Detection of accessory renal arteries with virtual vascular endoscopy of the aorta

Purpose: To evaluate the diagnostic accuracy of virtual vascular endoscopy (VVE) in the detection of accessory renal arteries. Methods: We retrospectively reviewed the CT angiography data sets of 67 pat ients (29 male and 38 female; age range 17-72 years, mean age 53 years) ima ged for the study of the renal arteries, and affected by renovascular hyper tension. All patients also had intraarterial digital subtraction angiograph y (DSA). CT angiography data sets were processed to obtain maximum intensit y projection (MIP) and surface-rendered VVE of the aorta. Axial images, MIP , and VVE were evaluated separately and in combination in the detection of accessory renal arteries. Their results in terms of sensitivity and specifi city were then compared with DSA. Results: Axial images had a sensitivity of 88% and specificity of 94% for a ccessory renal artery detection, MIP had a sensitivity of 88% and specifici ty of 98%, and VVE had a sensitivity of 63% and a specificity of 88% (p < 0 .05 vs DSA), but these increased to 88% and 98% respectively if endoscopic views were integrated with the other display techniques. Conclusion: VVE based on surface rendering does not add substantial benefit s to CTY angiography; by contrast MIP is the most accurate display techniqu e for the detection of accessory renal arteries.