Color-encoded semiautomatic analysis of multi-slice first-pass magnetic resonance perfusion: comparison to tetrofosmin single photon emission computed tomography perfusion and X-ray angiography

Abstract

Background and purpose: Cardiovascular magnetic resonance (CMR) perfusion can accurately detect coronary artery disease (CAD). However, the absence of efficient, easy-to-use and reliable image analysis software is an obstacle to its introduction into clinical practice. The aim of this study was to evaluate new color-encoded semiautomatic software for analysis of first-pass CMR perfusion in comparison to tetrofosmin myocardial single photon emission computed tomography (SPECT), using X-ray angiography as the standard of truth for the detection of CAD.

Methods: Thirty-two patients underwent both SPECT and CMR perfusion at rest and adenosine stress. Twenty of these patients also underwent X-ray angiography. Off-line CMR image analysis consisted of six steps to generate a color display of the myocardial perfusion reserve index (MPRI). The MPRI color-maps were analyzed visually and compared to SPECT.

Results: In comparison to X-ray angiography overall accuracy was 87% for CMR and 77% for SPECT perfusion to detect significant CAD (stenosis > or =70%). In comparison with SPECT sensitivity was 80%, specificity 91%, and the overall agreement 89% for CMR.

Conclusions: Post-processing of CMR perfusion data using new semiautomatic software to generate and display the MPRI visually as color-encoded images is feasible and fast. In this study it yielded higher accuracy than SPECT to detect significant CAD on X-ray angiography. Correlation between SPECT and CMR accuracy for detection of perfusion defects was high. This method may accelerate the time-consuming analysis of CMR perfusion data, thus enabling a more widespread clinical utility.