Iranian Risk Model as a Predictive Tool for Retinopathy in Patients with Type 2 Diabetes

• Published in: Canadian journal of diabetes 2015-10, Vol.39 (5), p.358-363
• Main Authors: Azizi-Soleiman, Fatemeh, PhD, Heidari-Beni, Motahar, PhD, Ambler, Gareth, PhD, Omar, Rumana, PhD, Amini, Masoud, MD, Hosseini, Sayed-Mohsen, PhD
• Format: Article
• Language: English
• Subjects: Adult; Blood Glucose; Blood Pressure; Body Mass Index; Cross-Sectional Studies; Diabetes Mellitus, Type 2 - complications; diabetic retinopathy; Diabetic Retinopathy - diagnosis; diabète de type 2; Educational Status; Endocrinology & Metabolism; facteurs de risque; Female; Glycated Hemoglobin A - metabolism; Humans; Iran; Logistic Models; Male; Middle Aged; Odds Ratio; Other; Proteinuria - complications; Risk Assessment; risk factors; ROC Curve; rétinopathie diabétique; type 2 diabetes
• ISSN: 1499-2671; 2352-3840
• DOI: 10.1016/j.jcjd.2015.01.290

Abstract Objective Diabetic retinopathy (DR) is the leading cause of blindness in patients with type 1 or type 2 diabetes. The gold standard for the detection of DR requires expensive equipment. This study was undertaken to develop a simple and practical scoring system to predict the probability of DR. Methods A total of 1782 patients who had first-degree relatives with type II diabetes were selected. Eye examinations were performed by an expert ophthalmologist. Biochemical and anthropometric predictors of DR were measured. Logistic regression was used to develop a statistical model that can be used to predict DR. Goodness of fit was examined using the Hosmer-Lemeshow test and the area under the receiver operating characteristic (ROC) curve. Results The risk model demonstrated good calibration and discrimination (ROC area=0.76) in the validation sample. Factors associated with DR in our model were duration of diabetes (odds ratio [OR]=2.14, confidence interval [CI] 95%=1.87 to 2.45); glycated hemoglobin (A1C) (OR=1.21, CI 95%=1.13 to 1.30); fasting plasma glucose (OR=1.83, CI 95%=1.28 to 2.62); systolic blood pressure (OR=1.01, CI 95%= 1.00 to 1.02); and proteinuria (OR=1.37, CI 95%=1.01 to 1.85). The only factor that had a protective effect against DR were body mass index and education level (OR=0.95, CI 95%=0.92 to 0.98). Conclusions The good performance of our risk model suggests that it may be a useful risk-prediction tool for DR. It consisted of the positive predictors like A1C, diabetes duration, sex (male), fasting plasma glucose, systolic blood pressure and proteinuria, as well as negative risk factors like body mass index and education level.