Glycated Albumin and Risk of Death and Hospitalizations in Diabetic Dialysis Patients

Background and objectives Relative to hemoglobin (Hb) A(1c) glycated albumin (GA) more accurately reflects glycemic control in patients with diabetes mellitus and ESRD. We determined the association between GA, HbA(1c) and glucose levels with survival and hospitalizations in diabetic dialysis patients. Design, setting, participants, & measurements Quarterly GA levels were measured for up to 2.33 years in 444 prevalent patients with diabetes and ESRD. Proportional hazard time-dependent covariate models were computed with adjustment for demographic characteristics, comorbidities, and laboratory variables. Similar analyses were performed for available HbA(1c) and monthly random serum glucose determinations. Results The participants were 53% male, 54% African American, 43% Caucasian, 90% on hemodialysis, with a mean (SD) age of 62 (12) years and median follow-up duration of 2.25 years. GA and HbA(1c) mean +/- SD 21.5% +/- 6.0%, median 20.4% and mean +/- SD 6.9% +/- 6.6%, median 1.6%, respectively. There were 156 deaths during the observation period. In best-fit models, predictors of death included increasing GA, increasing age, presence of peripheral vascular disease, decreasing serum albumin, and decreasing hemoglobin concentrations. HbA(1c) and random serum glucose concentrations were not predictive of survival. Increasing GA levels were associated with hospitalization in the 17 days after measurement, whereas HbA(1c) was not. Conclusions In contrast to the HbA(1c) and random serum glucose values, GA accurately predicts the risk of death and hospitalizations in patients with diabetes mellitus and ESRD. The GA assay should be considered by clinicians who care for patients with diabetes on dialysis. Clin J Am Soc Nephrol 6: 1635-1643, 2011. doi: 10.2215/CJN.11491210