The aim of this study was to determine the potential of contrast-enhanced magnetic resonance (MR) angiography in the evaluation of peripheral bypass grafts. Digital subtraction angiography (DSA) served as a standard of reference. Thirty-five patients with previous bypass graft surgery underwent DSA and contrast-enhanced MR angiography within 2 weeks. MR angiography was performed using a three-dimensional fast gradient-echo sequence after administration of gadopentetate dimeglumine. Every leg was divided into 11 segments and scored in five categories of stenosis. MR angiography findings were compared with those of DSA. A total of 38 bypass grafts and 454 segments in 27 patients were included in the evaluation. In 33 (87%) bypass grafts stenosis grading with both methods corresponded, and in 5 (13%) cases stenosis was overestimated on MR angiography. Agreement in detection of hemodynamically significant stenosis (stenosis =50%) was 94.7% with a sensitivity of 100% and a specificity of 91.3%. In 340 (83.0%) vascular segments there was conformity in graduation, in 69 (16.88%), there was a difference of one or more grades on MR angiograms. Forty-four segments (9.6%) were not assessable due to technical limitations. Contrast-enhanced MR angiography is an useful noninvasive tool in the detection of failing peripheral vascular bypass grafts.