Usefulness of glycated albumin as a predictor of mortality in chronic hemodialysis patients with diabetes: a multi-center, prospective cohort study

Background The association of glycated albumin (GA) with mortality is unclear in chronic hemodialysis patients with diabetes. We investigated the usefulness of GA by comparing it with hemoglobin A1c (HbA1c) in this patient population. Research design and methods This was a multi-center, prospective cohort study of 841 Japanese chronic hemodialysis patients with diabetes. There were 235 women and 606 men included, with a mean age of 64 years. The primary and secondary endpoints were the incidence of all-cause and cause-specific mortality, respectively. The hazard ratios of GA and HbA1c for the endpoints were estimated using the values at baseline and during the study period. Results During the mean follow-up period of 3.1 years, there were 184 deceased cases, in which 30 and 154 resulted from atherosclerotic cardiovascular disease (ASCVD) and non-ASCVD, respectively. The hazard ratio for a 1% increase in GA was 1.033 (95% confidence interval 1.006-1.060, p = 0.017) for all-cause mortality with a statistical significance when GA was treated as a time dependent variable, but not when the baseline levels or the mean levels during the follow-up period were used in the analysis (p = 0.815 and 0.517, respectively). GA was a significant predictor for ASCVD-related mortality in the above 3 models, but was not for non-ASCVD mortality. Higher levels of HbA1c were only associated with ASCVD-related mortality when HbA1c was treated as a time-dependent variable. Conclusions GA may be useful compared to HbA1c for predicting all-cause and ASCVD-related mortality in Japanese patients with diabetes undergoing chronic hemodialysis.