P-176 Morphokinetic Expressions Predict Day Five Embryo Quality and Implantation - A Prospective Observational Study

Abstract

Abstract Study question The aim of this prospective study was to investigate the value of human embryo morphokinetics as predictors of day 5 embryo quality and implantation. Summary answer Day 5 embryo quality was predicted best by Cleavage Synchronicity 2-8-cell stage (CS2-8). Implantation was predicted best by the time to the expanded blastocyst (tEB-tPNf). What is known already Time-lapse imaging technology enables embryologists to observe and assess not only embryo morphology but also kinetics and cleavage synchronicity. It was found that time intervals like time from 5- to 8-cell stage (s3 = t8 - t5) can predict blastocyst formation. Cetinkaya et al. defined four relative morphokinetic expressions describing cleavage synchronicity: Cleavage Synchronicity 2-8-, 4-8-, 2-4-cell stage (CS2-8, CS4-8, CS2-4) and DNA Replication time ratio (DR). In a retrospective study they found that CS2-8 followed by s3, and CS4-8 were better predictors of blastocyst quality than absolute cleavage time points (https://doi.org/10.1007/s10815-014-0341-x). Study design, size, duration This prospective observational study was conducted between December 2020 and October 2021 at a single university hospital. Patients between 18 and 45 years who underwent stimulated IVF or ICSI treatment with embryo culture in a time-lapse incubator were included. Exclusion criteria were endometriosis, PCOS, history of chemotherapy, PGT cycles and end of culture before day 5. Every patient was included only once (n = 100). Implantation was assessed for all patients with embryo transfer. Participants/materials, setting, methods Embryos (n = 298) reaching at least 8-cell stage were annotated for morphokinetics and divided into four groups based on day 5 morphology: top (TQ; n = 60), good (GQ; n = 69), poor quality (PQ; n = 105) or not reaching blastocyst stage (AE; n = 64). Absolute times were normalized to fading of pronuclei (tPNf). ROC-analysis was performed to test how morphokinetics predict quality and implantation (only known-implantation-data; n = 110). Main results and the role of chance To predict day 5 embryo quality, four morphokinetic expressions had an Area Under the Curve (AUC) of more than 0.7, including CS2-8 (AUC = 0.732), CS4-8 (AUC = 0.717), s3 (AUC = 0.724) and the start of blastulation (tSB-tPNf) (AUC = 0.725). All morphokinetic expressions mentioned were significantly different between TQ plus GQ versus PQ plus AE embryos (p < 0.001). Additionally, the data of each morphokinetic parameter were divided into four quarters using the 25th, 50th and 75th percentile. CS2-8 had a rate of 67.6% TQ plus GQ blastocysts in the fourth quarter (CS2-8 > 0.86). CS4-8, s3 and tSB-tPNf had the highest rate of TQ plus GQ blastocysts (64.3 %, 64.3 %, 72.9 %) in the first quarter (CS4-8≤0.20, s3≤3.01 h, tSB-tPNf ≤ 67.06 h). Implantation was predicted best by the time to reach the morula stage (tM-tPNf; AUC = 0.636, p = 0.049), start blastulation (tSB-tPNf; AUC = 0.660, p = 0.022) and expanded blastocyst (tEB-tPNf; AUC = 0.688, p = 0.009). The relative expressions CS2-8 and CS4-8 and the interval s3 were not significantly different between implanted and non-implanted embryos. Limitations, reasons for caution Morphokinetic annotation is open to inter- and intraobserver variability, although partially controlled by prospectivity. The ROC-analysis does not adjust for statistical dependence between embryos of one patient. The used morphological groups do not necessarily measure developmental potential. Embryos with KID showed less morphokinetic variation because of pre-selection for transfer. Wider implications of the findings The findings could contribute to an early decision model on transfer hierarchy that would help avoiding costs and a potential negative influence of prolonged blastocyst culture. Observing the kinetics of blastocyst formation (tM, tSB, tEB) might allow an improved prediction of implantation potential. Trial registration number DRKS00022983