Reproducibility of carotid atherosclerotic lesion type characterization using high resolution multicontrast weighted cardiovascular magnetic resonance.

PURPOSE Cardiovascular magnetic resonance (CMR) can characterize carotid atherosclerosis. The purpose of this study is to evaluate reader agreement of carotid atherosclerotic lesion types by CMR. METHODS Carotid arteries of 34 patients (29 men, 5 women; mean age, 53 years) were imaged on a 1.5-T scanner. Images with 4 contrast weightings (T1, T2, proton density, and 3-dimensional time-of-flight) were acquired on each axial slice of the carotid arteries. Modified AHA criteria were used for lesion type assessment on the 4 selected axial slices (1 from the common carotid artery, 1 from the carotid bifurcation, and 2 from the internal carotid artery). The modified AHA criteria are as follows: type I-II, near-normal wall thickness without calcification; type III, diffuse wall thickening or small eccentric plaque without calcification; type IV-V, plaque with a lipid rich necrotic core surrounded by fibrous tissue with possible calcification; type VI, complex plaque with a possible surface defect, hemorrhage, or thrombus; type VII, calcified plaque; and type VIII, fibrotic plaque without a lipid core and with possible small calcifications. RESULTS Of the 272 possible axial slices (34 patients x 2 arteries per patient x 4 slices per artery), 256 slices were available for lesion type assessment. The majority (94%) of the lesions were of type I-II and III. kappa was 0.80 and 0.60 for intra-reader and inter-reader agreement of lesion types, respectively. Inter-reader disagreement for type I-II and type III occurred in 82% of lesions. Weighted kappa was 0.92 and 0.83 for intra-reader and inter-reader agreement of lesion types, respectively. CONCLUSION The difference between type I-II and III lesions lies in the definition of the vessel wall. The moderate inter-reader agreement suggests further efforts such as establishment of normal carotid artery wall thickness by a quantitative measure are needed for carotid atherosclerotic lesion characterization.