Validity of Coronary Artery Disease Consortium Models for Predicting Obstructive Coronary Artery Disease & Cardiovascular Events in Patients with Acute Chest Pain Considered for Coronary Computed Tomographic Angiography

G Teressa
M Zhang
P Lavenburg
G Cantor
P Noack
J Yang
M Goyfman, Zucker School of Medicine at Hofstra/Northwell
E J. Feldmann
J Butler
M Poon, Zucker School of Medicine at Hofstra/Northwell

Abstract

© 2018 Although the majority of acute chest pain patients are diagnosed with noncardiac chest pain after noninvasive testing, identifying these low-risk patients before testing is challenging. The objective of this study was to validate the coronary artery disease (CAD) consortium models for predicting obstructive CAD and 30-day major adverse cardiovascular events (MACE) in acute chest pain patients considered for coronary computed tomography angiogram, as well as to determine the pretest probability threshold that identifies low-risk patients with <1% MACE. We studied 1,981 patients with no known CAD and negative initial troponin and electrocardiogram. We evaluated CAD consortium models (basic: age, sex, and chest pain type; clinical: basic + diabetes, hypertension, dyslipidemia, and smoking; and clinical + coronary calcium score [CAC] models) for prediction of obstructive CAD (≥50% stenosis on coronary CT angiogram) and 30-day MACE (Acute Myocardial Infarction, revascularization, and mortality). The C-statistic for predicting obstructive CAD was 0.77 (95% confidence interval [CI] 0.73 to 0.77) for the basic, 0.80 (95% CI 0.77 to 0.80) for the clinical, and 0.88 (95% CI 0.85 to 0.88) for the clinical + CAC models. The C-statistic for predicting 30-day MACE was 0.82 (95% CI 0.77 to 0.87) for the basic, 0.84 (95% CI 0.79 to 0.88) for the clinical, and 0.87 (95% CI 0.83 to 0.91) for the clinical + CAC models. In 47.3% of patients for whom the clinical model predicted ≤5% probability for obstructive CAD, the observed 30-day MACE was 0.53% (95% CI 0.07% to 0.999%); in the 66.9% of patients for whom the clinical + CAC model predicted ≤5% probability, the 30-day MACE was 0.75% (95% CI 0.29% to 1.22%). We propose a chest pain evaluation algorithm based on these models that classify 63.3% of patients as low risk with 0.56% (95% CI 0.15% to 0.97%) 30-day MACE. In conclusion, CAD consortium models have excellent diagnostic and prognostic value for acute chest pain patients and can safely identify a significant proportion of low-risk patients by achieving <1% missed 30-day MACE.