Accuracy and feasibility of contrast echocardiography for detection of perfusion defects in routine practice: comparison with wall motion and technetium-99m sestamibi single-photon emission computed tomography. The Nycomed NC100100 Investigators.

OBJECTIVES: We sought to assess the feasibility and accuracy of myocardial contrast echocardiography (MCE) using standard imaging approaches for the detection of perfusion defects in patients who had a myocardial infarction (MI). BACKGROUND: Myocardial contrast echocardiography may be more versatile than perfusion scintigraphy for identifying the presence and extent of perfusion defects after MI. However, its reliability in routine practice is unclear. METHODS: Fundamental or harmonic MCE was performed with continuous or triggered imaging in 203 patients with a previous MI using bolus doses of a perfluorocarbon-filled contrast agent (NC100100). All patients underwent single-photon emission computed tomography (SPECT) after the injection of technetium-99m (Tc-99m) sestamibi at rest. Quantitative and semiquantitative SPECT, wall motion and digitized echocardiographic data were interpreted independently. The accuracy of MCE was assessed for detection of segments and patients with moderate and severe sestamibi-SPECT defects, as well as for detection of patients with extensive perfusion defects (>12% of left ventricle). RESULTS: In segments with diagnostic MCE, the segmental sensitivity ranged from 14% to 65%, and the specificity varied from 78% to 95%, depending on the dose of contrast agent. Using both segment- and patient-based analysis, the greatest accuracy and proportion of interpretable images were obtained using harmonic imaging in the triggered mode. For the detection of extensive defects, the sensitivity varied from 13% to 48%, with specificity from 63% to 100%. Harmonic imaging remained the most accurate approach. Time since MI and SPECT defect location and intensity were all determinants of the MCE response. The extent of defects on MCE was less than the extent of either abnormal wall motion or SPECT abnormalities. The combination of wall motion and MCE assessment gave the best balance of sensitivity (46% to 55%) and specificity (82% to 83%). CONCLUSIONS: Although MCE is specific, it has limited sensitivity for detection of moderate or severe perfusion defects, and it underestimates the extent of SPECT defects. The best results are obtained by integration with wall motion. More sophisticated methods of acquisition and interpretation are needed to enhance the feasibility of this technique in routine practice.