P–282 A time lapse analysis of 36,671 embryos to compare the incidence of early stage abnormal cleavage events in ICSI and IVF derived embryos

Abstract

Abstract Study question Is this incidence of early stage abnormal cleavage events different between embryos created following ICSI compared with IVF? Summary answer Embryos derived from ICSI are more likely to exhibit abnormal cleavage compared with those from IVF. This difference is most marked in women ≥35 years. What is known already Time lapse imaging (TLI) has been instrumental in allowing detailed annotation of early embryo development to provide an objective aid for embryo selection in ART cycles. Amongst several abnormal cleavage events reported, rapid cleavage and multichotomous mitosis/direct cleavage, during the first days after fertilisation have been demonstrated to be associated with lower blastulation rates, reduced implantation potential, increased aneuploidy and poor pregnancy outcomes. With ICSI being utilised commonly, and being the insemination method of choice in some clinics, the incidence of abnormal cleavage was investigated in association with insemination method, ICSI or IVF. Study design, size, duration The incidence of abnormal cleavage events was evaluated in a large multicentre retrospective analysis of 36,671 embryos from 6689 patients treated in 8 IVF clinics enabled with time lapse imaging, between 2011 - 2019. This constituted 10931 IVF embryos and 25740 ICSI embryos. Participants/materials, setting, methods Following ICSI or after IVF fertilisation check, embryos were time-lapse imaged every 10 minutes and annotated using the EmbryoScope. Second cell cycle durations were calculated as follows: time to reach 3-cell (t3) from 2-cell (t2) (t3-t2 = cc2). These were analysed using a welch t-test as three groups of abnormal cleavage: direct cleavage/trichotomous mitosis (DC) - where cc2=0 hours(h), rapid cleavage within 2h (R2) - where 0<cc2<2h and rapid cleavage between 2–5h (R5) where 2h<cc2<5h. Main results and the role of chance The incidence of DC, R2 and R5 in the whole cohort of embryos was 5%, 8% and 9% respectively. In the subpopulation of IVF embryos the incidence of DC, R2 and R5 was 4%, 8% and 9% respectively. In the subpopulation of ICSI embryos the incidence of DC, R2 and R5 was 6%, 8% and 9% respectively. The incidence of DC was significantly higher in ICSI embryos compared with IVF (p < 0.001) whilst R2 and R5 were the same. ICSI derived embryos had a mean (± SE) cc2 value of 9.39 ± 0.03h, compared with 9.56 ± 0.05h for IVF embryos (p < 0.0038). Examination of data split by maternal age demonstrated that ICSI oocytes from women of advanced maternal age (≥35) also had significantly more embryos exhibiting rapid cleavages R2 and R5 than IVF oocytes (p < 0.007). There were no significant differences however, in rates of abnormal cleavages between ICSI and IVF in embryos from women aged <30 (p = 0.06). Male-factor diagnoses showed no significant differences in abnormal cleavage values between ICSI or IVF in all three abnormal cleavage categories. Limitations, reasons for caution This analysis could not control for all potential confounders therefore it is possible that the increased abnormal cleavages observed in this investigation are a result of another, or combination of factors. Despite quality assurance programs being in place across all clinics, there is a risk of annotation bias. Wider implications of the findings: There is a higher incidence of early abnormal cleavage in embryos derived from ICSI, particularly in those from women of increased age and this research may help elucidate the reasons for this and add to the debate regarding the appropriateness of the increasing use of ICSI. Trial registration number Not applicable