ABSTRACT
PURPOSE: To assess the clinical value of three-dimensional coronary MR angiography (CMRA) in the detection of significant coronary artery stenosis using conventional X-ray angiography as the standard reference. MATERIAL AND METHODS: Sixty-nine patients underwent X-ray coronary angiography and CMRA because of suspected or previously diagnosed coronary artery disease. MRI was performed with a 1.5-T whole body imaging system using ECG-triggered 3D gradient echo sequence with retrospective navigator echo respiratory gating and fat suppression. RESULTS: A total of 276 coronary artery segments were analyzed. The X-ray coronary angiography was normal in 22 patients. Significant proximal stenoses (exceeding 50%) or occlusions were present in 102 coronary artery segments. In all, 120 stenoses or occlusions were identified in CMRA. Sixteen percent of the coronary artery segments had to be excluded because of poor image quality. The overall sensitivity and specificity for MRA for identification of significant stenosis were 75% and 62%, respectively. CMRA correctly detected 89% of patients with at least one vessel disease, but 6 patients with coronary artery disease would have been missed. CONCLUSIONS: Because of the high data exclusion and false- negative case rate, CMRA with retrospective navigator echo triggering is at present not suitable as a clinical screening method in coronary artery disease.
