AIMS: To examine the performance of hemoglobin A1c (HbA1c) in diagnosing prediabetes and whether it was associated with cardiometabolic risk factors in Chinese adolescents without known diabetes. METHODS: This was a cross-sectional study including 933 adolescents (aged 11-16 years; 46.5% girls) with HbA1c of 6.4% or less and fasting plasma glucose (FPG) of 6.9 mmol/L or less. All participants underwent anthropometric and biochemical examinations. The individuals with an increased HbA1c (5.7%-6.4%) were given a 75-g oral glucose tolerance test (OGTT). RESULTS: Of 213 subjects with an increased HbA1c or increased FPG (5.6-6.9 mmol/L), only 4.7% (n = 10) met both criteria. Among the 68 students with increased HbA1c who agreed to undergo the OGTT, only 18 subjects (26.5%) were confirmed to be prediabetic, and the receiver operating characteristic curve for HbA1c to identify prediabetes diagnosed by OGTT had an AUC of 0.53, which was not statistically significant (P = 0.69). The waist circumference, FPG, uric acid, and homeostasis assessment model of insulin resistance were correlated with HbA1c (all P < 0.05). Hemoglobin A1c showed significant association with abdominal obesity and dyslipidemia and the clustering of cardiometabolic risk factors (odds ratios, 1.58-1.90; all P < 0.05). The cutoff point of HbA1c was 5.6% with a maximum accuracy in determining multiple cardiometabolic risk factors (sensitivity, 35.1%; specificity, 72.2%). CONCLUSIONS: There is poor agreement between HbA1c and FPG or OGTT in diagnosing prediabetes, but the former was associated with the presence of multiple cardiometabolic risk factors in Chinese adolescents.
AIMS: To examine the performance of hemoglobin A1c (HbA1c) in diagnosing prediabetes and whether it was associated with cardiometabolic risk factors in Chinese adolescents without known diabetes. METHODS: This was a cross-sectional study including 933 adolescents (aged 11-16 years; 46.5% girls) with HbA1c of 6.4% or less and fasting plasma glucose (FPG) of 6.9 mmol/L or less. All participants underwent anthropometric and biochemical examinations. The individuals with an increased HbA1c (5.7%-6.4%) were given a 75-g oral glucose tolerance test (OGTT). RESULTS: Of 213 subjects with an increased HbA1c or increased FPG (5.6-6.9 mmol/L), only 4.7% (n = 10) met both criteria. Among the 68 students with increased HbA1c who agreed to undergo the OGTT, only 18 subjects (26.5%) were confirmed to be prediabetic, and the receiver operating characteristic curve for HbA1c to identify prediabetes diagnosed by OGTT had an AUC of 0.53, which was not statistically significant (P = 0.69). The waist circumference, FPG, uric acid, and homeostasis assessment model of insulin resistance were correlated with HbA1c (all P < 0.05). Hemoglobin A1c showed significant association with abdominal obesity and dyslipidemia and the clustering of cardiometabolic risk factors (odds ratios, 1.58-1.90; all P < 0.05). The cutoff point of HbA1c was 5.6% with a maximum accuracy in determining multiple cardiometabolic risk factors (sensitivity, 35.1%; specificity, 72.2%). CONCLUSIONS: There is poor agreement between HbA1c and FPG or OGTT in diagnosing prediabetes, but the former was associated with the presence of multiple cardiometabolic risk factors in Chinese adolescents.