Day 3 embryo selection by metabolomic profiling of culture medium with near-infrared spectroscopy as an adjunct to morphology: a randomized controlled trial

Is the selection of a single Day 3 embryo by metabolomic profiling of culture medium with near-infrared (NIR) spectroscopy as an adjunct to morphology able to improve live birth rates in IVF, compared with embryo selection by morphology alone? The live birth rate after embryo selection by NIR spectroscopy and morphology is not significantly different compared with the live birth rate after embryos were selected by morphology alone. The elevated incidence of pregnancy and neonatal problems associated with a high-twinning rate after IVF can only be successfully reduced by the transfer of one embryo. Current embryo assessment methods are unable to accurately predict the reproductive potential of an individual embryo. Today, a number of techniques are said to be more accurate at selecting the best embryo. One of these new technologies is metabolomic profiling of spent embryo culture media with the use of NIR spectroscopy. A double-blind, randomized controlled trial was conducted between 2009 and 2011, and included 417 couples undergoing IVF with a single embryo transfer. Randomization was performed centrally just before Ovum Pick-Up (OPU), using a computerized randomization program. Both patient and physician were unaware of the treatment allocation. To ensure blinding, the allocations were placed in consecutively numbered, opaque envelopes. Patients were randomized (1:1) into either the control group (embryo selection by morphology only) or the treatment group (embryo selection by morphology plus NIR spectroscopy of embryo culture medium). At OPU, 208 patients were randomized to the morphology only group and 209 patients were randomized to the morphology plus viability score group. On Day 3, 163 patients in the control group and 146 patients in the treatment group met the inclusion criteria. The study was conducted in an academic hospital with IVF laboratory and three non-academic hospitals. Patient demographics and baseline characteristics were distributed equally over the two groups, except for embryo fragmentation, which was significantly higher in the treatment group. In the intention to treat analysis, the live birth rates were 31.7 and 26.8 for the control group and the treatment group, respectively (relative risk 0.84; 95 confidence interval 0.631.14, P 0.27). In the per protocol analysis, the live birth rates were 31.3 and 29.5 for the control group and the treatment group, respectively (relative risk 0.94; 95 confidence interval 0.671.32, P 0.73). For the treatment group, the embryological technicians independent choice (by morphology) of which embryo to transfer was recorded 138 times. In 75.4 (104 of 138) of the transfers, the embryo with the best morphology did not have the highest viability score. The live birth rate of these 104 transferred embryos was 30.8. A possible limitation of our study is the pre-selection of all embryos by morphology and dividing the cohort of available embryos into two groups: good quality embryos and poor quality embryos. As a consequence, we have probably selected for a better prognosis patient group. To avoid the use of incompetent embryo selection tools at the expense of the patient, an evidence-based proof of clinical usefulness is essential before the implementation of new diagnostic tools in IVF laboratories. Dutch Trial Registry, registry number NTR1178.