Incremental value of adenosine stress cardiac magnetic resonance in coronary artery disease detection

Introduction: Cardiac magnetic resonance myocardial perfusion imaging (CMR-MPI) is considered a stateof- the-art non-invasive modality for ischemia detection but its additive value in a multiple-test strategy in patients with suspected coronary artery disease (CAD) is not fully validated. We aimed to evaluate CMR-MPI integration with exercise treadmill test (ETT) for the diagnostic workup of patients with suspected CAD, having invasive fractional flow reserve (FFR) as reference standard. Methods: In this prospective single-center study, patients with suspected CAD underwent sequential ETT, CMR-MPI and X-ray invasive coronary angiography (XA). Significant CAD was defined by the presence of stenosis N40% with FFR <= 0.8 in vessels >2 mmor >= 90% stenosis/occlusion. Results: 80 symptomatic patients (68% male, 61 +/- 8 years) were enrolled. Compared to ETT, CMR-MPI showed similar sensitivity (81%) and higher specificity (93 vs. 58%, p < 0.001) for CAD detection (prevalence = 46%) translating into better diagnostic performance (AUC 0.87 vs. 0.70; p = 0.002). CMR-MPI improved accuracy independently of ETT in all patients with high pre-test probability and in intermediate-probability patients but those with a clearly positive-ETT (symptoms + ST-shift), in whom ETT correctly identified CAD. In the lowprobability group CMR-MPI was useful as a gatekeeper for XA after a positive-ETT. The best integrating protocol achieved a global accuracy of 89% (AUC 0.88) and was clearly superior to an approach based solely in ETT (AUC 0.70, p < 0.001), yet similar to isolated CMR-MPI (AUC 0.87, p = ns). Conclusions: CMR-MPI has high sensitivity and specificity for CAD detection and may be combined with ETT in a diagnostic workflow aiming to increase accuracy and reduce the number of unnecessary catheterizations. (C) 2013 Elsevier Ireland Ltd. All rights reserved.