1H NMR based profiling of spent culture media cannot predict success of implantation for day 3 human embryos

Authors:
Publisher website
Google Scholar
Abstract
Background Identification of a non-invasive technique to assess embryo implantation potential in assisted reproduction would greatly increase success rates and lead more efficiently to single embryo transfer. Early studies suggested metabonomic analysis of spent culture media could improve embryo selection. The goal of this study is to assess if embryo implantation can be predicted based on proton nuclear magnetic resonance (1H NMR) profiles of spent embryo culture media from patients undergoing transfer of multiple embryos on cycle day 3. Method We conducted a retrospective study in an academic assisted reproduction technology (ART) program and analyzed the data in a university research center. Two hundred twenty-eight spent culture media samples originating from 108 patients were individually analyzed. Specifically, five distinct sets (1 to 5) of different types of spent media samples (volume ~14 μL) from embryos that resulted in clinical pregnancy (positive heart rate at 6 weeks gestation) (n 1 = 29; n 2 = 19; n 3 = 9; n 4 = 12; n 5 = 33; n total = 102) and from embryos that did not implant (n 1 = 28; n 2 = 29; n 3 = 18; n 4 = 15; n 5 = 36; n total = 126) were collected on day 3 of embryo growth. The media samples were profiled using 1H NMR spectroscopy, and the NMR profiles of sets 1 to 5 were subject to standard uni- and multi-variate data analyses in order to evaluate potential correlation of profiles with implantation success. Results For set 1 of the media samples, a borderline class separation of NMR profiles was obtained by use of principal component analysis (PCA) and logistic regression. This tentative class separation could not be repeated and validated in any of the other media sets 2 to 5. Conclusions Despite the rigorous technical approach, 1H NMR based profiling of spent culture media cannot predict success of implantation for day 3 human embryos.

This publication has 31 references indexed in Scilit: