Assessment of carotid artery stenosis by MR angiography: comparison with x-ray angiography and color-coded Doppler ultrasound.

PURPOSE To compare magnetic resonance angiography (MRA) with duplex Doppler ultrasound (US) and x-ray angiography (XRA) in the evaluation of the carotid bifurcation. METHODS The carotid arteries of 61 patients were studied using MRA, US, or XRA; 31 of the patients underwent all three examinations. MRA included both 2D and 3D time-of-flight sequences. Internal and external carotid artery origins were graded normal, mild, moderate, severe, or critical stenosis, or complete occlusion by each of the three studies. RESULTS Spearman rank correlations of both internal and external carotid artery grades were 0.85 (MRA and XRA), 0.69 (MRA and US), and 0.73 (XRA and US). For internal carotid artery origins only, the correlations were 0.94 (MRA and XRA), 0.85 (MRA and US), and 0.82 (XRA and US). Of discrepancies in internal carotid artery interpretation greater than one grade, seven resulted from US error, three from MRA error, and one from XRA error. A 2-cm partially thrombosed aneurysm detected by US and MRA was missed by XRA. Of 16 possible ulcers on XRA, 11 were noted by MRA, none by US. CONCLUSIONS MRA and XRA are similar in assessment of carotid bifurcation stenosis. MRA, like US, permits direct visualization of plaque. This preliminary study suggests that MRA may be used to clarify equivocal findings of US, or replace XRA in presurgical planning.