Noninfarcted myocardium: correlation between dynamic first-pass contrast-enhanced myocardial MR imaging and quantitative coronary angiography.

Abstract

To determine the accuracy of first-pass contrast material-enhanced stress myocardial magnetic resonance (MR) imaging for depiction of myocardial ischemia in patients without myocardial infarction. First-pass contrast-enhanced MR images of the entire left ventricle were acquired in 104 patients at rest and during dipyridamole-induced stress by using an interleaved notched saturation technique. Coronary angiography was performed in all patients, and stress perfusion single photon emission computed tomography (SPECT) was performed in 69 patients. Receiver operating characteristic curve analysis was performed to compare the diagnostic accuracies of first-pass contrast-enhanced stress MR imaging and stress SPECT, with coronary angiography as the reference standard. The overall sensitivity of MR imaging for depicting at least one coronary artery with significant stenosis was 90% (69 of 77 patients). The sensitivities of MR imaging for depiction of single-, double-, and triple-vessel stenoses were 85% (33 of 39 patients), 96% (22 of 23 patients), and 100% (15 of 15 patients), respectively. The specificity of MR imaging for identification of patients with significant coronary artery stenoses was 85% (23 of 27 patients). The areas under the receiver operating characteristic curve for detection of significant stenosis in individual coronary arteries were 0.888 (observer 1) and 0.911 (observer 2) for MR imaging and 0.707 (observer 1, P <.001) and 0.750 (observer 2, P <.001) for SPECT. In patients without myocardial infarction, stress enhancement at dynamic MR imaging correlates more closely with quantitative coronary angiography results than does stress enhancement at SPECT.