IN-VIVO RADIOFREQUENCY-BASED ULTRASONIC TISSUE CHARACTERIZATION OF THE ATHEROSCLEROTIC PLAQUE

Background and Purpose: The ultrasonic image can offer unique information on the composition of atherosclerotic plaque, ie, the relative content of lipids, fibrous tissue, and calcific deposits. To date, however, the echographic assessment of plaque structure is based on a subjective, qualitative evaluation of the bidimensional images. We evaluated the feasibility and accuracy of assessing, in vivo, the acoustic properties of arterial carotid plaques by means of a suitably modified echographic apparatus allowing direct access to the radiofrequency signal. Methods: In 15 patients undergoing carotid thromboendarterectomy, the ultrasonic findings in 70 discrete sites (within the plaque, n=54; normal sites, n=11; or intraluminal thrombi, n=5) were correlated with the histological analysis (hematoxylin-eosin and Mallory trichrome stains) independently performed on the arterial samples. The pathological examination was carried out at a similar level of the insonation; the sites analyzed within the plaque were chosen because of their uniform echoic characteristics. In each ultrasonic region of interest selected from the echographic image, the integrated amplitude of the rectified radiofrequency signal was measured as the integrated backscatter index. Results: The intimal-medial layer of normal carotid wall (n=11) exhibited values of -32.5+/-9.4 dB. The integrated backscatter index in fatty sites (n=11, -40.3+/-5.4 dB) differed from that of fibrous (n=12, -23.8+/-5.0 dB) and calcified (n=26, -11.5+/-5.2 dB, P<.01 for all intergroup differences) sites. Intraluminal thrombotic sites (n=5, -42+/-5.1 dB, P<.01) differed from fibrous and calcified subsets (P<.01) but overlapped (P=NS) with fatty sites. Histological sampling also showed two sites of intraplaque hemorrhage that exhibited very low backscatter values (-53 and -58 dB) and three fibrofatty sites showing backscatter values (-28, -28, and -32 dB) intermediate between the fibrous and the fatty subsets. Conclusions: Quantitative analysis of integrated backscatter of the arterial wall is feasible in humans and provides an operator-independent assessment of plaque echoic structure. In particular, integrated backscatter is effective in distinguishing lipidic, fibrotic, and calcific components in human atherosclerotic plaques.