.72, 95%CI 0.31-1.14) compared to the non-PE group (mean 1.21, 95%CI 1.12-1.31). The number of top-quality blastocysts was also significantly (p = 0.003) lower in the PE group (mean 0.03, 95%CI -0.03-0.10) compared to the non-PE group (mean 0.14, 95%CI 0.11-0.17). Fertilisation rates per oocyte retrieved, fertilisation rates per oocyte inseminated and the top-quality blastocyst rate per oocyte retrieved were consistently lower in the PE group at 33.3% (95% CI 19.0-47.6), 26.9% (95% CI 14.8-39.0) and 0.7% (95% CI -0.8-2.1) compared to 58.7% (95% CI 55.8-61.7), 52.3% (95% CI 49.3-55.2) and 5.5% (95% CI 4.3-6.8) in the non-PE group. The negative association of PE with the primary outcome measure of this study was confirmed in the multivariable regression analysis where the number of usable blastocysts was significantly lower in the PE group (adjusted mean 0.14, 95%CI 0.03-0.25) compared to the non-PE group (adjusted mean 0.27, 95%CI 0.23-0.31) after adjusting for confounders. Limitations, reasons for caution This is a retrospective observational study, thus not all confounders may be accounted for. Despite being one of the largest studies conducted in poor responders, the sample size of unexpected PE cycles is relatively small due to its rarity. Therefore, this study may be underpowered to detect smaller effect sizes. Wider implications of the findings This study demonstrates for the first time that unexpected PE on the day of trigger (i.e. in cycles with low ovarian response) is associated with poorer oocyte health as reflected in the reduced fertilisation rates and fewer blastocysts available for transfer. This information can aid in better understanding this phenomenon. Trial registration number Not applicable