Utility of magnetic resonance arteriography for distal lower extremity revascularization

PURPOSE: Magnetic resonance arteriography (MRA) of the lower extremities affords several possible advantages over conventional contrast arteriography (CA). We hypothesized that MRA of the infrageniculate vessels was sufficiently accurate to replace CA before revascularization procedures in patients with limb-threatening ischemia. METHODS: Fifty-three extremities in 49 patients were prospectively evaluated before attempted infrageniculate revascularization procedures with preoperative infrageniculate time-of-flight MRA (cost, $170/study) and standard contrast arteriography (cost, $1310/study) of the aortoiliac and runoff vessels. Independent operative plans were formulated based on the MRA and CA results before the revascularization procedure. Intraoperative, prebypass arteriograms (IOA; cost, $46/study) were obtained in all patients to confirm the adequacy of the distal runoff. The preoperative plans formulated by the results of MRA and CA were compared with the actual procedure performed based on the IOA. All arteriograms (CA, MRA, IOA) were reviewed after the operation by two independent reviewers, and the number of patent vessel segments and those with < 50% stenosis was determined. RESULTS: Revascularization procedures were performed in 44 of 53 extremities (83%), and amputation was performed in nine extremities (17%) because of an absence of a suitable bypass target. The CA and MRA were equally effective in predicting the optimal operative plans as determined from IOA (CA, 42 of 53 [77%] vs MRA, 40 of 53 [75%]; p = 0.79). More patent vessel segments were seen on CA than MRA (reviewer A, 229 vs 174, kappa = 0.32; reviewer B, 321 vs 314, kappa = 0.46); however, a comparable number of segments were seen if the vessels of the foot were excluded. The accuracy (reviewer A, 78% vs 68%, p = 0.003; reviewer B, 75% vs 67%, p = 0.003) and sensitivity (reviewer A, 69% vs 51%, p = 0.001; reviewer B, 68% vs 46%, p = 0.0001) of CA relative to IOA were superior to those of MRA, although the specificity was comparable (reviewer A, 86% vs 90%, p = 0.31; reviewer B, 82% vs 87%, p = 0.52). The combination of MRA and IOA would have resulted in the optimal operative plan in 51 of the 53 cases (96%) and was comparable with CA and IOA (53 of 53; 100%; p = 0.50). Substitution of MRA and IOA for CA and IOA could potentially have saved an estimated $60,420. CONCLUSIONS: The combination of MRA and IOA provides an accurate, cost-efficient strategy for visualization of the infrageniculate vessels before revascularization procedures.