1998 年 11 巻 3 号 p. 98-102
A comparison of three-dimensional time-of-flight magnetic resonance angiography (MRA) with B-mode and colorcoded Doppler ultrasonography (US) was made in 56 carotid arteries in 28 individuals (male 24 and female 4, age 64.5±7.2 years) . The degree of stenosis of the carotid arteries was classified as normal: 0-19%, mild: 20-39%, moderate: 40-59%, severe: 60-79%, critical: 80-99%, and occlusion: 100%. MRA showed two false-positive cases, one false-negative case and one undergraded case. US showed no false-positive and no false-negative case. MRA was relatively insensitive for detection of mild stenosis, but showed good correlation with US in detection of severe, critical stenosis or occlusion. Comparison of US with conventional angiography demonstrated 100% accuracy for carotid stenosis (sensitivity 100%, specificity 100% for stenosis) . Comparison of MRA with US demonstrated 95% accuracy for carotid stenosis (sensitivity 98%, specificity 96% for stenosis) . It was concluded, that the diagnostic pitfalls of MRA are artifacts caused by turbulent nonlaminar flow, flow separation, reversal of flow and motion. US is useful for interpreting intraplaque morphology and for detecting turbulence or vortex flow using color Doppler flow imaging. However, the major limitation of US is the difficulty in detecting plaques in the high cervical carotid area. This study suggests that combined use of MRA and US is accurate for detection and evaluation of carotid stenosis in patients with cerebrovascular symptoms.