Purpose: To establish the minimum dose required for detection of renal artery stenosis using high temporal resolution, contrast enhanced MR angiography (MRA) in a porcine model. Materials and Methods: Surgically created renal artery stenoses were imaged with 3 Tesla MR and digital subtraction angiography (DSA) in 12 swine in this IACUC approved protocol. Gadobutrol was injected intravenously at doses of 0.5, 1, 2, and 4 mL for time-resolved MRA (1.5 x 1.5 mm2 spatial resolution). Region of interest analysis was performed together with stenosis assessment and qualitative evaluation by two blinded readers. Results: Mean signal to noise ratio (SNR) and contrast to noise ratio (CNR) values were statistically significantly less with the 0.5-mL protocol (P < 0.001). There were no statistically significant differences among the other evaluated doses. Both readers found 10/12 cases with the 0.5-mL protocol to be of inadequate diagnostic quality (? = 1.0). All other scans were found to be adequate for diagnosis. Accuracies in distinguishing between mild/insignificant (<50%) and higher grade stenoses (>50%) were comparable among the higher-dose protocols (sensitivities 7393%, specificities 62100%). Conclusion: Renal artery stenosis can be assessed with very low doses ( 0.025 mmol/kg bodyweight) of a high concentration, high relaxivity gadolinium chelate formulation in a swine model, results which are promising with respect to limiting exposure to gadolinium based contrast agents. J. Magn. Reson. Imaging 2012;36:704713. (C) 2012 Wiley Periodicals, Inc.