A LASSO-derived developing and validating risk model for chronic total occlusion in Asian population before coronary angiography.

Abstract

Despite several previous studies that have explored the predictors of high morbidity in coronary artery disease (CAD) and developed nomograms for CAD patients prior to coronary angiography (CAG), there is a lack of models available to predict chronic total occlusion (CTO). The aim of this study is to develop a risk model and a nomogram for predicting the probability of CTO prior to CAG. The study included 1,105 patients with CAG-diagnosed CTO in the derivation cohort and 368 patients in the validation cohort. Clinical demographics, echocardiography results, and laboratory indexes were analyzed using statistical difference tests. Independent risk factors affecting the CTO indication were selected using least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression analysis. A nomogram was built and validated based on these independent indicators. The performance of the nomogram was evaluated using area under the curve (AUC), calibration curve, and decision curve analysis (DCA). LASSO and multivariate logistic regression analysis revealed that 6 variables, including sex (male), lymphocyte percentage (LYM%), ejection fraction (EF), myoglobin (Mb), non-high-density lipoprotein cholesterol (non-HDL), and N-terminal pro-B-type natriuretic peptide (NT-proBNP), were independent predictors of CTO. The nomogram constructed based on these variables showed good discrimination (C index of 0.744) and external validation (C index of 0.729). The calibration curves and DCA demonstrated high reliability and precision for this clinical prediction model. The nomogram based on sex (male), LYM%, EF, Mb, non-HDL, and NT-proBNP could be used to predict CTO in CAD patients, enhancing the ability to predict their prognosis in clinical practice. Further research is needed to validate the efficacy of the nomogram in other populations.