Assessment of gadolinium-enhanced time-resolved three-dimensional MR angiography for evaluating renal artery stenosis.

OBJECTIVE The purpose of this study was to assess the image quality of gadolinium-enhanced time-resolved three-dimensional (3D) MR angiography and to evaluate its accuracy in revealing renal artery stenosis. SUBJECTS AND METHODS Thirty-nine patients underwent MR angiography using an ultrafast 3D Fourier transform spoiled gradient-recalled acquisition in the steady state (TR/TE range, 2.6/0.7--0.8). Five seconds after administration of 15--20 mL gadodiamide hydrate, four or five consecutive data sets with imaging times of 7.0--7.6 sec were acquired during a single breath-hold. A timing examination was not performed. Image quality was assessed using quantitative analysis (signal-to-noise, contrast-to-noise, and venous-to-arterial enhancement ratios) and qualitative analysis (presence of venous overlap, presence of artifacts, and degree of renal arterial enhancement). MR angiography depiction of the renal artery stenosis was evaluated using conventional angiography as the standard of reference. RESULTS On the best arterial phase, average aortic signal-to-noise ratio (+/-SD) was 74.5 +/- 24.4, aorta-to--inferior vena cava contrast-to-noise ratio was 70.8 +/- 23.4, and inferior vena cava--to-aorta venous-to-arterial enhancement ratio was 0.03 +/- 0.04. No venous overlap was seen in 38 of 39 patients. Substantial enhancement of renal arteries was seen in all patients without any noticeable artifacts. MR angiography correctly depicted the degree of stenosis in 44 of 47 normal arteries, 13 of 16 mildly stenotic arteries, five of five moderately stenotic arteries, three of four severely stenotic arteries, and one of one occluded artery. Sensitivity and specificity for revealing greater than 50% stenosis was 100%. CONCLUSION Time-resolved 3D MR angiography can provide high-quality arteriograms. Its performance in revealing renal artery stenosis is comparable with that of conventional angiography.