Multidetector computed tomography myocardial perfusion imaging during adenosine stress

OBJECTIVES The purpose of this study is to validate the accuracy of multidetector computed tomography (MDCT) to measure differences in regional myocardial perfusion during adenosine stress in a canine model of left anterior descending (LAD) artery stenosis, during first-pass, contrast-enhanced helical MDCT. BACKGROUND Myocardial perfusion imaging by MDCT may have significant implications in the diagnosis and treatment of coronary artery disease. METHODS Eight dogs were prepared with a LAD stenosis, and contrast-enhanced MDCT imaging was performed 5 min into adenosine infusion (0.14 to 0.21 mg/kg/min). Images were analyzed using a semiautomated approach to define the regional signal density (SD) ratio (myocardial SD/left ventricular blood pool SD) in stenosed and remote territories, and then compared with microsphere myocardial blood flow (MBF) measurements. RESULTS Mean MBF in stenosed versus remote territories was 1.37 +/- 0.46 ml/g/min and 1.29 +/- 0.48 ml/g/min at baseline (p = NS) and 2.54 +/- 0.93 ml/g/min and 8.94 +/- 5.74 ml/g/min during adenosine infusion, respectively (p < 0.05). Myocardial SD was 92.3 +/- 39.5 HU in stenosed versus 180.4 +/- 41.9 HU in remote territories (p < 0.001). There was a significant linear association of the SD ratio with MBF in the stenosed territory (R = 0.98, p = 0.001) and between regional myocardial SD ratio and MBF < 8 ml/g/min, slope = 0.035, SE = 0.007, p < 0.0001. Overall, there was a significant non-linear relationship over the range of flows studied (LR chi-square [2 degrees of freedom] = 31.8, p < 0.0001). CONCLUSIONS Adenosine-augmented MDCT myocardial perfusion imaging provides semiquantitative measurements of myocardial perfusion during first-pass MDCT imaging in a canine model of LAD stenosis.