Abstract
Background: A non-invasive predictive model has not been established to identify the severity of coronary lesions in young adults with acute coronary syndrome (ACS). Methods: In this retrospective study, 1088 young adults (≤45 years of age) first diagnosed with ACS who underwent coronary angiography were enrolled and randomized 7:3 into training or testing datasets. To build the nomogram, we determined optimal predictors of coronary lesion severity with the Least Absolute Shrinkage and Selection Operator and Random Forest algorithm. The predictive accuracy of the nomogram was assessed with calibration plots, and performance was assessed with the receiver operating characteristic curve, decision curve analysis and the clinical impact curve. Results: Seven predictors were identified and integrated into the nomogram: age, hypertension, diabetes, body mass index, low-density lipoprotein cholesterol, mean platelet volume and C-reactive protein. Receiver operating characteristic analyses demonstrated the nomogram’s good discriminatory performance in predicting severe coronary artery disease in young patients with ACS in the training (area under the curve 0.683, 95% confidence interval [0.645–0.721]) and testing (area under the curve 0.670, 95% confidence interval [0.611–0.729]) datasets. The nomogram was also well-calibrated in both the training (P=0.961) and testing (P=0.302) datasets. Decision curve analysis and the clinical impact curve indicated the model’s good clinical utility. Conclusion: A simple and practical nomogram for predicting coronary artery disease severity in young adults≤45 years of age with ACS was established and validated.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have