Densitometric quantitative analysis of intracoronary ultrasound images: anatomopathological correlation

Abstract

To establish if the videodensitometric analysis (VDA) of the intracoronary ultrasound images (IVUS) can predict the qualitative and quantitative composition of the atherosclerotic coronary plaques, thirty-one patients who had undergone anatomopathological study of directional coronary atherectomy (DCA) samples and pre- and post-intervention IVUS image were analyzed. The video IVUS images were digitized in a 512 x 512 matrix and analyzed for densitometric differences with an Automatic Image Analysis System (AIAS) (Vidas 2000, Zeiss Kontron). The components of the plaque were arbitrarily divided into three densitometric categories using a 256 gray scale: high density (HD) 121-255, medium (MD) 81-120 and low (LD) 30-80. The relative percentage of each component was automatically recorded. The DCA samples were microscopically examined and put into the AIAS. The components were divided into: collagenous tissue (CT); lipid-necrotic debris (LND); proliferative tissue (PT). The area of each component was expressed as a percentage of the total. Linear correlation analysis was applied. Comparison between the IVUS and the histological composition of the plaque showed that: HD corresponded to CT; MD to PT; LD to LND. The correlation between the percentage distribution of the densitometric categories and the anatomopathological components showed a correlation coefficient r = 0.91 between HD and CT; r = 0.87 between MD and PT; r = 0.88 between LD and LND. The VDA of the IVUS can distinguish three basic components of the atherosclerotic plaque: fibrous, lipid-necrotic and proliferative tissue, allowing absolute and relative quantitative analysis. This capability may be of interest for device selection and histopathological correlation.