Ultrasonographic morphology of unstable carotid plaque: Is the standard computer-assisted GSM analysis precise enough?

Abstract

Background: Carotid plaque instability continues to be a problem in perioperative risk assessments among patients qualified for CEA. The aim of the study was to compare different computer-assisted techniques from the aspect of preoperative and perioperative complications.
Material and methods: The authors have evaluated 3 programs (2 commercially available and 1 specifically designed for the purpose of the study). 130 carotid plaques were analyzed (derived from 125 patients who underwent operations for CEA). Apart from a computer analysis, the following elements were evaluated: microembolism (assessed by TCD), intraoperative evaluation of the plaque and its histopathological examination. A postoperative course was also assessed including the results of a CT and NMR scan performed whenever it was deemed to be necessary. All the results were analyzed with the use of multivariate analysis and Pearson’s chi-square tests.
Results: The application of computer-assisted techniques in US evaluations of carotid plaque morphology proves to be an important tool in the diagnosis of plaque instability features and perioperative risk assessment. The multivariate analysis revealed the superiority of pixel distribution analysis, as the most accurate and specific of the 3 computer programs evaluated. Standard GSM analysis is a valuable method, though less precise than pixel distribution analysis.