Erratum to: Glycated albumin as a useful clinical biomarker for glycemic variability in type 1 diabetes assessed by continuous glucose monitoring

This study was conducted to evaluate HbA1c and GA as clinical markers for glycemic variability under normal daily living conditions, not in a hospital setting, in patients with type 1 diabetes by continuous glucose monitoring (CGM). Twenty-one outpatients with type 1 diabetes (age 47.2 ± 16.6 years, with a mean duration of diabetes of 11.1 ± 7.0 years) underwent up to 72 h of the CGM. We analyzed the correlation between ten parameters [mean glucose, area under the curve (AUC) at 180/70 mg/dl, %High/Low, SD, J-index, %CV, high blood glucose index (HBGI)/low blood glucose index (LBGI), M-value, mean amplitude of glycemic excursions (MAGE) and mean of daily differences (MODD)] and the levels of glycemic markers (HbA1c, GA and the GA/HbA1c ratio). The average levels of HbA1c, GA and the GA/HbA1c ratio were 7.9 ± 0.9 %, 25.6 ± 4.4 % and 3.4 ± 0.5, respectively. The Pearson univariate correlation analysis showed that both the levels of HbA1c and GA, but not the GA/HbA1c ratio, were closely related to the CGM-measured mean glucose, AUC at 180, %High, HBGI and the M value. The GA levels also positively correlated with the SD (R = 0.46, P = 0.035), J-index (R = 0.67, P = 3.5 × 10−3) and MAGE (R = 0.45, P = 0.042), while the HbA1c levels did not. In addition, the GA/HbA1c ratio correlated significantly with the glycemic variability, SD (R = 0.58, P = 5.7 × 10−3), J-index (R = 0.45, P = 0.039), M-value (R = 0.46, P = 0.035) and MAGE (R = 0.60, P = 4.3 × 10−3). In conclusion, GA and the GA/HbA1c ratio could therefore be useful clinical markers for the blood glucose level and glycemic variability in patients with type 1 diabetes.