BACKGROUND: A quantitative technique is required to reduce the subjectivity and improve the reproducibility of stress echocardiography. Tissue Doppler imaging may offer these benefits, but its feasibility with exercise echocardiography (ExE) is undefined. This study sought the determinants of the exercise tissue Doppler velocity (TDV) response and the feasibility and accuracy of color TDV during ExE. METHODS AND RESULTS: Fifteen volunteers and 85 patients (age 60 +/- 10 years, 19 women) with known or suspected coronary artery disease were studied with standard 2-dimensional (2D) echocardiography and pulsed wave (PW) and color TDV before and after they underwent exercise treadmill testing. After the study PW TDV was measured in 6 basal segments, and off-line software was used to display color TDV data from all myocardial segments. Color TDV was compared with PW TDV in the basal segments at rest and stress with the use of linear regression. Color TDV in mid and basal segments was compared with wall motion on 2D echocardiography. The predictors of the TDV response to exercise were defined in a multiple linear regression. A logistic regression model was used to integrate clinical, exercise, and TDV variables for prediction of abnormal regional left ventricular function. Color and PW correlated well at rest (r = 0.81) and stress (r = 0.84), but PW was greater than color velocities at rest and stress. On the basis of 2D echocardiography, 752 myocardial segments were classified as normal in patients without evidence of coronary disease, 309 were normal in patients with abnormal wall motion in another territory, and 128 showed ischemia or scar. Segmental comparison of velocities assessed by color TDV showed that scar segments had a lower velocity than normal segments at rest and stress (P <.001). Ischemic segments had a lower peak TDV and less increment in velocity than normal segments. Heart rate, functional capacity, and regional dysfunction (scar or ischemia) were independent predictors of TDV at peak exercise. With the use of receiver operating characteristic analysis, the "correction" of TDV by these other variables increased the accuracy of the technique for the detection of regional left ventricular dysfunction. CONCLUSION: Color TDV is feasible during ExE. The correlation found between TDV and wall motion analysis of experienced observers indicates that TDV may be useful as a quantitative tool for interpretation of ExE.