Development of a multivariate model to predict significant coronary artery disease in Thai patients with left ventricular systolic dysfunction and determine the applicability of coronary angiography: a single-center, retrospective, case–control study

Abstract

Background Coronary angiography (CAG) or stress imaging has been performed in almost all Thai patients with left ventricular (LV) systolic dysfunction. If CAG results reveal insignificant coronary stenosis, such patients are diagnosed with nonischemic cardiomyopathy (NICM); however, CAG is considered to provide no benefit and may even harm these patients because it is invasive. Objectives To identify predictors associated with significant coronary artery disease (CAD) (stenosis) in Thai patients with LV systolic dysfunction without angina and without LV regional wall motion abnormality and create a prediction score. Method Retrospective data from patients at a single tertiary-care center with LV systolic dysfunction (LV ejection fraction <50%) diagnosed between August 2000 and October 2014 were separated into a group with ischemic cardiomyopathy (ICM) and a group with NICM according to CAG. Predictors associated with CAD found in normal populations were determined. Multivariate analysis was used to identify predictors associated with significant coronary stenosis in patients with LV systolic dysfunction to develop a model to create a prediction score. Results We included data registered from 240 Thai patients with LV systolic dysfunction. Predictors associated with ICM were age (>60 years), sex (male), and a history of diabetes mellitus (DM). Predictors associated with NICM were body mass index (BMI) >25 kg/m2 and the presence of left bundle branch block (LBBB) on electrocardiography. A simplified equation to predict significant CAD in patients with LV systolic dysfunction is: 3(male sex) + 3(age >60 y) – 5(BMI >25 kg/m2) - 5(LBBB) + 5(DM) - 5. The sensitivity and specificity of this score are 60.5% and 85.1%, respectively. Conclusion Our prediction score has modest sensitivity, but high specificity for predicting significant CAD and can be used to determine who should not undergo CAG.