Detection of coronary artery disease using real-time myocardial contrast echocardiography: a comparison with dual-isotope resting thallium-201/stress technectium-99m sestamibi single-photon emission computed tomography

Abstract

Real-time myocardial contrast echocardiography (MCE) has the potential to evaluate myocardial perfusion and wall motion (WM) simultaneously. The purposes of this study were to correlate the diagnostic value of MCE with radionuclide single-photon emission computed tomography (SPECT), and to assess the sensitivity and specificity of real-time MCE in detecting coronary artery disease (CAD). Seventy patients with clinically suspected CAD underwent MCE and SPECT at baseline and after dipyridamole infusion. Segmental perfusion with MCE using low mechanical index after 0.3-0.4-ml bolus injections of perfluorocarbon exposed sonicated dextrose albumin solution was performed. All patients had a dual-isotope (rest thallium-201, stress sestamibi) study performed both at baseline and after dipyridamole infusion, and 40 patients had subsequent quantitative coronary angiography. Abnormalities were noted in 27 patients (38.6%) by MCE, in 29 patients (41.4%) by WM analysis, and in 30 patients (42.9%) by SPECT imaging. When MCE and WM analysis were combined, the agreement with SPECT imaging improved from 75.7% (Kappa = 0.50) to 82.0% (Kappa = 0.62). In 40 patients (120 territories) who underwent coronary angiography, good perfusion concordance was achieved for the left anterior descending and left circumflex arteries, and was fair for the right coronary arteries. Compared with quantitative angiography, there was no difference in sensitivity, specificity, and accuracy in detecting significant CAD among the three modalities. The combination of MCE and WM had a better sensitivity (84%), specificity (93.3%), and accuracy (87.5%) than the MCE and WM analysis alone. However, the difference did not reach statistical significance. Real-time MCE has a good agreement with SPECT imaging for detecting CAD. The combination of MCE and WM appears to have higher sensitivity, specificity, and accuracy in detecting CAD than either technique alone.