Purpose. Exercise ECG is commonly used in patients with suspected coronary artery disease (CAD) to select candidates for stress imaging testing or coronary angiography. The diagnostic accuracy of exercise ECG can be improved by integrating clinical variables into predictive models of CAD. We hypothesized that the diagnostic accuracy could be further improved by including bio-humoral markers in predictive models based on clinical data and exercise ECG. Materials and Methods. A cohort of 428 patients with angina-like chest pain or equivalent symptoms, enrolled in the EVINCI (EValuation of INtegrated Cardiac Imaging) study, underwent clinical examination, exercise ECG test, bio-humoral characterization and coronary arteriography. In every patient 36 biomarkers linked to the atherosclerotic process were evaluated, and the integrated Duke clinical score (which incorporates exercise ECG results) was calculated. Independent markers of CAD were identified by logistic regression analysis. Severe obstructive CAD, i.e. >70% coronary stenosis in at least one vessel at coronary arteriography, or 30-70% stenosis with reduced fractional flow reserve, was the diagnostic endpoint. The diagnostic accuracy of the integrated models was evaluated by ROC curves analysis and differences in accuracy were assessed using Delong method. Results. The actual prevalence of severe obstructive CAD was 27%. The integrated clinical model showed a predictive accuracy of 66.7%. At multivariate analysis, High-Density Lipoprotein cholesterol, Aspartate Transaminase, Homeostasis Model Assessment Index, Interleukin-6, and Osteopontin were independent predictors of CAD. Adding these variables to the clinical model, the diagnostic accuracy increased up to 74.0% (p = 0.001). Conclusions. Five bio-humoral markers linked to the atherosclerotic process are independently associated with obstructive CAD. Adding these markers into predictive models significantly improves the diagnostic accuracy of clinical evaluation and exercise ECG for obstructive CAD.
Improving diagnostic yield of exercise ECG for obstructive coronary artery disease: integration with clinical and bio-humoral markers.
Chiara Caselli;Martina Marinelli;Daniela Giannessi;Daniele Rovai
2013
Abstract
Purpose. Exercise ECG is commonly used in patients with suspected coronary artery disease (CAD) to select candidates for stress imaging testing or coronary angiography. The diagnostic accuracy of exercise ECG can be improved by integrating clinical variables into predictive models of CAD. We hypothesized that the diagnostic accuracy could be further improved by including bio-humoral markers in predictive models based on clinical data and exercise ECG. Materials and Methods. A cohort of 428 patients with angina-like chest pain or equivalent symptoms, enrolled in the EVINCI (EValuation of INtegrated Cardiac Imaging) study, underwent clinical examination, exercise ECG test, bio-humoral characterization and coronary arteriography. In every patient 36 biomarkers linked to the atherosclerotic process were evaluated, and the integrated Duke clinical score (which incorporates exercise ECG results) was calculated. Independent markers of CAD were identified by logistic regression analysis. Severe obstructive CAD, i.e. >70% coronary stenosis in at least one vessel at coronary arteriography, or 30-70% stenosis with reduced fractional flow reserve, was the diagnostic endpoint. The diagnostic accuracy of the integrated models was evaluated by ROC curves analysis and differences in accuracy were assessed using Delong method. Results. The actual prevalence of severe obstructive CAD was 27%. The integrated clinical model showed a predictive accuracy of 66.7%. At multivariate analysis, High-Density Lipoprotein cholesterol, Aspartate Transaminase, Homeostasis Model Assessment Index, Interleukin-6, and Osteopontin were independent predictors of CAD. Adding these variables to the clinical model, the diagnostic accuracy increased up to 74.0% (p = 0.001). Conclusions. Five bio-humoral markers linked to the atherosclerotic process are independently associated with obstructive CAD. Adding these markers into predictive models significantly improves the diagnostic accuracy of clinical evaluation and exercise ECG for obstructive CAD.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
