Objective: Continuous glucose monitoring system (CGMS) is emerging as an important tool in management of diabetes. Estimated glycated hemoglobin (eA1c), a CGM-derived metric, can be widely used if it is highly matched with the measured glycated hemoglobin (A1c). This study aimed to evaluate the relationship between eA1c and A1c measured by high performance liquid chromatography.

Methods: Data were collected from T2D patients between July 2017 and December 2019. Patients with abnormal hemoglobin levels and acute diabetic complications were excluded. Enrolled patients were divided into 3 groups based on the difference of eA1c and A1c (ΔA1c=eA1c-A1c): matched group (ΔA1c from -0.099% to 0.099%), overestimated group (ΔA1c ≥ +0.1%), and underestimated group (ΔA1c ≤ -0.1%). The indices of glycemic variability (GV) were obtained from CGMS.

Results: The 102 enrolled patients were 55.9(12.1) [mean (SD)] years old, with the measured A1c of 6.89(1.25)% and eA1c of 6.20(0.65)%. A positive correlation between eA1c and measured A1c (r= 0.50, p<0.001) was found. Majority of the patients were allocated in the underestimated group (13.7%, 19.6% and 66.7% of the patients in the matched, overestimated and underestimated groups, respectively). ΔA1c was -0.027(0.06)%, 0.59(0.39)% and -1.20(1.03)% (p<0.001) in these 3 groups, respectively. Age, BMI, duration of disease, type of hypoglycemic therapy, duration and intensity of hypoglycemia, blood pressure, lipid profiles, and the indices of intra-day GV among the 3 groups were similar. Multivariate logistic regression analysis revealed that gender (β=7.99, p=0.01), hematocrit (β=4.28, p=0.01) and mean of daily differences (MODD) (β=-7.44, p=0.02) were independently related with ΔA1c.

Conclusions: eA1c was generally underestimated compared to measured A1c in T2D patients. Gender, hematocrit and MODD were identified as significant factors accounting for this discrepancy.


B. Lin: None. W. Xu: None. Z. Liu: None. D. Yang: None. Y. Su: None. Y. Chen: None. J. Yan: None. B. Yao: None.


National Key Research and Development Program of China (2016YFC1304801)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.