Re-evaluation of glycated hemoglobin and glycated albumin with continuous glucose monitoring system as markers of glycemia in patients with liver cirrhosis

Liver cirrhosis (LC) is frequently accompanied by glucose intolerance. The present study was designed to determine whether glycated hemoglobin A1c (HbA1c) and glycated albumin (GA) were predictive markers of glycemia, as determined by a continuous glucose monitoring system (CGMS), in patients with LC. A total of 30 patients with LC, including 3, 19, 5, 2 and 1 with LC due to hepatitis B virus, hepatitis C virus, non-alcoholic steatohepatitis, alcohol and unknown causes, respectively, were assessed by CGMS. The average, maximum and minimum blood glucose (BG) levels were measured by CGMS, and correlated with HbA1c and GA. The average, maximum and minimum BG in these individuals were 142±38.7, 209.3±65.7 and 85.1±25.4 mg/dl, respectively. HbA1c was significantly correlated with average BG (r=0.447, P=0.015) and maximum BG (r=0.523, P=0.004). In addition, GA was significantly correlated with average BG (r=0.687, P<0.001) and maximum BG (r=0.648, P<0.001). Neither HbA1c nor GA was significantly correlated with minimum BG. Correlation analysis yielded formulas by which HbA1c and GA were predictive of average BG in individuals with LC: Average BG=19.2 × HbA1c (%) + 36.5 and average BG=6.6 × GA (%) + 13.0, respectively. In conclusion, HbA1c and GA showed significant correlations with average and maximum BG, as determined by CGMS. The derived formulas allow for estimates of average BG based on HbA1c and GA, and may contribute to the control of glycemia in patients with LC.