Comparison of diagnostic methods for diabetes mellitus based on prevalence of retinopathy in a Japanese population: The Hisayama study

Aims/hypothesis. The aims of this study were to compare the ability of tests measuring fasting plasma glucose, 2-h plasma glucose and HbA₁c levels in predicting specific diabetic retinopathy, and to determine the cut-off level of each measurement for diagnosing diabetes in a Japanese population. Methods. In a total of 1637 subjects, fasting plasma glucose, 2-h plasma glucose and HbA₁c levels were measured in a 75-g oral glucose tolerance test, and diabetic retinopathy was assessed by ophthalmic examination. We calculated receiver operating characteristic (ROC) curves as well as the prevalence of diabetic retinopathy by deciles of the distribution of these glycaemic measurements. Results. Of the subjects, 37 (2.3%) had diabetic retinopathy. The prevalence of retinopathy dramatically increased in the tenth decile of each variable. Analysis with ROC curves showed that the optimal cut-off levels for diagnosis of diabetes were 6.4 mmol/l for fasting plasma glucose, 11.1 mmol/l for 2-h plasma glucose, and 5.7% for HbA₁c. The sensitivities for the cut-off point of the three measurements were identical (86.5%), and the specificities were similar (fasting plasma glucose 87.3%; 2-h plasma glucose 89.6%; HbA₁c 90.1%). The area under the ROC curve for 2-h plasma glucose (96.1%) was slightly but not significantly larger than that for fasting plasma glucose (90.0%) and that for HbA₁c (94.5%). Conclusions/interpretation. Our findings suggest that measuring fasting plasma glucose or HbA₁c is just as useful as measuring 2-h plasma glucose for the diagnosis of diabetes, and that the cut-off point for diagnostic fasting plasma glucose level is lower than that of the current diagnostic criteria.