No consistent glycaemic threshold for incident retinopathy across different populations
Document Actions
The assumption of the presence of a glycaemic threshold defining diabetes (e.g. fasting plasma glucose (FPG) >7.0 mmol/L) with high sensitivity for identifying retinopathy is paradoxically based on three previous studies that had important limitations in detecting retinopathy. This study was aimed to assess the diagnostic accuracy of current FPG thresholds in identifying both prevalent and incident retinopathy, and to provide updated data for the relation between FPG and retinopathy. To this end, the data of three cross-sectional adult populations were examined, including 3,162 subjects from the Blue Mountains Eye Study (BMES), 2,182 subjects from the Australian Diabetes, Obesity and Lifestyle study (AusDiab), and 6,079 from the Multi-Ethnic Study of Atherosclerosis (MESA). The overall prevalence of retinopathy was 11.5% in BMES, 9.6% in AusDiab, and 15.8% in MESA. There was inconsistent evidence for a uniform glycaemic threshold of prevalent and incident retinopathy, with analyses rather suggesting a continuous relation. The current FPG cutoff of 7.0 mmol/L had a sensitivity <40% for detecting retinopathy, with a specificity ranging from 80.8% to 95.8%. Based on these findings, the authors conclude that the widely used diabetes cutoff of 7.0 mmol/L did not accurately identify people with and without retinopathy. Consequently, the criteria used for diagnosing diabetes in the fasting state could need reassessment as they rely on retinopathy risk thresholds.
Abstract
