|
KMID : 1101420170490020114
|
|
Korean Journal of Clinical Laboratory Science
2017 Volume.49 No. 2 p.114 ~ p.120
|
|
|
The Cutoff Value of HbA1c in Predicting Diabetes and Impaired Fasting Glucose
|
|
Kwon Se-Young
Ra Young-Ak
|
|
|
Abstract
|
|
|
|
There have been many studies to develop methods for predicting diabetes and to prevent diabetes. The validity of glycated hemoglobin (HbA1c), one of the commonly known tools in predicting diabetes, has been verified by many previous studies. In this study, we examined the cutoff value of HbA1c for diabetes and impaired fasting glucose (IFG). Based on this study, we proposed a proper clinical guideline and evaluated the validation of the guideline. Excluding those without blood glucose and HbA1c data, we used the data of 5,161 subjects (2,281 men and 2,880 women) over the age of 20 years from the 2015 Korean National Health and Nutrition Examination Survey. The correlation efficient of fasting plasma glucose (FPG) and HbA1c was 0.79, indicating a strong relationship. Howeve, the correlation efficient of FPG and HbA1c was low, showing 0.27 in non-diabetes, 0.39 in IFG, and 0.66 in diabetes, showing a strong relationship. The cutoff value of HbA1c for predicting diabetes using ROC curve was 6.05% (sensitivity 84.6%, and specificity 92.0%), and AUC was 0.941 (0.937 in men, and 0.946 in women). The cutoff value of HbA1c for predicting IFG using ROC curve was 5.55% (sensitivity 64.5%, and specificity 70.0%), and AUC was 0.733 (0.708 in men, and 0.764 in women). Therefore, it may not be appropriate to apply the guidelines for diagnosing IFG since sensitivity and specificity were below 70%. For future studies retarding the cutoff value of HbA1c in predicting IFG, high sensitivity and specificity are expected if we segment the reference range of IFG.
|
|
|
KEYWORD
|
|
|
Cutoff value, HbA1c, Diabetes, Impaired fasting glucose
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|