Renal artery stenosis: Evaluation with conventional angiography versus gadolinium-enhanced MR angiography

PURPOSE: To evaluate the interobserver and intermodality variablity of conv entional alangiography and gadolinium-enhanced magnetic resonance (MR) angi ography in the assessment of renal artery stenosis. MATERIALS AND METHODS: Fifty-four patients underwent conventional angiograp hy and gadolinium-enhanced three-dimensional gradient-echo MR angiography, Three angiographers blinded to each other's interpretations and the MR angi ographic findings assessed the conventional angiograms for renal artery ste nosis. Similarly, three blinded MR imagers evaluated the MR angiograms. RESULTS: Interobserver variability for the degree of renal artery stenosis in the 107 kidneys evaluated was not significantly different between the tw o modalities. The mean SD of the degree of stenosis was 6.9% at MR angiogra phy versus 7.5% at conventional angiography (alpha less than or equal to .0 5, P >.05). In 70 kidneys (65%) the average degree of stenosis reported by the readers for the two modalities differed by 10% or less. In 22 cases (21 %), the degree of stenosis was overestimated with MR angiography by more th an 10% relative to the results of conventional angiography. In 15 cases (14 %), the degree of stenosis was underestimated with MR angiography by more t han 10%. CONCLUSION: Gadolinium-enhanced MR angiography permits evaluation of renal artery stenosis with an interobserver variability comparable with that of c onventional angiography.