Peripheral Arterial Plaque Classification With Intravascular Ultrasound: A Radiofrequency Data Analysis.

OBJECTIVES: Spectral analysis of backscattered radiofrequency waves from intravascular ultrasound (IVUS) has shown potential for in vivo atherosclerotic plaque classification for coronary arteries. However, it is unknown if this technology can be successfully applied to lower extremity peripheral arteries. The aim of this research was to determine if peripheral plaque stratification is feasible and accurate.
METHODS: Arterial specimens from above knee (n=55) and below knee (n=19) amputations were obtained within 24 hours of amputation in patients with severe atherosclerotic disease. Arteries were pressure perfused and IVUS imaged ex vivo. Sutures were placed on the vessel at corresponding IVUS cross sections of interest and captured for later correlation with histology. Following IVUS imaging, arteries were pressure perfused-fixed with sutured locations sent for histology. IVUS images and Movat pentachrome stained histology cross sections (n=456) were evaluated to locate fibrous, fibro-fatty, necrotic core, and dense calcium plaque. Homogenous plaque regions of interest (ROI), as determined by histology, were chosen from near perfect matches as determined by two observers. Fibrous (n=172), dense calcium (n=41), and necrotic core (n=36) regions of interest were identified from popliteal (n=19) and tibial artery (n=4) atherosclerotic plaque. Spectral parameters for each ROI were calculated using an autoregressive method. An atherosclerotic plaque classification algorithm was then generated using the spectral parameters.
RESULTS: Peripheral arteries were found to have scant regions of fibro-fatty plaque as opposed to coronary arteries. Predictive accuracies were 83.5% for fibrous, 92.0% for dense calcium, and 89.2% for necrotic core plaque regions. Overall accuracy was 85.0%. Tissue maps could be accurately reconstructed with the use of predictions of plaque composition from the autoregressive classification scheme.
CONCLUSIONS: Lower extremity peripheral plaque composition can be predicted through the use of IVUS radiofrequency data analysis. Autoregressive classification schemes provide comparable accuracies to previously reported coronary accuracies. These techniques may allow for real-time in vivo analysis of IVUS data plaque characterization to be extended to lower extremity arteries.