Analysis of the morphological dynamics of blastocysts after vitrification/warming: defining new predictive variables of implantation.

Objective: To describe the morphological dynamics of vitrified/warmed blastocysts and to identify quantitative morphological variables related to implantation. Subsequently, by using the most predictive parameters, to develop a hierarchical model by subdividing vitrified/warmed blastocysts into categories with different implantation potentials.

Design: Observational, retrospective, cohort study.

Setting: University-affiliated private IVF center.

Patient(s): The study included 429 vitrified/warmed blastocysts with known implantation data, which were evaluated by time-lapse imaging. Blastocysts were routinely placed in EmbryoScope (Vitrolife) immediately after warming until transfer.

Intervention(s): None.

Main Outcome Measure(s): Embryos were vitrified and warmed by the Cryotop method (KitazatoBiopharma). The studied variables included the initial and minimum thicknesses of zona pellucida (μm), the initial and maximum areas (μm), the area of inner cell mass (μm), expansion (whether the embryo reexpands or not after warming), and collapsing or contraction after warming. After defining the optimal ranges according to the consecutive quartiles with the highest probability of implantation, a logistic regression analysis was performed by combining the former variables and the blastocyst morphological classification criteria defined by the Spanish Association of Embryologists into A, B, C, or D categories.

Result(s): Reexpansion of vitrified/warmed blastocysts correlated strongly with implantation (44.6% for reexpanded vs. 6.5% for the blastocysts that did not reexpand after warming). Throughout the logistic regression analysis, the model identified the maximum blastocyst area, odds ratio (OR) = 0.41 (95% confidence interval [CI], 0.22-0.77), followed by the initial area, OR = 0.62 (95% CI, 0.35-1.08) as the most predictive variables related to implanting embryos. Blastocyst morphology was not considered relevant in our model. The hierarchical tree model subdivided embryos into four categories, A-D, with lowering expected implantation potentials (from 47.3% for A to 14.2% for D).

Conclusion(s): The analysis of warmed blastocysts by time-lapse imaging may provide objective quantitative markers for the blastocyst implantation potential. We propose a hierarchical model to classify vitrified/warmed blastocysts according to their implantation probability. The observed correlations and the proposed algorithm should be validated in a prospective trial to evaluate its efficacy.