Abstract: Objective：To explore the relationship between the embryodynamic parameters using time-lapse system and the pregnancy outcomes, so as to establish a model of the optimal embryo selection. Methods: A total of 137 cycles with the time-lapse system (IVF/ICSI-ET cycle) were retrospectively analyzed from December 2016 to December 2017. The clinical data including basic information and pregnancy outcomes were collected. The morphokinetic parameters using the time-lapse system (t2、t3、t4、t5、t6、t7、t8、cc2、s2、cc3、s3) were compared between the implanted embryos and the unimplanted embryos. Based on those parameters related with embryo implantation, a multivariate classification model was developed. This model was then primarily verified. Results: ①Totally 171 embryos had one or more early abnormal division event，in which 16 were considered to be implantable, and then transplanted. There were 4 embryos implantated succesfully. ②There were no significant differences in all of 11 morphokinetic parameters between the implanted embryos and the unimplanted embryos (P>0.05). ③The t6 (OR=1.462, 95%CI: 1.093-1.954, P=0.010) and t7 (OR=0.803, 95%CI: 0.653-0.989, P=0.039) could be the best predictors of implantation. A classification model was developed with these two parameters combined with the exclusion standard of abnormal division event. The implantation rate was declined as the embryonic grade dropped. The area under the receiver operating characteristic curve (AUC) was 0.649 (95%CI: 0.528-0.770). Conclusions: Two of morphokinetic parameters (t6, t7) could be closely related to the implantation potential. We tried to develop an embryo selecting model with these two parameters combined with exclusion standard of abnormal division events. It is true that this model should be furtherly validated in our centers.

Key words: Fertilization in vitro, Sperm injections, intracytoplasmic, Embryo transfer, Time-lapse system, Kinetics, Pregnancy outcome