Absolute Versus Relative Myocardial Blood Flow by Dynamic CT Myocardial Perfusion Imaging in Patients With Anatomic Coronary Artery Disease

OBJECTIVE. The purpose of this study was to evaluate differences in the diagnostic accuracy of absolute and relative territorial myocardial blood flow (MBF) derived from stress dynamic CT myocardial perfusion imaging (MPI) for the detection of significant coronary artery stenosis. MATERIALS AND METHODS. Dynamic CT MPI and coronary CT angiography (CTA) datasets from a multicenter registry of 137 patients (mean age, 60.9 +/- 8.4 years; 88 men) with suspected or known coronary artery disease were retrospectively analyzed. For each coronary territory, absolute MBF and the MBF relative to remote myocardium (MBF ratio) were calculated. Coronary CTA datasets were visually assessed for significant stenosis (>= 50% luminal narrowing) in consensus by two observers. RESULTS. Significant stenosis was detected in 137 of 411 (33.3%) vessels. Mean absolute MBF and MBF ratio were statistically significantly lower in territories supplied by arteries with stenosis (80.7 +/- 33.7 vs 140.0 +/- 38.4 mL/100 mL/min and 0.52 vs 0.89, respectively; both p < 0.0001). ROC analysis showed better discrimination by MBF ratio than by absolute MBF (AUC, 0.925 vs 0.882; p = 0.0022) and increased sensitivity (90.7% vs 82.4%; p < 0.04) and specificity (93.1% vs 80.5%; p < 0.03) for MBF ratio and absolute MBF cutoff values of 0.71 or less and 103 mL/100 mL/min or less, respectively. CONCLUSION. In stress dynamic CT MPI, relative MBF provides superior diagnostic accuracy compared with absolute territorial MBF values for coronary CTA-detected significant coronary artery stenosis.