Improved artery-vein separation with acceleration-dependent preparation for non-contrast-enhanced magnetic resonance angiography.

PURPOSE: To compare the use of acceleration-dependent and velocity-dependent flow-preparation for non-contrast-enhanced magnetic resonance angiography (NCE-MRA), investigating both image quality and the ability to discriminate between arteries and veins. We develop an acceleration-dependent NCE-MRA method known as acceleration dependent vascular anatomy for non-contrast-enhanced MRA (ADVANCE-MRA). METHODS: Acceleration-dependent and velocity-dependent images were acquired using a constant and pulsatile flow-phantom and from the lower legs of six healthy volunteers and one patient with peripheral vascular disease. The volunteer images were assessed both by quantitative signal measurements and qualitative scoring by a radiologist. RESULTS: In the phantom, acceleration-dependent preparation depicted pulsatile but not constant flow, while velocity-dependent preparation depicted both. In the volunteers and the patient, the velocity-dependent preparation was unable to separate the arterial and venous signals completely, with some overlap of arterial and venous signals for all acquired flow sensitizations whereas the acceleration-dependent preparation gave complete artery-vein separation over a wide range of flow sensitizations. Acceleration-dependent preparation received the best overall qualitative scores for arterial image quality and venous contamination. CONCLUSION: Acceleration-dependent NCE-MRA improves arterial image quality and reduces venous contamination, compared with velocity-dependent NCE-MRA, and warrants further investigation in patients.

Initial clinical evaluation of a non-contrast-enhanced MR angiography method in the distal lower extremities.

PURPOSE: To report the initial experience and diagnostic performance applying a novel flow-dependent non-contrast-enhanced MR angiography (NCE-MRA) method, in patients with suspected peripheral vascular disease, in comparison with established contrast-enhanced MRA (CE-MRA). METHODS: The lower legs of 34 patients were imaged at 1.5 T with both NCE-MRA and CE-MRA. The NCE-MRA method consisted of a cardiac-gated balanced-SSFP sequence with controllable flow-suppression preparation. Flow-suppressed and unsuppressed datasets were subtracted to obtain angiograms. Two experienced radiologists assessed both NCE-MRA and CE-MRA images, first independently and then in consensus to resolve significant disagreements. Signal loss, vessel conspicuity, vascular disease, venous contamination, artifacts, and diagnostic confidence were assessed. RESULTS: Using the CE-MRA as the "gold standard," the per-segment sensitivity and specificity for detection of significant disease were 81.7% and 90.9%, respectively. Mean diagnostic confidence (scale 0-4) was 3.4 for NCE-MRA and 3.9 for CE-MRA. Most vessel segments were well visualized but the popliteal arteries often suffered some technique-related signal loss. CONCLUSIONS: The NCE-MRA method was able to visualize most vessel segments with good or excellent confidence, few artifacts, and excellent background suppression, giving moderate agreement with CE-MRA. However, some segments were poorly visualized, probably due to flow profile distortion in these patients.