Glycaemic and haemoglobin A1c thresholds for detecting diabetic retinopathy: the fifth Korea National Health and Nutrition Examination Survey (2011).

AIMS Few representative population-based data are available regarding glycaemic and HbA1c thresholds for detecting diabetic retinopathy (DR) in Asia. We investigated the association between DR and fasting plasma glucose (FPG) and HbA1c levels among Korean adults. METHODS Using data from the Korea National Health and Nutrition Examination Survey (2011), a total of 5212 adults (≥19 years old) were analysed. When participants had diabetes mellitus and/or a suspicion of DR in two-field nonmydriatic fundus photography, seven standard photographs were obtained after pupil dilatation (75.9% of men, 75.0% of women among the subjects). DR was defined as the presence of ≥1 retinal microaneurysms or blot haemorrhages with or without more severe lesions. Receiver operating characteristic (ROC) curves were used to determine the optimal cut-off value for HbA1c or FPG. RESULTS The overall glycaemic thresholds for DR were 6.3mmol/l for FPG and 6.2% (44mmol/mol) for HbA1c. The optimal thresholds did not differ by age group. The sensitivities and specificities were 82.6% and 91.2% for FPG and 93.9% and 89.7% for HbA1c, respectively. The diagnostic discrimination was better for HbA1c than FPG for DR-area under curve: 0.908 for FPG and 0.953 for HbA1c (p=0.007). After being controlled for other covariates, the odds ratio for the risk of DR increased significantly in a consistent way from 6.2% (44mmol/mol) for HbA1c and 6.3mmol/l for FPG. CONCLUSIONS According to these nationally representative data, the current diabetes diagnostic values for FPG and HbA1c based on DR may be lower for the Korean population.