Three-dimensional transabdominal ultrasound identification of aortic plaque.

Abstract

Tree-dimensional (3-D) reconstruction of acquired tomographic images in adults has recently been described. With an adaptation of this technique, we performed 3-D reconstruction of transabdominal images of the abdominal aorta to test the hypotheses that 3-D reconstruction of the abdominal aorta is feasible and that 3-D images have incremental value over 2-D in the detection of atheromatous plaque. Twenty-one patients undergoing contrast aortography (Aogram) for clinical indications (1 abdominal aorta (AA) aneurysm, 5 peripheral vascular disease, 1 renal artery stenosis, 14 renal donors) were studied using a 5-MHz annular array probe fitted to a mechanical registration device. In 13 of 21 patients, adequate 2-D ultrasound slices were acquired around a 180 degrees rotation and stored as a volumetric data set using a dedicated computer and 3-D images were reconstructed off-line. Three-dimensional and planar images were blindly compared with Aograms using the following scale: grade 1, normal; grade 2, increased echodensity of the intimal surface; grade 3, local intimal thickening and/or luminal irregularity; and grade 4, protruding mass. Analogous 3-D images were produced in all 13 patients with branching vessels visible in 3 of 13. In 10 patients, the Aogram was interpreted as normal. Compared with Aogram, blindly interpreted 3-D images were compared and correctly identified normal AA in 8 of 10 and atherosclerotic plaque (grade 3 or 4) in 2 of 3. Discordant results were present in 2 of 10 normal aortas and 1 of 3 disease aortas. When 2-D (planar) images were compared with Aograms, 8 of 10 identified normal AA and 3 of 3 aortas with grade 3 or 4 plaque. Thus, in 2 patients, 3-D and planar images suggested atherosclerotic changes not seen by Aogram. Transabdominal 3-D imaging of the abdominal aorta is a feasible technique. Early data suggest that 3-D imaging may distinguish normal from moderate to severe disease, but currently has no demonstrable incremental value over conventional 2-D images. These early results in a small number of patients suggest that this promising technique warrants further evaluation.