Session M1.1

Quantification of Myocardial Perfusion Using Multi-Detector Computed Tomography: Validation against Invasive Coronary Angiography

N Kachenoura*, T Gaspar, JA Lodato, DME Bardo, B Newby,
S Gips, N Peled, RM Lang, V Mor-Avi

University of Chicago
Chicago, IL, USA

Multi-detector computed tomography (MDCT) is an alternative to invasive coronary angiography (ICA). While CT coronary angiography (CTCA) has been validated against ICA and SPECT myocardial perfusion imaging (MPI), the potential of MDCT to evaluate perfusion has not been explored. We sought to: 1) develop and validate a technique for quantitative assessment of myocardial perfusion from MDCT images for the evaluation of coronary artery disease (CAD), 2) identify underlying causes of the detected perfusion abnormalities, and 4) determine the added value of MDCT perfusion when combined with CTCA. Methods. We studied 84 consecutive patients undergoing CTCA (64 pts with ICA; 20 controls), irrespective of calcium scores or coronary stents. MDCT perfusion was measured in terms of attenuation in 16 myocardial segments. Abnormal perfusion was automatically detected by comparison with normal values obtained in the control group. Accuracy of detection was determined against ICA findings (significant CAD: >50% stenosis), combined with known prior myocardial infarction (MI) to take into account the possibility of abnormal perfusion despite patent arteries. Results. Perfusion abnormalities were detected by MDCT at rest in 29/64 patients in 47 vascular territories, of which 36 (77%) were confirmed by the reference technique. Of these 36 abnormalities, 10 (28%) were associated with prior MI, while 26 (72%) corresponded to significant coronary stenosis. The relatively large number of false positive perfusion abnormalities (11/47) was related to image quality: the proportion of images with suboptimal quality (72%) was higher than in true abnormal territories (31%). Despite the decrease in specificity from 0.79 to 0.68 due to these false positives, the addition of resting MDCT perfusion to CTCA improved its diagnostic accuracy on a patient basis (sensitivity from 0.87 to 0.96 and accuracy from 0.84 to 0.88). Conclusions. Myocardial perfusion, even at rest, is a potentially valuable addition to MDCT tools for the evaluation of CAD, especially in patients in whom the interpretation of CTCA is limited by high calcium scores and/or stents. Importantly, this added benefit comes without additional radiation or contrast load.

(Abstract Control Number: 288)