The aim of this study was to validate the measurement of glucose appearance rate using [6,6-2H2]glucose i.v. infusion in lactating dairy cows. Sample enrichments were analysed by gas chromatography/mass spectrometry. Linearity (enriched solutions) and specificity (enriched plasma) were good: for enrichments ranging between 1.6 and 6.3 mol% excess, the slopes were about 1 and the ordinates at the origin were not different from zero. For a plasma enriched at 3.74 mol% excess, repeatability and long term intralaboratory reproducibility coefficients of variation were 1.31 and 1.90%, respectively. The appearance rates were calculated by two models. The values provided by the steady-state model were not different from those provided by the non-steady-state Steele model. Both models can be used because the treatment effects were similarly discriminated regardless of the model. In our experiments analysing the nutritional effects on Ra in mid-lactating cows, the precision of the method (1.90%) was not the limiting factor to detect a significant difference in Ra compared to the statistical precision obtained with the experimental scheme (4 x 4 and 5 x 5 Latin square design). We conclude that in lactating dairy cows, the measurement of glucose fluxes with this method is relevant and minimally invasive for the animals.