Optimal Cutoff of Hemoglobin A1c for Diagnosing Diabetes and Prediabetes in Obese Children and Adolescents

Presentation Number: OR30-5
Date of Presentation: April 4th, 2017

Hyo-Kyoung Nam*1, Won-Kyoung Cho2, Jae Hyun Kim3, Young-Jun Rhie4, Sochung Chung5, Kee-Hyoung Lee1 and Byung-Kyu Suh2
1Korea University College of Medicine, Seoul, Korea, Republic of (South), 2College of Medicine, The Catholic University of Korea, Seoul, Korea, Republic of (South), 3Inje University College of Medicine, Ilsan Paik Hospital, Goyang-si, Korea, Republic of (South), 4Korea University College of Medicine, Ansan City, Korea, Republic of (South), 5Konkuk University Medical Center, Konkuk University School of Medicine, Seoul, Korea, Republic of (South)


Introduction: Traditionally plasma glucose level during oral glucose tolerance test (OGTT) has been used as a tool for the diagnosis of diabetes and prediabetes. Hemoglobin A1c (HbA1c) has been recently recommended as an alternative diagnostic method in adults. However, application of HbA1c for diagnosing diabetes in children and adolescents were controversial. Therefore, we evaluated performance of HbA1c diagnosing diabetes and prediabetes and determined the optimal HbA1c cutoff values for detecting these diseases.

Methods: Data were obtained from retrospective review of medical records of the subjects who had undergone OGTT and HbA1c simultaneously between January 2010 and June 2016 in 6 general hospitals. Subjects were categorized according to plasma glucose level: normoglycemia with fasting blood glucose (FPG) < 100 mg/dL and post 2-hr plasma glucose (PP2) < 140 mg/dL; prediabetes with FPG between 100 and 125 mg/dL or PP2 140 and 199 mg/dL; diabetes FPG ≥126 mg/dL or PP2 ≥ 200 mg/dL. Demographic, anthropometric and biochemical parameters were compared among groups. HbA1c performance for diagnosing prediabetes and diabetes was evaluated via receiver operating characteristic (ROC) analysis.

Results: A total of 389 subjects (boys 217, 55.8%) were enrolled. Normoglycemia was 197 (50.6%), prediabetes 121 (31.1%) and diabetes 71 (18.3%). Mean FPG, PP2 and HbA1c were 109.7±45.9 mg/dL, 162.8±93.9 mg/dL and 6.3±2.1%, respectively. Z-score of body mass index was 2.3±0.6 for normoglycemia, 2.2±0.7 for prediabetes and 2.0±0.5 for diabetes, respectively (p<0.001). The kappa coefficients for agreement between the OGTT and HbA1c results were 0.46 [95% confidence interval (CI) 0.42-0.53]. In the ROC analysis, the optimal HbA1c cutoff points for prediabetes and diabetes were 5.8% (AUC 0.795, 95% CI 0.750-0.840) and 6.2% (AUC 0.972, 95% CI 0.949-0.995), respectively. The sensitivity and specificity of each cutoff values were 64.1% and 83.8% for prediabetes and 91.5% and 93.7% for diabetes, respectively.

Conclusions: The optimal cutoff of HbA1c for diagnosing diabetes and prediabetes were 5.8% and 6.2%, respectively. Usefulness of adult criteria of HbA1c for diagnosis of prediabetes and diabetes in children and adolescents remains to be clarified due to disparities between the results of OGTT- and HbA1c-based tests.


Nothing to Disclose: HKN, WKC, JHK, YJR, SC, KHL, BKS