RESUMO
BACKGROUND AND AIMS: Positive predictive value (PPV) of adenosine stress cardiac magnetic resonance (CMR) for coronary artery disease (CAD) is unsatisfactory. We investigated the impact of coronary caliber variability on this limitation in CMR performance. METHODS AND RESULTS: 206 consecutive patients with myocardial ischemia during CMR and subsequent coronary angiography (CA) were studied. Patients were examined in a 1.5-T scanner. After adenosine infusion, myocardial first-pass sequence using gadolinium-based contrast agent was performed and compared with rest perfusion. CAD was invasively confirmed in 165 [true positive (TP); PPV, 80.1%] and ruled out in 41 patients [false positive (FP)]. TP and FP were comparable for pre-test risk and CMR findings. We found a significant association between FP CMR and the presence of a small caliber coronary vessel (proximal diameter < one standard deviation below the mean) supplying the area of ischemia (chi-square 42.6, p < 0.0001). A small caliber artery ipsilateral to the ischemic region was a predictive parameter for FP versus TP discrimination (ROC area, 0.84 ± 0.04 vs. 0.59 ± 0.05; p < 0.0001). Further increment in diagnostic accuracy was achieved by including proximal ipsilateral/contralateral coronary diameter ratios (ROC area, 0.90 ± 0.03; p < 0.03). CONCLUSIONS: Small caliber coronary arteries found as normal variations in right-dominant or left-dominant circulation may account for hypoperfusion in the absence of coronary stenosis and thus may cause FP adenosine stress CMR results. Non-invasive assessment of proximal coronary diameters in the vessel supplying the area of ischemia could reduce FP rates, raise the diagnostic accuracy of CMR for CAD and minimize subsequent superfluous CA.