Proximal great vessels of aortic arch: comparison of three-dimensional gadolinium-enhanced MR angiography and digital subtraction angiography.

Abstract

To prospectively compare dynamic three-dimensional (3D) gadolinium-enhanced magnetic resonance (MR) angiography and digital subtraction angiography (DSA) for the detection of ostial stenosis of the craniocervical vessels. Thirty-three patients with carotid stenosis of more than 50% at sonography prospectively underwent both MR angiography and DSA. The overall quality of each DSA and MR angiographic study was analyzed. For each craniocervical vessel (brachiocephalic, common carotid, subclavian, and vertebral arteries) (n = 231), ostial stenosis was graded as follows: normal, mild (<50%), moderate to severe (>50%), or occlusion. MR angiographic and DSA results were compared by means of the Spearman rank correlation coefficient (Rs). The overall diagnostic quality of MR angiography was excellent or adequate. Three studies were inadequate because of a poor signal-to-noise ratio (13 of 231 arteries) or a coverage error (five of 231 arteries). Findings at MR angiography and DSA agreed on the degree of stenosis (Rs = 0.82, P <.001). No cases of stenosis of more than 50% were missed at MR angiography. However, some discrepancies were noted between vertebral arteries and the other craniocervical vessels. The sensitivity and specificity for stenosis of more than 50% in other craniocervical vessels were 100% and 98%, respectively. The sensitivity and specificity for stenosis of more than 50% in the vertebral arteries were 100% and 85%, respectively. Findings at MR angiography tended to result in overestimation of the degree of ostial stenosis, especially in vertebral arteries (10 [15%] of 66 arteries). MR angiography is useful to rule out ostial stenosis of the craniocervical vessels. MR angiography is an adequate diagnostic tool for ostial stenosis, except in the vertebral artery.