摘要
目的在胸部CT扫描中,肺部纯磨玻璃密度结节(pure ground glass nodule,pGGN)十分常见,且各期临床处理方式不同。本研究探讨pGGN良恶性与高分辨率CT表现的相关性。方法选取2012-05-01-2014-09-01山东省肿瘤医院收治的59例肺pGGN患者作为研究对象,共59个pGGN。由3名放射科医生回顾分析pGGN的HRCT特征,采用t检验、χ2检验、Fisher确切概率法和Logistic回归分析等统计学方法,分析其与侵袭性之间相关性。依据组织病理学结果,将pGGN的侵袭性分为浸润前病变(包括不典型腺瘤样增生及原位癌)与浸润性病变(包括微浸润性腺癌及浸润性腺癌)。应用ROC曲线分析pGGN直径诊断效能。结果 pGGN毛刺征、边缘分叶、血管增粗、血管牵拉征、密度及胸膜凹陷征在浸润前病变与浸润性病变两组之间差异具有统计学意义,均P值<0.05;充气支气管和形态差异无统计学意义,均P值>0.05;Logistic多因素回归分析显示,边缘分叶、血管牵拉征和血管增粗为鉴别肺腺癌浸润前病变和浸润性病变的独立影响因素,均P值<0.05。ROC曲线分析显示,鉴别浸润前病变与浸润性病变病灶大小的最佳界值为0.945cm,曲线下面积为0.688,诊断的敏感度为95.10%,特异度为44.40%。结论 CT特征可以预测pGGN的良恶性,边缘分叶、血管牵拉征、血管增粗和直径>0.945cm在预测pGGN侵袭性方面具有重要意义。
OBJECTIVE To analyze the correlation between the invasion ability and the high resolution computer tomography（HRCT） performance of the persistence of pure ground glass nodule（pGGN）. METHODS A total of 59 pa- tients （Shandong Cancer Hospital,2012-05-01--2014-09-01） were enrolled in this study. Three radiologists reviewed the characteristics of these HRCTs of pGGN. According to histopathological results, pGGN invasion was divided into pre-in- vasive lesions （including atypical adenomatous hyperplasia and carcinoma in situ） and invasive lesions （including micro-in- vasive adenocarcinoma and invasive adenocarcinoma）. The diagnostic efficacy of pGGN diameter was analyzed by receiver operating characteristie（ROC） curve. RESULTS The pGGN burr sign, margin lobulation, thickening of blood vessels, vascular traction sign,density and pleural depressions between the pre-invasive lesions and invasive lesions had significant differences （P〉0.05）, while the pneumatic bronchus had no significant differences. Logistic multivariate regression analy- sis showed that margin lobulation,vascular traction and thickening of blood vessels were the independent influencing fac tors of lung lesions and invasive lesions （P〈0.05）. ROC curve analysis showed that the best cutoff value of pre-invasive lesion and invasive lesions was 0. 945 cm, the area under the curve was 0. 688, the sensitivity of diagnosis was 95.10% and the specificity was 44.40%. CONCLUSION CT features can be used to predict the benign and malignant pGGN. Margin lobulation, vascular trauma, thickening of blood vessels and diameter greater than 0. 945 cm are important in predicting the invasiveness of pGGN.
作者
范明新
步玉兰
张学成
戚元刚
FAN Ming-xin;BU Yu-lan;ZHANG Xue-cheng;QI Yuan-gang(Department of Radiology,A f filated Hospital of Shandong Academy of Medical Sciences,Jinan 250031,P.R.China;Department of Radiology,Affiliated Hospital of Wei fang Medical University,Wei fang 261031,P.R.China;Department of Radiology,Shandong Cancer Hospital Affiliated to Shandong University,Shandong Academy of Medical Sciences,J inan 250114,P.R.China)
出处
《中华肿瘤防治杂志》
CAS
北大核心
2018年第15期1094-1097,共4页
Chinese Journal of Cancer Prevention and Treatment
关键词
肺肿瘤
肺腺癌
计算机体层摄影
病理学
lung neoplasms
adenocarcinoma
computed tomography
pathology
