In addition, oral inflammatory burden with biomarkers described in this publication may also be combined with other biomarkers either circulating or from the saliva that can be implemented to generate predictive models to identify subjects in the early stages of either developing autoimmunity or glucose intolerance. To the best of our knowledge, this is the first large population based study to examine the association between DR and CKD and its components among a representative Korean histological experimental diabetic population. In this cross-sectional study, conducted as part of he 2008�C2010 KNHANES, CKD and its components, including proteinuria, were positively associated with the risk of DR and VTDR. In this study, we found that the prevalence of CKD in this diabetic population was 10.62%. Recent studies show a worldwide increase in CKD. The prevalence of CKD in Korean adults seems to be lower than in adult Western and East Asian populations. However, because the duration and control of diseases have significant differences among these studies. Simple and direct comparison has some limitations. Another Korean national data set shows that the prevalence of CKD in KNHANES IV was 4.5% in men and 6.3% in women. However, because DM is one of the most important risk factors, the prevalence of CKD in the DM population seems to be higher, as observed in our data. Subjects with CKD were more likely to be older. Regular exercise and higher education were lower, whereas obesity, lower HDL, higher blood pressure, and longer DM duration were more frequent in participants with DM. Recently, one national cross-sectional study that performed in Spanish primary care centers reported that the prevalence was 18.0% for eGFR<60 ml/min/1.73 m2 and 15.4% for albuminuria. A study performed in Japan reported that 25.2% of type 2 DM patients had a GFR below 60 ml/min/1.73 m2. Different prevalence rates of CKD among these studies may due to the different methodology applied and the racial differences among patients. In addition, the decreased prevalence of CKD in Korean adults in a recent survey can be partly explained by the combined effect of not only active management of chronic diseases such as hypertension, diabetes and dyslipidemia, but also the increased number of subjects who were regularly engaged in exercise. However, we defined CKD using a single measurement of the estimated GFR based on the MDRD equation and of proteinuria by a dipstick test. In the present study, the prevalence of DR in this diabetic population was 11.39%. The prevalence of VTDR in this population was 2.20%. Previously reported studies have shown considerable variability in DR prevalence rates among individuals with DM, with rates ranging from 17.6% to 33.2%. Differences in study methodologies, population characteristics, and ascertainment and classification of DR have made direct comparisons between studies difficult. In addition, they concluded that the prevalence estimates of any DR and VTDR were highest in African Americans and lowest in Asians as in our data. Differences in the prevalence of DR among ethnic groups have also been reported. These findings were consistent with our data, but their study was hospital-based cohort study and had a limitation of only using the fundoscopic findings instead of the reference method for DR diagnosis. Our study should be interpreted with consideration of the following limitations.