Abstract: Objective To evaluate the methodology and performance of serum glycated albumin(GA) determination by enzymatic method as routine method and the correlation of glycosylated hemoglobin A_1c (HbA_1c) with GA, in order to provide a better method for the early diagnosis and effect monitoring of diabetes mellitus. Methods The enzymatic method for GA was evaluated for precision, linearity, bias, carryover rate and drift according to the Clinical and Laboratory Standards Institute (CLSI) EP10-A3. The GA and HbA_1c were determined in 336 patients with diabetes mellitus. The correlation between GA and HbA_1c was analyzed. According to American Diabetes Association (ADA) standard (HbA_1c<7% as blood glucose controlling good, HbA_1c 7%-8% as blood glucose controlling common, HbA_1c>8% as blood glucose controlling poor), the cut-off value of GA/HbA_1c ratio was evaluated by receiver operating characteristic (ROC) curve. Results The total mean imprecision was 1.62%, and the mean bias was 7.58%.The intercept, slope, nonlinearity, carryover rate and drift of enzymatic method were acceptable. The nonlinearity parameter had forward bias when determining high-value, which should be observed further, however, there was no influence on its clinical diagnosis. The correlation analysis indicated that GA was well correlated with HbA_1c[correlation coefficient (r)=0.879, P<0.01]. The optimal cut-off value of GA/HbA_1c ratio for blood glucose controlling was 2.60. The sensitivity was 86.8%, and the specificity was 70.1%. GA/HbA_1c ratio>2.60 was considered as blood glucose controlling poor. Conclusions The routine determination of GA by enzymatic method is available. GA can be used into the diagnosis and effect evaluation of GA. GA/HbA_1c ratio is valuable to monitor blood glucose controlling.

Key words: Glycated albumin, Glycosylated hemoglobin, Enzymatic method, Diabetes mellitus