P-253 Shorter duration of compaction during human in-vitro preimplantation embryo development is associated with a higher clinical pregnancy rate

Abstract Study question Is the duration of compaction, KID- (Known Implantation Data) and iDAScore (Intelligent Data Analysis for embryo evaluation) associated to clinical pregnancy rate? Summary answer Duration of compaction, KID- or iDAScore can be used to select the best embryo for transfer to increase clinical pregnancy rate. What is known already The development of human preimplantation embryos follows a programmed timeline in which a series of critical events occur. Compaction is a typical event at 3 to 4 days post fertilisation that is characterised by flattening of the blastomeres and the formation of tight junctions between the blastomeres. It is considered the first morphological event in the differentiation process of the embryo. Time-lapse technology introduced continuous monitoring of the embryo throughout development in the IVF laboratory. Evaluation of the developmental events combined with calculating KID- and iDAScore can optimise the selection of the most competent embryos for transfer and/or cryopreservation. Study design, size, duration This single-centre retrospective observational study included 158 IVF/ICSI cycles with fresh single embryo transfer (SET) was performed between December 2018 and November 2021. Embryos were cultured during 5 days in cleavage/blastocyst medium (Coopersurgical) in the EmbryoScope + (Vitrolife). Transferred embryos were evaluated for start of compaction, time to full compaction and duration of compaction. Embryo quality was calculated using KID- and iDAScore. These parameters were compared between the clinically pregnant and non-pregnant group (primary outcome). Participants/materials, setting, methods Only IVF/ICSI cycles with ejaculated sperm and fresh SET on day 5 were included. MNC, IVM and PGT cycles were excluded. Time zero was the start of injection or insemination. Pregnancy was confirmed by hCG and clinical pregnancy was defined by gestational sac visualisation at ultrasound. GraphPad Prism and R-studio were used for statistical analysis. For prediction of clinical pregnancy, univariate logistic regression was used. Other significant differences were determined using t-test. Main results and the role of chance Out of 158 fresh ET, 101 (63.9%) had a positive hCG, of which 88 (55.6%) achieved clinical pregnancy. All 158 transferred blastocysts were annotated to calculate KID- and iDAScore. There was no statistical difference in age between the two groups (34.7 years vs 35.0 years; p = 0.69). Start of compaction was heterogeneous (between 50.9 and 98.3 hours post injection/insemination; mean=76.5±7.7), as well as the blastomere number at its initiation (between 4 and 16 blastomeres; mean=11.8±2.1). Univariate logistic regression showed that each individual parameter, i.e. duration of compaction (p = 0.02), KID-score (p = 0.001) and iDAScore (p = 0.0006) was different between the clinically pregnant and non-pregnant group. The total duration of compaction was significantly shorter in the clinical pregnant group (mean=8.6±3.4 hours vs 10.2±4.7 hours; p = 0.01; t-test). In the pregnant group the KIDscore (mean=7.7±1.4 vs 6.7±2.3; p = 0.0007) and iDAScore (mean=8.9±0.7 vs 8.3±1.3; p = 0.0002) were significantly higher. During partial compaction, cells were rather excluded (93%) than extruded from the process; 17 embryos underwent this process, 10 of which resulted in a clinical pregnancy. Limitations, reasons for caution As this is a retrospective study, the influence of uncontrolled variables cannot be excluded. In the future, different models will be applied that can combine duration of compaction, KID- and iDAScore in a larger study. Wider implications of the findings Our analysis confirms previous findings that KID- and iDAScore are good predictors of clinical pregnancy. We also show that duration of compaction can be used as a potential predictor for pregnancy, especially in IVF clinics that have no access to KID- or iDAScore. Trial registration number not applicable