Fructosamines are formed when glucose reacts non-enzymatically with amino groups on proteins, and previous studies have indicated that the serum fructosamine concentration could be of importance in the diagnosis of canine diabetes mellitus. Owing to the connection between the protein/albumin concentration and serum fructosamine concentration, it has been suggested that the serum fructosamine concentration should be corrected for the protein/albumin concentration. Thus, the purpose of the present study was to evaluate the uncorrected serum fructosamine concentration and various protein and albumin corrections of the serum fructosamine concentration in the separation of dogs with diabetes mellitus from dogs with other diseases that presented with clinical signs suggestive of diabetes mellitus. The evaluation was assisted by relative operating characteristic curves (ROC curves), which may be used to compare various diagnostic tests under equivalent conditions (equal true positive ratios or false positive ratios) and over the entire range of cutoff values. A total of 58 dogs (15 dogs with diabetes mellitus and 43 dogs with other diseases) were included in the study. Serum fructosamine concentration, serum total protein concentration and serum albumin concentration were measured in each dog, and various corrections of the serum fructosamine concentration for protein or albumin concentration were made. Comparing the ROC curves of the uncorrected and each corrected serum fructosamine concentration indicated that there was no decisive difference between the uncorrected and the corrected serum fructosamine concentrations in discriminating between dogs with and without diabetes mellitus. Hence, correcting the serum fructosamine concentration as a routine procedure cannot be advocated from the results of the study. Moreover, the sensitivity and specificity of the uncorrected serum fructosamine concentration were very high, 0.93 and 0.95, respectively, further evidence of the value of the uncorrected serum fructosamine concentration in the diagnosis of canine diabetes mellitus.
