Sverzellati N, Calabrò E, Chetta A, Concari G, Larici A R, Mereu M, Cobelli R, De Filippo M, Zompatori M
Section of Radiology, Department of Clinical Sciences, University of Parma, Barbieri-Ospedale Maggiore di Parma, Via Gramsci 14, Parma, Italy.
Radiol Med. 2007 Dec;112(8):1160-72. doi: 10.1007/s11547-007-0213-x. Epub 2007 Dec 13.
The aim of this study was to assess the accuracy of some computed tomography (CT) quantitative indices (histogram features, ranges of density and one novel volumetric index) in the discrimination between normals and patients affected by lung fibrosis, and to compare their morphologic-functional relationship with the visual score one.
We analysed thin-section CTs and pulmonary function tests (PFTs) of six healthy subjects and 31 patients affected by lung fibrosis, including 17 with a usual interstitial pneumonia pattern (UIP group), and 14 with a predominant pattern of ground-glass opacities without honeycombing (non-UIP group). Presence and extent of various CT findings were assessed by the visual score as well as by CT computer indices.
Together with the histogram features, fibrosis ratio (defined as the ratio of nonfibrotic CT lung volume divided by total CT lung volume) contributed to objectively differentiate fibrotic lungs from normal lungs. The range of density 700 to 400 HU showed the greatest degree of correlation with physiologic abnormality in the non-UIP group. In the UIP group, the lone visual score provided prediction of functional impairment.
The visual score is still the main radiological method of quantifying the extent of abnormalities in patients with UIP, whilst the range of density 700 to 400 HU can be helpfully applied in a predominant pattern of ground-glass and reticular opacities without honeycombing.
本研究旨在评估一些计算机断层扫描（CT）定量指标（直方图特征、密度范围和一种新的体积指数）在鉴别正常人与肺纤维化患者方面的准确性，并将它们的形态-功能关系与视觉评分进行比较。
我们分析了6名健康受试者和31名肺纤维化患者的薄层CT及肺功能测试（PFT）结果，其中包括17例具有普通间质性肺炎模式的患者（UIP组）和14例以磨玻璃影为主且无蜂窝状改变的患者（非UIP组）。通过视觉评分以及CT计算机指标评估各种CT表现的存在情况和范围。
除直方图特征外，纤维化比率（定义为非纤维化CT肺体积与总CT肺体积之比）有助于客观地区分纤维化肺与正常肺。在非UIP组中，700至400 HU的密度范围与生理异常的相关性最高。在UIP组中，仅视觉评分就能预测功能损害。
视觉评分仍是量化UIP患者异常程度的主要影像学方法，而700至400 HU的密度范围可有效地应用于以磨玻璃影和网状影为主且无蜂窝状改变的情况。
