Can HbA1c replace OGTT for the diagnosis of diabetes mellitus among Chinese patients with impaired fasting glucose?

Background. HbA1c ≥ 6.5% has been recommended as a diagnostic criterion for the detection of diabetes mellitus (DM) since 2010 because of its convenience, stability and significant correlation with diabetic complications. Nevertheless, the accuracy of HbA1c compared to glucose-based diagnostic criteria varies among subjects of different ethnicity and risk profile. Objectives. This study aimed to evaluate the accuracy of HbA1c for diagnosing DM compared to the diagnosis by oral glucose tolerance test (OGTT) and the optimal HbA1c level to diagnose DM in primary care Chinese patients with impaired fasting glucose (IFG). Methods. A cross-sectional study was carried out in three public primary care clinics in Hong Kong. About 1128 Chinese adults with IFG (i.e. FG level between 5.6 and 6.9 mmol/l in the past 18 months) were recruited to receive paired OGTT and HbA1c tests. Sensitivities and specificities of HbA1c at different threshold levels for predicting DM compared to the diagnosis by OGTT were evaluated. A receiver operating characteristic (ROC) curve was used to determine the optimal cut-off level. Results. Among the 1128 subjects (mean age 64.2±8.9 year, 48.8% male), 229 (20.3%) were diagnosed to have DM by OGTT. The sensitivity and specificity of HbA1c ≥6.5% were 33.2% and 93.5%, respectively, for predicting DM diagnosed by OGTT. The area under the ROC curve was 0.770, indicating HbA1c had fair discriminatory power. The optimal cut-off threshold of HbA1c was 6.3% for discriminating DM from non-DM, with sensitivity and specificity of 56.3% and 85.5%, respectively. HbA1c ≥ 5.6% has the highest sensitivity and negative predictive value of 96.1% and 94.5%, respectively. Conclusions. HbA1c ≥ 6.5% is highly specific in identifying people with DM, but it may miss the majority (66.8%) of the DM cases. An HbA1c threshold of