To determine the correlation of quantitative assessment of coronary narrowings with left ventricular functional impairment induced by exercise, 57 patients with 1-vessel coronary artery disease and without evidence of collateral flow were studied. A significant relation was observed between minimal cross-sectional area, percent area stenosis, minimal lumen diameter, percent diameter stenosis and the percentage of segmental area change from rest to peak exercise in a vascular distribution territory (r = 0.76, p less than 0.001; r = -0.55, p less than 0.001; r = 0.56, p less than 0.001; r = -0.75, p less than 0.001, respectively). For minimal cross-sectional area, the best cut-off value to separate significantly patients who had a decrease in contractility at peak exercise testing from those who had a normal response was 2 mm2 (p less than 0.001); for percent cross-sectional area stenosis, it was 75% (p less than 0.001); for minimal lumen diameter, it was 0.7 mm (p less than 0.001); and, for percent diameter stenosis, it was 85% (p less than 0.001). High cut-off values for angiographic variables are necessary to separate significantly patients who have a decrease in contractility at peak exercise testing from those who have a normal response. Several patients with mild coronary stenoses may have either normal or abnormal wall motion during exercise. Thus, exercise echocardiography is a useful tool in detecting the presence of fairly severe anatomic narrowing, whereas it is of limited clinical use in the assessment of intermediate coronary atherosclerotic lesions.