期刊文献+

胸部CT影像学特征对肺腺癌亚实性结节脏层胸膜侵犯的预测价值 被引量:1

Predictive value of spiral computed tomography features in visceral pleural invasion in pulmonary adenocarcinoma appearing as sub-solid nodules
下载PDF
导出
摘要 目的:探讨胸部CT影像学特征在肺腺癌亚实性结节(SSN)脏层胸膜侵犯(VPI)中的临床预测价值。方法:回顾性收集2016年5月-2021年12月在本院经手术切除和病理确诊为肺腺癌SSN的患者103例。男34例,女69例,年龄25~82岁,平均(58.5±10.2)岁。病理组织学诊断结果VPI阳性组19例(18.45%),VPI阴性组84例(81.55%)。记录患者的年龄、性别等临床资料。所有患者术前均行胸部CT检查。阅读CT影像学征象,包括位置、径线、密度、分叶征、毛刺征及胸膜凹陷征等。结节与胸膜关系(NPR)分为四种亚型:Ⅰ型,1条线相连;Ⅱ型,多条线相连;Ⅲ型,窄基底相贴;Ⅳ型,宽基底相贴。单因素分析比较VPI阳性组和阴性组在临床和CT影像学特征间的差异,将单因素分析差异有统计学意义的指标纳入多因素Logistic回归分析,通过受试者操作特征(ROC)曲线建立预测模型。结果:103例患者包括非实性结节49例(47.57%)和部分实性结节54例(52.43%)。VPI阳性组和阴性组的性别和年龄差异无统计学意义(P=0.883、0.151)。非实性结节VPI发生率为8.16%(4/49),低于部分实性结节27.78%(15/54),差异有统计学意义(P=0.01)。NPR四种亚型发生VPI的比例分别为Ⅰ型6.82%(3/44)、Ⅱ型15.00%(3/20)、Ⅲ型28.00%(7/25)和Ⅳ型42.86%(6/14),差异具有统计学意义(P=0.011)。结节密度和NPR分型是预测肺腺癌SSN的VPI状态的独立危险因素。ROC结果显示,结节密度诊断VPI的曲线下面积(AUC)为0.663,敏感度为78.9%,特异度为53.6%;NPR分型诊断VPI的AUC为0.726,敏感度为68.4%,特异度为69.0%;两者联合诊断VPI的AUC为0.804,敏感度为73.7%,特异度为70.2%。结论:肺腺癌亚实性结节的CT影像学特征有助于脏层胸膜侵犯的术前评估,结节密度和结节与胸膜关系分型是预测脏层胸膜侵犯的独立危险因素。 Objective:To investigate the predictive value of spiral computed tomography(CT)features for visceral pleural invasion(VPI)in pulmonary adenocarcinoma appearing as sub-solid nodules.Methods:A total of 103 patients(34 males and 69 females,age range 25~82 years,mean age 58.5±10.2 years)with sub-solid lung adenocarcinoma diagnosed by surgical resection and pathology in our hospital from May 2016 to December 2021 were retrospectively collected.There were 19 cases(18.45%)in VPI positive group and 84 cases(81.55%)in VPI negative group.Patient's clinical data including age and gender were recorded.All patients underwent chest spiral CT examination before surgery.Spiral CT radiological features as location,size,attenuation,lobulated sharp,speculated sign,pleural indentation sign,vacuole sign and air-bronchogram sign were recorded,respectively.Nodule and pleura relationship(NPR)was subdivided into four subtypes:Ⅰ,connected by one line;Ⅱ,connected by multiple lines;Ⅲ,narrow base;Ⅳ,wide base.Univariate analysis was performed to compare the differences in clinical and radiological parameters between VPI positive and negative groups.Multivariate logistic regression was carried out to analyze the independent risk factors,and receiver operating characteristic(ROC)curve was used to conducted the predictive models.Results:There were 49(47.57%)non-solid nodules and 54(52.43%)partial solid nodules.No significant difference was shown in gender and age between VPI positive and negative groups(P=0.883,P=0.151).The incidence of VPI in non-solid nodules was 8.16%(4/49),which was lower than that of part-solid nodules 27.78%(15/54)(P=0.01).The proportion of VPI in the four NPR subtypes were 6.82%(3/44),15.00%(3/20),28.00%(7/25),42.86%(6/14)for typeⅠ,Ⅱ,ⅢandⅣ,respectively(P=0.011).Nodule attenuation and NPR subtype were independent risk factors in the differential diagnosis of VPI status.ROC results showed that the area under the curve(AUC),sensitivity and specificity were 0.663,78.9%,53.6%and 0.726,68.4%,69.0%in the nodule attenuation model and NPR subtype model,respectively.The combined model constructed by nodule attenuation and NPR subtype showed better diagnostic performance with AUC,sensitivity and specificity were 0.804,73.7%,and 70.2%,respectively.Conclusion:The chest spiral CT radiological features of SSN were helpful for preoperative evaluation of VPI status.Nodule attenuation and NPR subtype were independent risk factors for VPI positive groups.
作者 聂中新 段晓蓓 邝琼莲 区丽琼 寺江烽 谭国强 柳学国 龙晚生 陈相猛 NIE Zhong-xin;DUAN Xiao-bei;KUANG Qiong-lian(Department of Radiology,Jiangmen Central Hospital,Guangdong 529030,China)
出处 《放射学实践》 CSCD 北大核心 2024年第2期195-200,共6页 Radiologic Practice
基金 国家自然科学基金(62176104) 广东省基础与应用基础研究基金(2021A1515220172) 广东省医学科研基金(B2022113)。
关键词 体层摄影术 X线计算机 肺肿瘤 肺腺癌 胸膜侵犯 亚实性肺结节 Tomography,X-ray computed Lung neoplasms Lung adenocarcinoma Pleural invasion Sub-solid pulmonary nodules
  • 相关文献

参考文献8

二级参考文献29

共引文献80

同被引文献11

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部