BACKGROUND: In vitro glycation of transferrin leads to increased oxidative stress by impairing iron-binding antioxidant capacity. The aim of this study is to develop a method to evaluate in vivo transferrin glycation in diabetes. METHODS: We adapted the nitroblue tetrazolium assay to measure in micro-well plates the fructosamine content of transferrin isolated from serum by immunocomplexation. RESULTS: Introduction of the immunocomplexation step did not affect the analytical performance of the fructosamine measurement and analytical variability was lower than 7%. The diabetic group (n=107) had significantly higher transferrin glycation (1.39+/-1.12 versus 0.79+/-1.09 micromol fructosamine/g transferrin in the non-diabetic group, n=91, p<0.0005) and this was most pronounced in type 1 diabetes (1.95+/-1.02 versus 1.06+/-1.04 micromol fructosamine/g transferrin in type 2, p<0.0005). Transferrin glycation was associated with parameters of glycaemic control but did not correlate with serum iron or total iron-binding capacity. Total iron-binding capacity was lower in type 1 diabetes (63+/-9 versus 69+/-12 micromol/l in type 2, p<0.05) and was mainly determined by transferrin concentration. CONCLUSIONS: These results indicate that the adapted nitroblue tetrazolium assay combined with immunocomplexation of serum transferrin is suitable to detect differences in in vivo transferrin glycation between non-diabetic, type 1 and type 2 diabetic subjects