Prediction of coronary artery disease risk in patients with type 2 diabetes mellitus: a mathematical model

Coronary artery disease (CAD) is one of the most prevalent and life-threatening complications in individuals with type 2 diabetes mellitus (T2DM). The aim of this study was to develop a mathematical model for predicting the risk of CAD in patients with T2DM. A total of 242 patients with T2DM, aged 30-80 years, were examined. The following parameters were analyzed: patient age, age at T2DM onset, disease duration, fasting glucose levels, glycated hemoglobin levels, lipid profile parameters, blood pressure, presence of diabetic complications, lifestyle factors, family history, and parental exposure to famine in 1932-1933. The predictive mathematical model for CAD development in T2DM patients was constructed using receiver operating characteristic (ROC) analysis and multiple logistic regression. ROC analysis identified the prognostic value of each of the eight key independent variables, which do not depend on the patient's current health status and can be considered independent at the time of CAD diagnosis. In the second phase of the study, a formula for calculating CAD probability was developed, incorporating the most informative variables with predictive significance. These included the patient’s age, T2DM duration, presence of chronic kidney disease, paternal history of T2DM, maternal famine exposure during pregnancy, rural residence, and patient sex. The developed formula was used to predict CAD risk, and its sensitivity, specificity, and diagnostic performance were evaluated. The model demonstrated high predictive accuracy (AUC=0.792 [0.734–0.842], chi-square =65.1; p<0.001). The probability of CAD development was determined with an accuracy of 72.3%, and the model’s predictive efficiency was 73.6%. The obtained results allowed us to establish statistically significant associations between the studied risk factors and CAD development in T2DM patients. Based on these findings, we have developed a mathematical model for predicting CAD risk in T2DM patients, which can be implemented in clinical practice.