Tang Jizheng, Liu Chunquan, Wang Peihao, Cui Yong
Department of Thoracic Surgery, Beijing Friendship Hospital of Capital Medical University, Beijing 100050, China.
Zhongguo Fei Ai Za Zhi. 2021 Feb 20;24(2):94-98. doi: 10.3779/j.issn.1009-3419.2021.102.05. Epub 2021 Jan 29.
Preoperative diagnosis and differential diagnosis of small solid pulmonary nodules are very difficult. Computed tomography (CT), as a common method for lung cancer screening, is widely used in clinical practice. The aim of this study was to analyze the clinical data of patients with malignant pulmonary nodules and intrapulmonary lymph nodes in the clinical diagnosis and treatment of <1 cm solid pulmonary nodules, so as to provide reference for the differentiation of the two.
Patients with solid pulmonary nodules who underwent surgery from June 2017 to June 2020 were analyzed retrospectively. The clinical data of 145 nodules (lung adenocarcinoma 60, lung carcinoid 2, malignant mesothelioma 1, sarcomatoid carcinoma 1, lymph node 81) were collected and finally divided into two groups: lung adenocarcinoma and intrapulmonary lymph nodes, and their clinical data were statistically analyzed. According to the results of univariate analysis (χ² test, t test), the variables with statistical differences were selected and included in Logistic regression multivariate analysis. The predictive variables were determined and the receiver operating characteristic (ROC) curve was drawn to get the area under the curve (AUC) value of the area under the curve.
Logistic regression analysis showed that the longest diameter, Max CT value, lobulation sign and spiculation sign were important indicators for distinguishing lung adenocarcinoma from intrapulmonary lymph nodes, and the risk ratios were 106.645 (95%CI: 3.828-2,971.220, P<0.01), 0.980 (95%CI: 0.969-0.991, P<0.01), 3.550 (95%CI: 1.299-9.701, P=0.01), 3.618 (95%CI: 1.288-10.163, P=0.02). According to the results of Logistic regression analysis, the prediction model is determined, the ROC curve is drawn, and the AUC value under the curve is calculated to be 0.877 (95%CI: 0.821-0.933, P<0.01).
For <1 cm solid pulmonary nodules, among many factors, the longest diameter, Max CT value, lobulation sign and spiculation sign are more important in distinguishing malignant pulmonary nodules from intrapulmonary lymph nodes.
小实性肺结节的术前诊断及鉴别诊断极具难度。计算机断层扫描（CT）作为肺癌筛查的常用方法，在临床实践中被广泛应用。本研究旨在分析直径＜1cm实性肺结节临床诊治中恶性肺结节及肺内淋巴结患者的临床资料，为两者的鉴别提供参考。
回顾性分析2017年6月至2020年6月接受手术的实性肺结节患者。收集145个结节（肺腺癌60个、肺类癌2个、恶性间皮瘤1个、肉瘤样癌1个、淋巴结81个）的临床资料，最终分为肺腺癌组和肺内淋巴结组，并对其临床资料进行统计学分析。根据单因素分析（χ²检验、t检验）结果，选取具有统计学差异的变量纳入Logistic回归多因素分析。确定预测变量并绘制受试者工作特征（ROC）曲线，得出曲线下面积（AUC）值。
Logistic回归分析显示，最大直径、最大CT值、分叶征及毛刺征是鉴别肺腺癌与肺内淋巴结的重要指标，风险比分别为106.645（95%CI：3.828 - 2971.220，P＜0.01）、0.980（95%CI：0.969 - 0.991，P＜0.01）、3.550（95%CI：1.299 - 9.701，P = 0.01）、3.618（95%CI：1.288 - 10.163，P = 0.02）。根据Logistic回归分析结果确定预测模型，绘制ROC曲线，计算曲线下AUC值为0.877（95%CI：0.821 - 0.933，P＜0.01）。
对于直径＜1cm的实性肺结节，在诸多因素中，最大直径、最大CT值、分叶征及毛刺征在鉴别恶性肺结节与肺内淋巴结方面更为重要。
Zotero 插件