Remy-Jardin M, Remy J, Artaud D, Fribourg M, Duhamel A
Department of Radiology, Hôpital Calmette, Lille, France.
Radiology. 1998 Sep;208(3):761-70. doi: 10.1148/radiology.208.3.9722857.
To evaluate the application of the volume-rendering technique to airway disease.
Seventy-four spiral computed tomographic (CT) examination (1- to 5-mm section thickness; pitch, 1.7-20.0) were performed for known or suspected abnormality of the airways, including (a) benign tracheobronchial stenosis (group 1, n = 47), (b) complex airway lesions (group 2, n = 15), and (c) bronchiectasis (group 3, n - 12). Two radiologists independently compared overlapping transverse CT scans and volume-rendered bronchographic images for detection of airway abnormalities and identification of lesion morphology and extent, with bronchoscopic correlation.
Bronchographic image quality was graded as excellent (73%) or good (24%), whereas motion-related artifacts led to poor image quality in 3% of cases (trapezoid parameters included window width of -700 to -350 HU, triangular trapezoid shape, parenchymal opacification of 70%, unshaded algorithm). Compared with transverse CT scans, volume-rendered images (a) provided supplemental group 3, even, seven), (b) improved confidence in the interpretation of congenital airway abnormalities in six cases (8%) (all from group 2), and (c) corrected interpretive errors in four cases (5%) (group 1, two cases; group 2, two).
Volume-rendered images improved the recognition of mild changes in airway caliber and the understand of complex tracheobronchial anomalies.
评估容积再现技术在气道疾病中的应用。
对已知或疑似气道异常的患者进行了74次螺旋计算机断层扫描（CT）检查（层厚1至5毫米；螺距1.7至20.0），包括（a）良性气管支气管狭窄（第1组，n = 47），（b）复杂气道病变（第2组，n = 15），以及（c）支气管扩张（第3组，n = 12）。两名放射科医生独立比较重叠的横轴CT扫描图像和容积再现支气管造影图像，以检测气道异常并确定病变形态和范围，并与支气管镜检查结果进行对照。
支气管造影图像质量被评为优秀（73%）或良好（24%），而在3%的病例中，与运动相关的伪影导致图像质量较差（梯形参数包括窗宽-700至-350 HU、三角形梯形形状、实质内70%的模糊度、非阴影算法）。与横轴CT扫描相比，容积再现图像（a）在第3组中提供了补充信息（均匀，7例），（b）提高了对6例（8%）先天性气道异常解释的可信度（均来自第2组），并且（c）纠正了4例（5%）的解释错误（第1组，2例；第2组，2例）。
容积再现图像提高了对气道管径轻微变化的识别能力以及对复杂气管支气管异常的理解。
