双能量CT预测下咽鳞状细胞癌隐匿性淋巴结转移  

Dual energy CT for the prediction of occult lymph node metastases in cN0 hypopharyngeal squamous cell carcinoma

作　　者：应海峰 陈炜越 李程超 夏水伟 卢陈英[1] 纪建松[1] YING Haifeng;CHEN Weiyue;LI Chengchao;XIA Shuiwei;LU Chenying;JI Jiansong(Department of Radiology,Zhejiang Key Laboratory of Imaging and Interventional Medicine,the Fifth Affiliated Hospital of Wenzhou Medical University,Lishui Central Hospital,Lishui 323000,China)

机构地区：[1]温州医科大学附属第五医院(丽水市中心医院)放射科全省影像与介入医学重点实验室,浙江丽水323000

出　　处：《温州医科大学学报》2024年第12期1012-1018,共7页Journal of Wenzhou Medical University

基　　金：浙江省卫生健康科技计划项目(2024KY568)。

摘　　要：目的:探讨双能量CT在预测下咽鳞状细胞癌(HPSCC)隐匿性淋巴结转移(OLNM)中的应用价值。方法:回顾性收集2019年1月至2023年4月期间在温州医科大学附属第五医院经病理证实的c N0期HPSCC患者89例,根据颈淋巴结清扫结果,分为有OLNM组和无OLNM组。在动、静脉双能量CT图像上测量病灶在40、50、60、70 ke V时的CT值(表示为CT_(40ke V)、CT_(50ke V)、CT_(60ke V)、CT_(70ke V))、碘浓度(IC)、有效原子序数(Zeff),并计算标准化碘浓度(NIC)及能谱曲线斜率(λ_(Hu));同时收集临床病理资料。使用最小绝对值收敛和选择算子(LASSO)回归筛选c N0期HPSCC患者OLNM的潜在影响因素,随后代入多因素Logistic回归分析中分别建立临床模型、双能量CT模型和联合模型,并绘制列线图。根据ROC曲线的AUC、灵敏度、特异度和准确度评价预测模型的性能。结果:分化程度和浸润深度在两组之间差异有统计学意义(均P<0.05)。OLNM组动脉期CT_(40ke V)、CT_(50ke V)、CT_(60ke V)、IC、NIC、Zeff和λ_(Hu)均高于无OLNM组(均P<0.05)。经LASSO筛选获得6个潜在影响因素,包括分化程度、浸润深度、CT_(40ke V)、NIC、Zeff、λ_(Hu)。ROC曲线分析显示,临床模型的AUC、敏感度、特异度和准确度分别为0.733、87.88%、46.43%、56.27%;双能量CT模型的AUC、敏感度、特异度和准确度分别为0.816、81.82%、73.21%、76.18%;列线图的AUC、敏感度、特异度和准确度分别为0.891、81.82%、82.14%、82.02%。结论:双能量CT动脉期多个参数对HPSCC患者OLNM均有一定的预测价值,进一步结合临床病理特征建立的列线图模型具有更高的诊断性能。Objective:The purpose of this study was to investigate the application value of dual-energy CT in predicting occult lymph node metastasis(OLNM)in hypopharyngeal squamous cell carcinoma(HPSCC).Methods:A total of 89 patients with cN0 HPSCC confirmed by pathology in the Fifth Affiliated Hospital of Wenzhou Medical University from January 2019 to April 2023 were retrospectively collected,who were divided into OLNM and non-OLNM groups according to the results of neck lymph node dissection.The CT values of the lesions at 40,50,60,and 70 keV(referred to as CT_(40 keV),CT_(50 keV),CT_(60 keV),and CT_(70 keV),respectively),iodine concentration(IC),and effective atomic number(Zeff)were measured on arterial and venous dual-energy CT images.Standardized iodine concentration(NIC)value and spectral curve slope(λ_(Hu))were calculated.Pathological and clinical data were simultaneously obtained.The least absolute shrinkage and selection operator regression algorithm(LASSO)was used to screen potential influencing factors for OLNM in cN0 HPSCC patients.Subsequently,the clinical model,dual-energy CT model,and combined model were established by multivariate logistic regression analysis,and the nomogram was drawn.The performance of the prediction models was evaluated based on the area under the curve(AUC)of ROC curve,sensitivity,specificity,and accuracy.Results:There was significant difference in the degree of differentiation and infiltration depth between the two groups of patients(all P<0.05).The arterial phase CT_(40 keV),CT_(50 keV),CT_(60 keV),IC,NIC,Zeff,and λ_(Hu) in the OLNM group were higher than those in the non-OLNM group(all P<0.05).Six potential influencing factors were obtained by LASSO screening,including degree of differentiation,infiltration depth,CT_(40 keV),NIC,Zeff,and λ_(Hu).ROC analysis showed that the AUC,sensitivity,specificity,and accuracy of the clinical model were 0.733,87.88%,46.43%,and 56.27%,respectively;the AUC,sensitivity,specificity,and accuracy of the dual energy CT model were 0.816,81.82%,73.21%,a

关 键 词：下咽癌 鳞状细胞癌 双能量CT成像 隐匿性淋巴结 

分 类 号：R445.3[医药卫生—影像医学与核医学]