Morphokinetic analysis of euploid blastocysts: searching for non-invasive criteria of embryo implantation additional to chromosomal assessment

Study question: Can we define any morphokinetic parameter able to increase our predictive power upon implantation potential of euploid blastocysts cultured in time-lapse? Summary answer: Blastomeres’ symmetry at 4 cell-stage and blastocyst morphological quality after full-expansion were significantly correlated with the implantation of euploid blastocysts. What is known already: To date, the assessment of a normal chromosomal constitution through comprehensive-chromosome-testing techniques represents a powerful predictive parameter upon blastocyst’s reproductive potential. However, ≈50% of euploid blastocysts fail to implant. The study of preimplantation embryonic development, via morphokinetic parameters in time-lapse, do not provide to date solid evidence of correlation with the clinical outcomes after IVF. Although, the additional predictive value of morphokinetic in preimplantation-genetic-testing (PGT-A) cycles conducted in time-lapse is yet to be assessed. Study design, size, duration: Retrospective exploratory study. All transferred euploid blastocysts obtained after consecutive PGT-A cycles conducted in a time-lapse system between March 2014-March 2017 were included (n = 320). All morphokinetic parameters of embryo evaluation were compared among implanted and not implanted euploid blastocysts. Logistic regression analyses and receiver operating curve (ROC) analysis were conducted. Power calculations for significant differences were performed with the software G*Power v3.1. Participants/materials, setting, methods: 130 implanted and 190 not implanted euploid blastocysts were included and confronted for the following timings: tPB2(2ndPolar-Body extrusion), tPNf (pronuclei formation), t2,t3,t4,t5, t8 (2-,3-,4-,5- and 8-cell division), cc1(t3-t2), cc2(t4-t3), cc3(t5-t4), tM(morula), tSB(starting-blastulation), tB(full-expansion). Furthermore, blastomeres’ symmetry and fragmentation at t4, and blastocyst’s quality at tB were evaluated and blindly-confirmed by two independent operators. The results were corrected through univariable and multivariable logistic regression analyses for all putative confounders including maternal age. Main results and the role of chance: Significant differences were reported between implanted and not implanted euploid blastocysts for cc2 (1.3 ± 2.3 h,0-12.6 h versus 2.5 h±4.3 h,0-26.5 h, p<0.01), cc3 (12.4 ± 5.2 h,0.3- 45.5 h versus 10.7 ± 6.5 h,0-43.4 h, p = 0.02) and tM (88.7 ± 10.5 h,66.9- 121.7 h versus 91.4 ± 10.0 h,61.3-123.1 h, p = 0.03). Furthermore, low fragmentation (<10%; n = 111/243, 45.7%) at t4 resulted in higher ongoing implantation rates after euploid blastocyst transfer than high-fragmentation (≥10%; n = 9/77,24.7%; p<0.01 and power = 92%). Similarly, even blastomeres’ symmetry (n = 114/241, 47.3%) at t4 resulted into a better reproductive outcome than uneven one (n = 16/79,20.3%; p<0.01 and power>99%). At last, blastocysts’ quality at tB was also correlated with a positive ongoing implantation (excellent: n = 86/184,46.7%; good: n = 21/49,42.9%; average: n = 17/51,33.3% and poor: n = 6/36,16.7%; p<0.01 and power = 95%). The univariable logistic regressions confirmed these data. Conversely, the multivariable logistic model outlined only blastomeres’ symmetry (even/uneven, OR = 0.36, 95%CI:0.19-0.67; p<0.01) and blastocyst’s quality (excellent/good/average/ poor, OR = 0.77, 95%CI:0.61-0.98; p = 0.04) as predictive of implantation success/failure. The ROC curve analysis defined an area under the curve (AUC) for this model of 0.65, 95%CI:0.60-0.71, p<0.01. Limitations, reasons for caution: The main limitations of this study are its retrospective nature and the subjectivity of blastomeres’ symmetry definition at t4. Possibly, these data should be confirmed through a computational/mathematical assessment. Moreover, the extension of the sample size may allow the evaluation also of anomalous patterns of development, such as direct/reverse cleavage. Wider implications of the findings: A model including blastomeres’ symmetry at 4cell-stage and blastocyst’s quality at full-blastulation may represent a valuable selection tool to predict euploid blastocysts’ reproductive competence for any PGT-A cycle, regardless the adoption of a time-lapse system. Nonetheless, these data should be confirmed from a prospective study. Trial registration number: None.