The objective of this study was to assess the cost-effectiveness of magnetic resonance angiography (MRA) imaging for renal artery stenosis (RAS) in people with progressive renal failure (PRF). We created a simulation model to determine the incremental cost-effectiveness of MRA screening in PRF compared with the fallback strategy of not screening. Costs, probabilities, and utilities were estimated from the literature and from institutional data. A three-state Markov model was used to simulate the progression from PRF to end-stage renal disease and death. In our baseline analysis, assuming a sensitivity of 0.85 and a specificity of 0.8 of MRA for RAS, we obtained an incremental cost-effectiveness of MRA screening compared with no screening of $2,214 per quality-adjusted life year saved, which is less than many commonly performed procedures. Under our baseline assumptions, if the receiver-operating characteristic curve of MRA for RAS is better than the chance curve, then MRA screening would be cost-effective. The analysis was most sensitive to assumptions about renal function after correction of RAS and prevalence of RAS, although the results show that MRA remains cost-effective for reasonable ranges of these assumptions. The use of MRA in PRF would be a worthwhile investment of resources in comparison with many currently funded procedures. The expense and morbidity associated with end-stage renal disease make any reasonable way of delaying or preventing the disease worth examining in detail.