Serum glycated albumin predicts the progression of diabetic retinopathy--a five year retrospective longitudinal study.
Mots clés
Abstrait
OBJECTIVE
To assess the predictive value of glycated albumin (GA) and other risk factors on a progression of diabetic retinopathy (DR).
METHODS
In this retrospective longitudinal study, we enrolled the subjects with type 2 diabetes who had undergone fundus photography twice with a 5-years gap between January 2006 and December 2012, and had been measured consecutively for hemoglobin A1c (HbA1c) and GA levels every 3 or 6 months. The subjects were divided into two groups with or without a progression of DR. The mean HbA1c and mean GA were calculated separately by the sum of all measured values divided by the numbers of values throughout the study period.
RESULTS
Of the 359 subjects, progression group showed significantly higher diabetes duration (8.41±5.72 vs. 6.46±5.77, P<0.01), baseline HbA1c (9.13±2.71 vs. 8.41±2.32, P<0.05), fasting plasma glucose (8.71±2.78 vs. 7.94±2.63, P<0.05), 2h-postprandial glucose (15.12±11.20 vs.13.14±4.72, P<0.05), eGFR (114.81±39.15 vs. 103.23±32.18, P<0.01), mean HbA1c (8.32±1.69 vs. 7.39±1.35, P<0.01) and mean GA (22.66±5.92 vs. 19.83±5.18, P<0.01) than non-progression group. The frequencies of subjects with DR progression increased obviously with the increment of baseline HbA1c, mean HbA1c and mean GA according to quartile stratification of the above three glucose parameters. Multivariable binary logistic regression analysis investigated that the factors affected the DR progression were the presence of DR at baseline (OR=0.391, P=0.005), mean HbA1c (OR=1.389, P=0.021), mean GA (OR=1.087, P=0.039) and eGFR (OR=1.008, P=0.045). The optimal cut-off values of mean HbA1c and GA to predict DR progression were 7.27% and 21.85%, respectively.
CONCLUSIONS
The presence of DR at baseline, poor glycemic control, glycated albumin, and impaired renal function predicted DR progression in patients with type 2 diabetes.
