Performance of the Traditional Age, Sex, and Angina Typicality-Based Approach for Estimating Pretest Probability of Angiographically Significant Coronary Artery Disease in Patients Undergoing Coronary Computed Tomographic Angiography: Results From the Multinational Coronary CT Angiography Evaluation for Clinical Outcomes: An International Multicenter Registry (CONFIRM)

Study Questions:

How does the presence and extent of obstructive coronary artery disease (CAD), as detected by computed tomographic angiography (CTA), compare to currently used algorithms for predicting the prevalence of obstructive CAD?


CTA was performed in 14,048 consecutive patients with suspected CAD (no patient had known CAD). Chest pain was classified as typical angina, atypical angina, or nonangina. Traditional risk factors of age, gender, hypertension, diabetes, etc., were also tabulated. CTA was analyzed on the 16-segment American Heart Association (AHA) coronary model and stratified by diameter stenosis >50% (CAD50) and >70% (CAD70). Patients were stratified into four age groups (30-39, 40-49, 50-59, and ≥60). Patients presenting with isolated dyspnea without pain were tabulated separately. The expected probability of suspected CAD was tabulated from the current American College of Cardiology/AHA Clinical Practice Guidelines for the Management of Patients With Chronic Stable Angina.


Of the 7,719 males in the population, 3,114 were asymptomatic, 582 had nonanginal pain, 2,612 had atypical angina, 805 had typical angina, and 606 presented with dyspnea. Among the 6,329 females, 1,577 were asymptomatic, 671 had nonanginal pain, 2,611 had atypical angina, 825 had typical angina, and 645 dyspnea only. Among males, the overall prevalence of CAD50 was 23%, 21%, 25%, 19%, 40%, and 29% for the five symptomatic groups referenced above. For CAD70, the overall CAD prevalence was 12% with prevalence for the symptom groups of 10%, 17%, 9%, 27%, and 16%. Among women, CAD50 was present in 13% overall and in 13%, 12%, 11%, 19%, and 13% in the symptom groups, and CAD70 was present in 6% of the overall population including 6%, 7%, 5%, 11%, and 6% for the symptom groups. Typical angina was associated with the highest prevalence of CAD50 (40% in men, 19% in women) and CAD70 (27% in men, 11% in women) compared to other symptom categories (p < 0.001). ACC/AHA guidelines predicted a 51% prevalence of CAD50 for the overall group compared to the actual observed prevalence of 18%, and a prevalence of 42% for CAD70 versus 10% observed (p < 0.001). Among patients with typical angina, CAD50 was predicted to be present in 86%, but was found on CTA in only 29% and CAD70 was predicted to be present in 71%, but found in only 19%. The overestimation of disease prevalence by the guideline probability was found at all of the participating centers for both men and women across all age subgroups.


Current guidelines for clinical prediction of the presence of CAD substantially overestimate the actual presence of disease in patients presenting for evaluation with CTA.


This large multicenter trial nicely demonstrates a substantial and clinically relevant overestimation of the prevalence of CAD by the currently recommended ACC/AHA algorithm, which is an integral part of guidelines for evaluation of patients with stable angina, and are also utilized as part of multiple other documents including appropriate use criteria for stress echocardiography and myocardial perfusion. CTA is well established as an accurate method for determining the presence and severity of CAD, and there is no reason to think that invasive coronary arteriography performed in this same patient population would provide different results. The authors propose several reasons for the overestimation by the current clinical guidelines, including the fact that they are based on studies utilizing invasive coronary arteriography. In many instances, patients being referred for invasive coronary arteriography will represent a different clinical subset than those being more electively evaluated with CTA and probably do represent a population with higher risk for significant obstructive disease than the population noted here. Of note, in addition to overestimation by the ACC/AHA guidelines, other clinical predictors, which include multiple risk factors, also overstated the prevalence of disease as actually found. Previous algorithms for predicting the prevalence of CAD have all demonstrated a hierarchy of presence based on increasing age as well as a higher prevalence of CAD for males across all age ranges, which was confirmed here, although at substantially lower absolute levels of prevalence. This very well done study provides information that is crucial for understanding the performance of other noninvasive strategies for the detection of CAD, and caution is advised in utilizing noninvasive strategies for predicting the presence of CAD as it applies to the appropriateness of subsequent testing.

Clinical Topics: Heart Failure and Cardiomyopathies, Invasive Cardiovascular Angiography and Intervention, Noninvasive Imaging, Prevention, Stable Ischemic Heart Disease, Atherosclerotic Disease (CAD/PAD), Interventions and Coronary Artery Disease, Interventions and Imaging, Angiography, Echocardiography/Ultrasound, Nuclear Imaging, Hypertension, Chronic Angina

Keywords: Coronary Artery Disease, Angina, Stable, Echocardiography, Stress, Risk Factors, Constriction, Pathologic, Dyspnea, Prevalence, Registries, Coronary Angiography, Chest Pain, United States, Diabetes Mellitus, Hypertension

< Back to Listings