Noninvasive quantitative evaluation of coronary artery stent patency using 64-row multidetector computed tomography

Abstract

Many studies have used multidetector computed tomography (MDCT) angiography to evaluate coronary stents qualitatively but not quantitatively. This study sought to validate a method of quantitatively evaluating stent patency by using 64-row compared with invasive coronary angiography (ICA) and to evaluate the stent size threshold of MDCT in detecting stent patency. Stented lesions (n=122) in 55 patients (age, 65+/-10 years; 90% men) who underwent both 64-row MDCT and ICA were studied. Density measurements in Hounsfield units (HUs) and stent diameters in millimeters were recorded in the stented segments, with the density of the ascending aorta (AO) taken as a reference. The ratio of the average of stent's proximal, middle, and distal densities to mean AO density was defined as the AS/AO HU. Threshold values for the detection of stent patency were examined by using receiver operator characteristic (ROC) curve analysis. One hundred six of 122 stents were interpretable. By ICA, 24 stents were found to have in-stent restenosis (22 interpretable and 2 noninterpretable with MDCT). The ROC curve showed that the optimal cutoff value of AS/AO HU to predict stent patency on MDCT was 0.81 with sensitivity of 90.9%, specificity of 95.2%, and the optimal stent diameter cutoff value was > or = 2.5 mm with a sensitivity of 91.8% and a specificity of 93.8%. With 64-row MDCT, coronary stent patency can be evaluated quantitatively with high sensitivity and specificity and with adequate diagnostic accuracy in stents > or = 2.5 mm in diameter.