Coronary plaque composition as assessed by greyscale intravascular ultrasound and radiofrequency spectral data analysis

Objectives: (i) To explore the relation between greyscale intravascular ultrasound (IVUS) plaque qualitative classification and IVUS radiofrequency data (RFD) analysis tissue types; (ii) to evaluate if plaque composition as assessed by RFD analysis can be predicted by visual assessment of greyscale IVUS images. Methods: In 120 IVUS-RFD cross-sections, a sector of the plaque with homogenous tissue composition (e.g., fibrous, fibrofatty, necrotic core, and dense calcium) was selected. Two experienced observers analyzed twice the corresponding greyscale IVUS images to: (1) classify the selected sectors according to greyscale IVUS plaque type classification and (2) predict the tissue type expected in the sector by RFD analysis. Results: In the greyscale IVUS plaque type classification, the observers agreed in 90/120 sectors (κ = 0.64). Calcified, soft and mixed plaques by greyscale IVUS classification were mainly composed of dense calcium, fibrofatty, and necrotic core, respectively, in the RFD analysis. The plaques classified in greyscale IVUS as fibrous were actually fibrous tissue by IVUS RFD in only 30% of the cases. Overall, high interobserver variability in the prediction of RFD results by visual assessment of greyscale IVUS images (κ = 0.23 for observer 1 and 0.55 for observer 2) was found. Sensitivities for detection of calcified tissue and NC by greyscale IVUS visual assessment were 88% and 58%, respectively. Conclusions: High interobserver variability in the prediction of tissue type by visual assessment of greyscale IVUS images was observed. This underlines the need of quantitative methods for the analysis of the ultrasound characteristics of coronary plaque components.