Assessment of the implantation of day-2 human embryos by morphometric nonsubjective parameters

Abstract

OBJECTIVE: To demonstrate the usefulness of image analysis in designing objective embryonic morphometric variables. DESIGN: Retrospective study of 214 top-quality day-2 embryo photographs from 50 double-embryo transfers resulting in no pregnancy (group 0) and 57 resulting in twin pregnancy (group 1). SETTING: Human reproduction unit. PATIENT(S): Study of 107 in vitro fertilization-intracytoplasmic sperm injection (IVF-ICSI) cycles in women age <36 years with double-embryo transfer of top-quality embryos. Only the first cycle of IVF-ICSI was included. INTERVENTION(S): Standard IVF-ICSI protocols. MAIN OUTCOME MEASURE(S): The embryo photographs were analyzed using the ImageJ program. The effects of the embryo variables and the clinical variables on embryo implantation were evaluated using a stepwise dichotomous logistic regression. RESULT(S): Significant differences were observed, owing to the women's ages, internal perimeter, roundness factor, and zona pellucida thickness. Embryos with smaller internal perimeter, circular shape, and thinner zona pellucida were more likely to implant. CONCLUSION(S): Morphometric variables lower the subjectivity of the current embryo grading systems. These variables are nonsubjective factors to consider when predicting implantation. Embryo image analysis is an accurate tool that can improve IVF-ICSI outcomes and reduce the number of twin pregnancies.