Reproducibility in the assessment of noncalcified coronary plaque with 256-slice multi-detector CT and automated plaque analysis software

Abstract

We aimed to assess intra- and inter-observer reproducibility in evaluating volume and characteristics of non-calcified coronary plaques (NCPs) using a 256-slice multi-detector computed tomography (MDCT) angiography and dedicated automated plaque analysis software in asymptomatic individuals. Forty-two NCPs from 39 patients with a vessel diameter >2mm were evaluated using a 256-slice MDCT with dedicated automated plaque analysis software. Two independent observers analyzed the characteristics of NCPs, including plaque volume (vol), mean CT number of the NCPs in Hounsfield units (HU(mean)), and remodeling index (RI). One of the observers repeated the evaluation of all datasets after an interval of at least 4weeks. Bland-Altman analysis and concordance correlation coefficients (CCCs) were used to determine intra- and inter-observer variability. For vol measurements, the 95% limits of agreement were -21.6 and 13.2mm(3), and -24.6 and 20.3mm(3) for intra- and inter-observer variability, respectively. For HU(mean) measurements, the 95% limits of agreement were -22.2 and 20.8HU, and -21.1 and 21.0HU for intra- and inter-observer variability, respectively. For RI measurements, the 95% limits of agreement were -0.38 and 0.39, and -0.51 and 0.36 for intra- and inter-observer variability, respectively. The CCCs was very high for all measurements, ranging from 0.90 to 0.98. Using 256-slice MDCT with dedicated automated plaque analysis software, intra- and inter-observer reproducibility were excellent in evaluating the volume and characteristics of NCP in asymptomatic individuals.