Correlation of Glycated Albumin and Glycated Hemoglobin with Glycemic Control in Patients with Diabetic Chronic Kidney Disease on Hemodialysis

Background and aims. Diabetes mellitus (DM) is the most common cause of end-stage renal disease (ESRD) worldwide and is the primary etiology of ESRD in Indonesia. It is estimated that there are 10 million patients with DM in Indonesia and about 25-40% of diabetics develop diabetic nephropathy (DN) within 25 years. Poor glycemic control is associated with increased mortality in a large observational study among diabetics on hemodialysis. Glycated hemoglobin, or known as HbA1c, was the currently recommended biomarker to monitor long-term glycemic control in diabetes mellitus guideline. However, it becomes underestimated in patients with DN on hemodialysis (DN-HD), because of 20- 50% reduction of erythrocyte lifespan, clinical use of iron therapy, and effects of recombinant human erythropoietin. Glycated albumin (GA), ketoamine formed via a non-enzymatic glycation reaction of serum albumin, is free from interference by erythrocyte lifespan or erythropoietin therapy and subsequently can be used as an alternative biomarker to monitor glycemic control in DN-HD. As albumin has a half-life of only 14-21 days* compared to Hb 60-100 days in well ESKD patients, thus GA may provide a better reflection of recent diabetic control than HbA1c. The aim of this study was to analyze the correlation between GA and HbA1c with glycemic control in DN-HD. Methods The study was an analytical observational study with a cross-sectional design. The subjects were consecutive patients with DN-HD who visited Hemodialysis Unit at Mohammad Hoesin General Hospital Palembang during August-November 2014. The glycemic control index was determined by the average value of 2 times a week pre-HD random blood glucose for 4 weeks (aRBG). Results. The subjects were 25 patients with an average age of 56.16±7.49 years old. The average value of GA was 26.94±7.74%. GA was strongly correlated with aRBG with r=0.776; p=0.000. After correcting for age, sex, and BMI, the correlations became significantly very strong (r=0.809, p <0.001). The simple linear regression for the relationship between GA and aRBG was aRBG=4,62×GA+42.74 (R2=0.602, P<0.001), estimating that a 1% increase of GA was associated with 4.62mg/dL increase of aRBG. After correcting for age, sex, & BMI, the correlations between HbA1c and aRBG were significant (r=0.852, p<0.001). Conclusion. GA was strongly correlated with glycemic control in patients with DN-HD and HbA1c was correlated better.