ABSTRACT
BACKGROUND: To create a more effective selection standard for early embryos, we developed a new grading system consisting of conventional morphological evaluation in combination with analysis of blastomere nuclei. METHODS: A total of 744 embryos used during 459 cycles of embryo transfer on day 2 and blastocyst transfer were subjected to retrospective analysis. The overall implantation rate was 15.5% (115/744). Morphological evaluation of the embryos was performed on day 2 by referring to both the size of blastomere and fragmentation (conventional method) and the nucleic features of the blastomeres--either multinucleated or anucleic (nuclei counting method). The implantation rate for every transferred embryo and blastocyst was examined. RESULTS: Although a high implantation rate was observed with the highest quality embryos as judged by either the conventional method (24.1%; 57/237) or the nuclei counting method (26.1%; 104/399), the nuclei counting method predicted implantation rate better than the conventional method. The embryos that were considered to be high quality according to the conventional method, but low quality according to the nuclei counting method, had a limited implantation success rate of 6.3% (4/66). Also, after blastocyst transfer, implantation occurred most often when high quality embryos evaluated by the nuclei counting method were used (25.5%; 25/98), while the blastocysts from low quality embryos seldom implanted (3.2%; 2/63). CONCLUSIONS: When choosing which embryo to transfer, the normality of blastomere nuclei may be a more important index of quality than standard fragmentation features and/or blastomere uniformity analysis. When choosing among embryos, if nucleic status is identical, then embryos with the least fragmentation should be chosen. Moreover, in blastocyst transfer, a blastocyst whose nuclei were judged normal on day 2 should be selected on day 5 over any other blastocysts.
