Indices of glycemic variability as the basis for building a prognostic model for the development of diabetic complications

Abstract

The use of mathematical indices of glycemic variability (IGV) opens up new possibilities in predicting diabetic complications, which allows more accurate correction of therapy and prevention of the development of acute and chronic conditions. Aim. To analyze the predictive capabilities of glycemic variability indices to assess the development and progression of diabetic complications Material and methods. The study included 307 patients with type 1 diabetes mellitus (DM1). In all patients, glycated hemoglobin (HbA1c) was assessed, as well as the main indicators of glycemic control and IGV, which were selected to predict the formation and progression of diabetic complications. Statistical analysis was carried out using the SPSS 23.0 program. To build a model for predicting diabetic complications, the logistic regression method was used. Results. During the work, there was a decrease in HbA1c from 9.0 to 8.0% (p 0.005), as well as a change in the main parameters of carbohydrate metabolism and IGV. A predictive model for diabetic complications was built on the basis of HbA1c, mean glycemia and IGV – SD, CONGA, LI, LBGI, HBGI, MODD, MAGE, ADDR, MAG at the first study visit. The predictive model for the development of diabetic complications was considered significant at p 0.05. The resulting model showed a high sensitivity – 92% and a sufficient specificity of 85%. Not all parameters turned out to be statistically significant, however, with the exclusion of some, the sensitivity and specificity of the model decrease, which indicates the importance of each of the IGVs in predicting diabetic complications. Concludion. By analyzing the IGV, and not just the standard methods for assessing carbohydrate metabolism, the doctor can more accurately judge the compensation for diabetes and give the patient individual recommendations for treatment. Evaluation of GV, in particular its mathematical indices, play a significant role in predicting the development and progression of diabetic complications in patients with DM1 in childhood and adolescence.