The purpose of this study was to determine whether myocardial contrast echocardiography (MCE) can be used to detect coronary artery disease (CAD) during rest and pharmacological stress in humans through the use of venous injections of contrast. Thirty patients with known or suspected CAD underwent MCE and 99mTc-sestamibi single-photon emission computed tomography (SPECT) at baseline and after dipyridamole (0.56 mg x kg(-1)) infusion. Ten myocardial segments (5 each in the apical two- and four-chamber views) from the two sets of images using both methods were scored for myocardial perfusion as follows: 1=normal, 0.5=mildly reduced, and 0=severely reduced. The information from baseline and postdipyridamole images was then used to determine whether an abnormal segment was irreversible (similar abnormal perfusion at baseline and after dipyridamole) or reversible (perfusion better at baseline compared with after dipyridamole). Concordance between segmental scores was 92% (kappa=.99) for both methods. Concordance between normal perfusion and reversible or irreversible segmental defects was 90% (kappa=.80). Agreem between the two methods for each of the three vascular territories in each patient was 90% (kappa=.77), while agreement for the presence or absence of CAD in each patient was 86% (kappa=.86). In the 4 patients with disagreement, the perfusion scores were 0.5 for SPECT and 1.0 for MCE. This study shows that MCE, with venous injection of contrast, can define the presence of CAD during rest and pharmacological stress. The location of perfusion abnormalities and their physiologic relevance (reversible or irreversible) by MCE is similar to that provided by SPECT. MCE, therefore, holds promise for the noninvasive assessment of myocardial perfusion in humans.