OBJECTIVE To determine whether spiral CT angiography allows accurate, quantitative evaluation of anatomic abnormalities, including detection of additional lesions, delineation of plaque morphology, and estimation of degree of internal carotid artery stenosis. METHODS Spiral CT angiography with a maximum intensity projection technique was compared with selective digital subtraction angiography (DSA) in 92 carotid arteries. The category of stenosis was determined according to the North American Symptomatic Carotid Endarterectomy Trial: mild (0% to 29%), moderate (30% to 69%), severe (70% to 99%), and occlusion (100%). RESULTS In 78 (85%) of the 92 cases, spiral CT angiography and selective DSA demonstrated the same degree of stenosis. All occlusions (n = 19) were diagnosed correctly with spiral CT angiography. Spiral CT angiography agreed with selective DSA in the classification of stenosis in 59% of the group with mild stenosis, in 82% of the group with moderate stenosis, and in 90% of the group with severe stenosis. In the groups with mild (n = 13), moderate (n = 9), and severe (n = 27) stenosis, correlation of spiral CT angiography with selective DSA was significant. Calcified plaques were readily diagnosed with the use of spiral CT angiography but delineation of ulcers was poor. Tandem lesions were not visible owing to the limited coverage. CONCLUSIONS Spiral CT angiography is useful for the detection of proximal internal carotid stenoses that are greater than 30%. Depiction of mild stenoses appears to be limited. CT is superior for the detection of calcified plaques but it is not useful for the detection of ulcers.