Semitransparent Volume-rendering CT Angiography for Lesion Display in Aortoiliac Arteriosclerotic Disease

Abstract

The potential of semitransparent volume-rendering (STVR) computed tomographic (CT) angiography was evaluated for the assessment of hemodynamically significant stenoses in aortoiliac arteriosclerotic disease. In a prospective study, 76 patients (57 men, 19 women; mean age, 70 years) underwent single-detector (n = 26) or multidetector (n = 50) CT angiography of the aortoiliac region. Intraarterial digital subtraction angiography (DSA) was performed in each patient. STVR images with semitransparent display of arterial lumen (opacity, 50%) and vascular calcifications (opacity, 20%), as well as maximum-intensity projection (MIP), frontal/sagittal curved planar reformation (CPR), and MIP/axial studies were independently reviewed for hemodynamically significant lesions (> or =70% cross-sectional area reduction). DSA combined with invasive pressure measurement was used as the standard of reference. Vessel wall calcifications were classified according to a four-point scale (0, not calcified; 1, mildly calcified; 2, moderately calcified; 3, severely calcified). Of the 380 reviewed vessel sections, 28 represented 70%-99% stenoses and 14 represented occlusions. For detecting >/==" BORDER="0">70% lesions, STVR as well as CPR and MIP/axial studies revealed significantly higher specificity (91%-94%), positive predictive value (PPV; 0.62-0.72), and accuracy (90%-94%) than MIP (specificity, 59%; PPV, 0.27; accuracy, 64%; P <.001) in moderately or severely calcified vessel sections. In noncalcified or mildly calcified sections, the specificity of MIP was comparable with the other rendering techniques (96%-98%; P values ranging from.34 to.77). No significant sensitivity differences were noticed among the CT angiography modalities: STVR, 79%; MIP, 88%; CPR, 83%; and MIP/axial, 93% (P values ranging from.12 to.78). Negative predictive values ranged from 0.97 to 0.99 (P values ranging from.14 to 1). Median calcification scores in sections with overestimated lumen narrowings were significantly higher (3, severely calcified) than in sections with correctly graded lumen width (1, mildly calcified) with all CT angiography modalities (P <.05). With use of STVR, visualization of vascular lumen can be improved by rendering highly transparent mural calcifications. Hence, for three-dimensional presentation of aortoiliac arteriosclerotic disease, STVR studies should be preferred to MIPs as supplements to review of axial-source images.