Diagnostic accuracy of quantitative cardiac MRI evaluation compared to stress single-photon-emission computed tomography

Background : Cardiac MRI (cMRI) perfusion is a promising non-invasive tool to assess myocardial ischemia. The accuracy of quantitative cMRI perfusion has been recently demonstrated, but to date no previous study has compared this technique with stress single-photon-emission computed tomography (SPECT). The aim of this study was to evaluate the diagnostic accuracy of myocardial perfusion reserve (MPR) based on cMRI compared with SPECT. Methods : We examined 24 patients who underwent coronary angiography, stress SPECT and cMRI perfusion. Qualitative assessment of both SPECT and cMRI images, quantification of cMRI perfusion, and quantitative coronary angiography (QCA) were independently performed. MPR was calculated using Fermi deconvolution technique. Accuracy of quantitative and qualitative data was examined to detect > 50% diameter stenosis (DS) by QCA. Results : Qualitative analysis was obtained in 198 segments and quantitative analysis was performed in 171 segments. Significant coronary artery disease (CAD) was present in 81.8% of patients. Visual cMRI assessment yielded sensitivity of 74.4% and specificity of 79.4% to predict > 50%DS, while SPECT showed sensitivity of 67.4% and specificity of 81.3%. The sensitivity for SPECT in the right coronary artery territory and apex was low compared to cMRI. Sensitivity and specificity for detection of significant CAD were 89.5% and 46.6% for MPR (cutoff 1.92). Area under the curve was 0.75 for MPR ( P