Computer-assisted ultrasonographic analysis of carotid plaques in relation to cerebrovascular symptoms, cerebral infarction, and histology.

The objective of this study was to determine the relationship between plaque echogenicity as measured by computer and the incidence of cerebrovascular symptoms and cerebral infarction. The correlation between carotid plaque echogenicity and plaque histology was also evaluated. In this prospective nonrandomized study, 38 consecutive patients with 54 atherosclerotic carotid plaques producing 50-99% stenosis were reviewed. The ultrasonic images of the plaques were digitized and transferred to a computer. A histogram for each plaque representing its composition was obtained. The median of the gray scale (GSM) of each histogram was used as measure of plaque echogenicity. All patients had a computed tomography (CT) brain scan performed to determine the presence of cerebral infarction. Twenty-eight plaques were examined histologically to determine the deposition of calcium, hemorrhage, cholesterol, and amorphous granular material. It is possible to identify carotid plaques at high risk for development of cerebrovascular symptoms and cerebral infarction by the computerized measurement of plaque echogenicity. This method may be used to improve the criteria of patients selection for carotid endarterectomy.