A Predictive Model for Diabetic Retinopathy in Patients with Type 2 Diabetes Mellitus

Abstract

Objective: To elaborate a predictive model taking into account the association between glycosylated hemoglobin and albuminuria with the progression of diabetic retinopathy.

Methods: An analytical case-control study was conducted in Centro Oftalmológico, of Santiago de Cuba, in patients with a diagnosis of diabetic retinopathy, in the period from October 2017 to October 2019. The mean, standard deviation, odds ratio, attributable risk in exposed and population were used. Binary logistic regression was performed to evaluate the probability of progression, while the model was performed an internal validity.

Results: In the probability of progression of the studied factors, the coefficients associated with the variables were found to be significant. The OR values were greater than one for a significance level of 0.05. The values of glycosylated hemoglobin greater than 7 % are associated with four times more probability of progression of diabetic retinopathy; with values of microalbuminuria, the probability of progression is 3, 5 %; and, if both factors are present in combination, the risk of progression of diabetic retinopathy is 14.5 times higher.

Conclusions: Glycosylated hemoglobin levels above 7 % and microalbuminuria are predictors of progression of diabetic retinopathy, whose combined occurrence increases the risk in patients with type 2 diabetes mellitus.