Differentiating transient hyperglycemia from diabetic hyperglycemia can be difficult in cats since single blood glucose measurements reflect only momentary glucose concentrations, and values may be elevated because of stress-induced hyperglycemia. Glycated protein measurements serve as monitors of longer-term glycemic control in human diabetics. Using an automated nitroblue tetrazolium assay, fructosamine concentration was measured in serum from 24 healthy control cats and 3 groups of hospitalized cats: 32 euglycemic, 19 transiently hyperglycemic, and 12 diabetic cats. Fructosamine concentrations ranged from 2.1 - 3.8 mmol/L in clinically healthy cats; 1.1 - 3.5 mmol/L in euglycemic cats; 2.0 - 4.1 mmol/L in transiently hyperglycemic cats; and 3.4 to >6.0 mmol/L in diabetic cats. Values for with-in-run precision at 2 fructosamine concentrations (2.64 mmol/L and 6.13 mmol/L) were 1.5% and 1.3%, respectively. Between-run coefficient of variation was 3.8% at a fructosamine concentration of 1.85 mmol/L. The mean fructosamine concentration for the diabetic group differed significantly (P=0.0001) from the mean concentrations of the other 3 groups. Poorly regulated or newly diagnosed diabetic cats tended to have the highest fructosamine values, whereas well-regulated or over-regulated diabetic cats had values approaching the reference range. As a single test for differentiating nondiabetic cats from diabetic cats, fructosamine was very sensitive (92%) and specific (96%), with a positive predictive value of 85% and a negative predictive value of 98%. Serum fructosamine concentration shows promise as an inexpensive, adjunct diagnostic tool for differentiating transiently hyperglycemic cats from poorly controlled diabetic cats.