OBJECTIVES: To assess the cost-effectiveness, from the UK NHS perspective, of magnetic resonance angiography (MRA), duplex ultrasound (DUS) and computed tomography (CT) compared with contrast angiography (CA) in assessing the extent and location of stenosis and subsequently formulating a treatment plan for patients with peripheral arterial disease (PAD). METHODS: A probabilistic decision tree was developed in order to estimate the cost per QALY (in £2004) associated with each diagnostic method for assessment of the whole leg, and the arteries above and below the knee. Input parameters were obtained from a systematic review, other published sources and expert opinion. Lack of data to extrapolate the results to a longer period led to consideration of a 1-year time horizon, therefore discounting was not performed. RESULTS: DUS was the dominant strategy for the assessment of the whole leg, with a cost per QALY of £13,646. MRA appeared to be more cost-effective for assessment of the arteries above the knee, with a cost-per-QALY equal to £8,628 with 2D-TOF MRA, and an incremental cost per additional QALY equal to £37,024 when 2D-TOF MRA was compared to DUS. For below the knee comparisons, results were uncertain, with DUS being more likely to be cost-effective at commonly accepted cost-per-QALY threshold values. CONCLUSIONS: The cost-effectiveness of the diagnostic tests was dependent on the area of the leg being assessed, with DUS being dominant for comparisons of the whole leg and cost-effective for below the knee comparisons, and MRA being cost-effective for above the knee assessments.