In vivo plaque characterization using intravascular ultrasound-virtual histology in a porcine model of complex coronary lesions.

Objective - To determine the accuracy of detection of different tissue types of intravascular ultrasound - virtual histology (IVUS-VH) in a porcine model of complex coronary lesions. Methods and Results - Coronary lesions were induced by injecting liposomes containing human oxidized low-density lipoprotein into the adventitia of the arteries. IVUS-VH imaging was performed in vivo at 8.2 +/- 1.6 weeks after injection. A total of 60 vascular lesions were analyzed and compared with their correspondent IVUS-VH images. Correlation analysis was performed using linear regression models. Compared with histology, IVUS-VH correctly identified the presence of fibrous, fibro-fatty, and necrotic tissue in 58.33%, 38.33%, and 38.33% of lesions, respectively. The sensitivity of IVUS-VH for the detection of fibrous, fibro-fatty, and necrotic core tissue was 76.1%, 46%, and 41.1% respectively. A linear regression analysis performed for each individual plaque component did not show strong correlation that would allow significant prediction of individual values. Conclusions - In a porcine model of complex coronary lesions, IVUS-VH was not accurate in detecting the relative amount of specific plaque components within each individual corresponding histological specimen.