A predictive model for blastocyst formation based on morphokinetic parameters in time-lapse monitoring of embryo development

Purpose

The aim of the study was to create a predictive model of blastocyst development based on morphokinetic parameters of time-lapse embryoscope monitoring.

Methods

Time-lapse recordings of 432 embryos (obtained from 77 patients), monitored in Embryoscope, were involved in the study. Patients underwent in vitro fertilization according to standard procedure between June 2012 and April 2013. A retrospective analysis of morphokinetic features, focused on duration of time from the Intracytoplasmic Sperm Injection (ICSI) procedure to consecutive embryo division for 2, 3, 4 and 5 blastomeres, as well as time intervals between each division, was conducted. All embryos were observed for 5 days.

Results

Based on the distribution of analyzed morphokinetic parameters and number of embryos developed to blastocyst, a range denoting the possibility of an embryo reaching blastocyst stage was determined. According to the obtained results, univariate and multivariate logistic regression analyses were performed. Based on the times of division for two and five blastomeres and intervals between the second and third division, a multivariate predictive model was created. The predictive equation was constructed based on the parameters of logistic regression analysis (odds ratios). Statistically significant differences (p < 0.001) in the size of the prediction parameter between the group of embryos developed to blastocyst (the median value: Me = 9.95, and quartiles: Q₁ = 7.59, Q₃ = 12.30) and embryos that did not develop to the blastocyst stage (Me = 4.66, Q₁ = 2.33, Q₃ = 8.19) were found. A Receiver Operating Characteristic (ROC) curve was created for the constructed predictive model. The Area Under the Curve was AUC = 0.806 with a 95 % confidence interval (0.747, 0.864). The predictive model constructed in this study has been validated using an independent data set, which indicates that the model is reliable and repeatable.

Conclusions

Time-lapse imaging presents a new diagnostic tool for parametric evaluation of embryo development, from the oocyte stage, through fertilization, up to the blastocyst stage. The assessment of morphokinetic parameters can help us to provide more accurate information about the reproductive potential of embryos. It allows for early selection of embryos with high reproductive potential and shortens embryo incubation.