Abstract
INTRODUCTION: The relationship between HbA1c and diabetic retinopathy is expected to differ between different races. This study was designed to verify whether HbA1c or fasting plasma glucose (FPG) is more effective in detecting diabetic retinopathy in a Japanese population. MATERIALS AND METHODS: The study subjects underwent health examinations between 2008 and 2009 with fasting. Of these participants, we analyzed the data for 2,921 Japanese men who had undergone an ophthalmologic examination. Retinopathy was classified into 7 categories according to a simplified diabetic retinopathy scale. The odds ratios of retinopathy according to the eight groups of FPG and HbA1c were estimated using multiple logistic regression analysis adjusted for age. Receiver operator characteristic analysis was used to evaluate each value associated with the presence or absence of retinopathy. Results and Discussion. The odds ratios (95% CI) of retinopathy for HbA1c level categories, in ascending order, were 1.0 (ref.), 0.88 (0.28-2.75), 1.27 (0.44-3.69), 1.52 (0.48-4.79), 1.89 (0.52-6.85), 2.70 (0.66-11.10), 4.10 (0.80-21.00), and 6.34 (2.37-16.97) where the odds ratios significantly increased with HbA1c ≥ 6.8%. The area under the curve (SE) for FPG and HbA1c was almost the same, at 0.668 (0.043) and 0.680 (0.043), respectively. CONCLUSIONS: It was clarified that the higher the level of HbA1c, the higher the prevalence of retinopathy, and there was no clear threshold. The detection ability of retinopathy was almost the same, suggesting that it is possible to detect the risk of retinopathy by HbA1c only.