Non-invasive viability assessment of day-4 frozen–thawed human embryos using near infrared spectroscopy

This study investigated if metabolomic profiling of culture media using near infrared (NIR) spectroscopy was related to live-birth rates after single-embryo transfer of frozen–thawed embryos. Analysis of culture media of frozen–thawed embryos was performed by NIR spectroscopy. A viability score was calculated using a predictive multivariate algorithm of fresh day-5 embryos with known pregnancy outcomes. This algorithm generated with fresh day-5 embryos could help to identify the live-birth group from the no live-birth group. Multivariable regression models that tested the predictive ability of the viability score for live birth showed an odds ratio in the crude analysis of 1.50 (P=0.008), after adjustment for embryo morphology, 1.44 (P=0.022), and after adjustment for all variables, 1.71 (P=0.005); based on a 0.1 step increase in viability scores. In conclusion, higher viability scores resulted in higher live-birth rates. An algorithm generated from fresh embryos might be used to predict viability of frozen–thawed embryos. Frozen–thawed embryos have different metabolic activity which is related to implantation potential. Therefore, this method might be useful to select the best embryo for transfer within a group of embryos with similar morphology.