Background-The aims of our study were to (1) examine how data from exercise treadmill testing (ETT) can identify patients who have coronary plaque or stenosis, using CT angiography (CTA) as the reference standard, and (2) identify patient characteristics that may be used in selecting ETT versus CTA. Methods and Results-The Rule Out Myocardial Infarction Using Computer-Assisted Tomography (ROMICAT) trial was an observational cohort study of acute chest pain patients presenting to the emergency department with normal initial troponin and a nonischemic ECG. Univariate and multivariable analyses were performed to assess the relationship of baseline clinical data and ETT parameters with coronary plaque and stenosis on CTA. Of the 220 patients who had ETT (mean age, 51 years; 63% men), 21 (10%) had positive results. A positive ETT had a sensitivity of 30% and specificity of 93% to detect Ͼ50% stenosis. The sensitivity increased to 83% after excluding uninterpretable segments and evaluating the ability to detect a Ͼ70% stenosis. Predictors of plaque included older age, male sex, diabetes, hypertension, hyperlipidemia, lower functional capacity, and a lower Duke Treadmill Score. Both a positive ETT and a low Duke Treadmill Score were significant univariate and multivariable predictors of stenosis Ͼ50% on CTA Whereas the prevalence of stenosis by CTA was greater among patients with more risk factors, coronary stenosis was not present among men Ͻ40 years old or women Ͻ50 years old or individuals who achieved at least 13 metabolic equivalents on ETT.
Conclusions-Among