A Glycemic Control Performance Study For Assessment Of Glycated Albumin And Glycated Hemoglobin In Diabetic Patients Undergoing Hemodialysis

Objective Foreign studies suggested that glycated hemoglobin (HbAlc) underestimate the actual glycemic control levels in maintenance hemodialysis (MHD) patients, because of anemia and using of erythropoietin (EPO), and recommended that glycated albumin (GA) should be an alternative marker, but there wasn't domestic verification study yet. We assess glycemic control performance of GA and HbAlc in MHD patients with diabetes mellitus (DM) and without DM by regarding 7 measure values of their blood glucose during a whole day as reference standard, and look for the non-blood glucose factors of them. Methods The cases were enrolled from blood purification centers of four hospitals. At the time registered and 2 months later when follow-up,7 glycemic levels were measured before and at 2 hours after three meals, and at bedtime, in the day before dialysis, so as to calculate the mean blood glucose. At the next day (dialysis day), the indicators such as HbAlc, hemoglobin (Hb), albumin (ALB), and so on were measured. We also retained plasma and serum samples to detect GA concentrately. At last,109 cases who meet inclusion criteria were divided to non-diabetes mellitus (NDM) group (n=41) and DM group (n=68), and then were investigated. Results There were no difference in these 2 groups by comparing indicators including age, gender, vintage and other general clinical information. In the patients of population and DM group, both GA and HbAlc were significantly correlated with mean blood glucose (p<0.01), but the distinction between them didn't be given by their correlation coefficients r. By comparing the area under curve (AUC) of receiver operating characteristic (ROC) curve in DM patients, assessment performance of HbAlc is better than GA (p<0.05). We found that the effects of Hb and EPO dose on HbAlc, ALB on GA, and inflammatory markers such as high sensitivity C-reactive protein (HsCRP) on GA and HbAlc were all not obvious. Conclusion Although using HbAlc in MHD patients may underestimate actual glycemic level, but in terms of correlation with blood glucose, it isn't lower than GA, and it is still a better indicator than latter in assessing glycemic control performance, Therefore, HbAlc cannot be replaced currently, but the control target in MHD patients may need to be reevaluated separately. Hb and EPO dose haven't obvious effect on HbAlc. The effects of ALB on GA, inflammation on GA and HbAlc isn't obvious in our study too, it is still difficult to draw convincing conclusions.