Abstract

The results of a prospective study comparing ultrasound, intra-arterial digital subtraction angiography, and magnetic resonance angiography in the assessment of the degree of extracranial internal carotid artery stenosis are reported in patients with symptoms of recent carotid territory ischaemia. A total of 70 patients and 137 vessels were examined by all three techniques. The results obtained by each technique were reported blind. The mean difference (SD) for the comparison of magnetic resonance angiography and digital subtraction angiography was -0.7 (14)%, for ultrasound and digital subtraction angiography 3.1 (15)%, and for magnetic resonance angiography and ultrasound -3.8 (15)%. The level of agreement was greater for the more tightly stenosed vessels. With the assumption that the results of the digital subtraction angiogram reflect the true situation, the sensitivity and specificity in the detection of > or = 30% stenoses were 93% and 82% with ultrasound and 89% and 82% with magnetic resonance angiography; for stenoses > or = 70% 93% and 92% with ultrasound and 90% and 95% with magnetic resonance angiography; and for stenoses of 70-99% 89% and 93% with ultrasound and 86% and 93% with magnetic resonance angiography. For occlusion the values were 93% and 99% with ultrasound and 80% and 99% with magnetic resonance angiography. Increased sensitivity and specificity were obtained when analysis was confined to those vessels in which ultrasound and magnetic resonance angiography were in agreement over classification. It is thus possible to accurately categorize the degree of stenosis of the extracranial internal carotid artery from a combination of ultrasound and magnetic resonance angiography. The adoption of this combination for the investigation of patients before carotid endarterectomy removes the risk associated with conventional angiography and represents an important advance in the management of carotid stenosis.