Development of a nomogram for predicting the risk of left ventricular diastolic function in subjects with type-2 diabetes mellitus

Abstract

Left ventricular diastolic dysfunction (LVDD) can be affected by many factors, including epicardial adipose tissue (EAT), obesity and type-2 diabetes mellitus (T2DM). The aim of this study was to establish and validate an easy-to-use nomogram that predicts the severity of LVDD in patients with T2DM. This is a retrospective study of 84 consecutive subjects with T2DM admitted to the Endocrinology Department, the First People’s Hospital of Zunyi City between January 2015 and October 2020. Several echocardiographic characteristics were used to diagnose diastolic dysfunction according to the 2016 diastolic dysfunction ASE guidelines. Anthropometric, demographic, and biochemical parameters were collected. Through a least absolute shrinkage and selection operator (LASSO) regression model, we reduced the dimensionality of the data and determined factors for the nomogram. The mean follow-up was 25.97 months. Cases were divided into two groups, those with LVDD (31) and those without (53). LASSO regression identified total cholesterol (Tol.chol), low-density lipoprotein (LDL), right ventricular anterior wall (RVAW) and epicardial adipose tissue (EAT) were identified as predictive factors in the nomogram. The ROC curve analysis demonstrated that the AUC value for most clinical paramerters was higher than 0.6. The nomogram can be used to promote the individualized prediction of LVDD risk in T2DM patients, and help to prioritize patients diagnosed with echocardiography.