Diagnostic impact of four postprocessing techniques in evaluating contrast-enhanced three-dimensional MR angiography

The purpose of this study was to determine the added diagnostic value of various three-dimensional (3D) data viewing techniques when analyzing contrast-enhanced 3D MR angiography. Twenty patients (mean age, 62 years) with symptomatic peripheral vascular disease were assessed with breath-hold, contrast-enhanced 3D MR angiography and catheter angiography, which served as the standard of reference. After an initial interpretation of the 3D MR angiographic data sets based only on standardized maximum intensity projections (MIP), the diagnostic gain of the stepwise addition of interactive multiplanar reformations, shaded-surface displays (SSD), and virtual intraarterial endoscopy (VIE) images was calculated. Time required for each step of postprocessing was measured. Pathologic changes were revealed by catheter angiography in 60 vascular segments (50 severe stenoses, seven aneurysms, and three occlusions). The average postprocessing times were MIP, 8 min (range, 5-12 min); multiplanar reformations, 9 min (range, 3-11 min); SSD, 15 min (range, 8-25 min); and VIE, 40 min (range, 18-63 min). Addition of multiplanar reformations to MIPs resulted in the greatest gain of diagnostic accuracy, from 92% to 96%, and diagnostic confidence. When analysis was based on all four techniques, receiver operating characteristic curve analysis revealed only minimal improvements in diagnostic confidence, whereas diagnostic accuracy remained unchanged at 96%. Accurate and time-effective analysis of contrast-enhanced 3D MR angiography should be based on MIP algorithms and multiplanar reformations. Additional evaluation with VIE or SSD techniques is time-consuming and provides little diagnostic gain.