多发肺结节主病灶良恶性判断的多因素分析

Multivariate Analysis of the Judgment of Benign and Malignant Main Lesions of Multiple Pulmonary Nodules

摘要: 目的：分析多发肺结节主病灶6 mm ≤ 直径 ≤ 20 mm患者的临床资料、影像学特征及病理特点，探讨影响其良恶性的独立危险因素并建立预测模型，帮助临床医师早期识别多发肺结节患者主病灶的良恶性，尽可能及早并准确地作出诊断和治疗。研究方法：回顾性收集西安医学院第一附属医院心胸外科2021年1月至2023年9月期间经手术治疗且病理明确的多发肺结节主病灶6 mm ≤ 直径 ≤ 20 mm患者的临床资料(年龄、性别、吸烟史、肿瘤个人或家族史)、影像学特征(大小、结节密度、生长位置、影像特征)及病理资料，根据病理结果将患者分为良性组和恶性组，通过单因素、多因素分析筛选出影响多发肺结节主病灶6 mm ≤ 直径 ≤ 20 mm良恶性的独立危险因素，并建立预测模型分析其预测效能。结果：本研究共纳入患者87例，其中良性组27人，恶性组60人，女性39例(44.8%)，男性48例(55.2%)，单因素分析结果表明，年龄、吸烟史、家族史、临床症状、CT值、结节位置、结节性质、结节数量、结节边界、毛刺征、分叶征、血管纠集征、支气管截断征、胸膜凹陷征、钙化均为影响主病灶良恶性的因素(P < 0.05)，多因素Logistic回归分析表明，年龄(OR = 1.086, P = 0.042, 95％CI: 1.003~1.176)、亚实性结节(OR = 7.307, P = 0.046, 95%CI: 1.032~51.719)、毛刺征(OR = 5.860, P = 0.048, 95%CI: 1.019~33.685)为影响主病灶良恶性的独立危险因素，进一步建立预测模型表明其对主病灶恶性组的预测效能(AUC: 0.885, P = 0.000, 95%CI: 0.795~0.978，灵敏度：77.8%，特异度：93.3%)优于年龄(AUC: 0.644, P = 0.032, 95%CI: 0.518~0.771，灵敏度：77.8%，特异度：51.7%)、毛刺征(AUC: 0.529, P = 0.070, 95%CI: 0.396~0.661，灵敏度：40.7%，特异度：65.0%)、亚实性结节(AUC: 0.826, P = 0.000, 95%CI: 0.733~0.919，灵敏度：85.2%，特异度：78.3%)等指标，预测效能更好。结论：年龄、结节性质、毛刺征与主病灶良恶性密切关系，年龄越大、亚实性结节、有毛刺征的主病灶肺结节恶性的可能性更大，分别为良性肺结节的1.086、7.037、5.860倍。

Abstract: Objective: The clinical data, imaging characteristics and pathological characteristics of patients with the main lesions of multiple pulmonary nodules with 6 mm ≤ diameter ≤ 20 mm were analyzed, the independent risk factors affecting the benign and malignant lesions were explored and the prediction model was established to help clinicians identify the benign and malignant main lesions of patients with multiple pulmonary nodules early, and make diagnosis and treatment as early and accurate as possible. Study Method: A retrospective review was conducted of patients with the main lesions of multiple pulmonary nodules with 6 mm ≤ diameter ≤ 20 mm who underwent treatment and had a confirmed pathological diagnosis at the Department of Cardiothoracic Surgery of The First Affiliated Hospital of Xi’an Medical College from January 2021 to September 2023, and their clinical data (age, gender, smoking history, personal or family history of tumors), imaging characteristics (size, nodule density, growth location, imaging signs) and pathological data were included. According to the pathological results, patients were divided into benign and malignant groups. Univariate and multivariate analyses were conducted to identify independent risk factors affecting the benign and malignant nature of the main lesions of multiple pulmonary nodules with 6 mm ≤ diameter ≤ 20 mm, and a predictive model was established to analyze their predictive efficacy. Results: A total of 87 patients were included in this study, including 27 in the benign group and 60 in the malignant group. There were 39 females (44.8%) and 48 males (55.2%). The results of univariate analysis showed that age, smoking history, family history, clinical symptoms, CT value, nodule location, nodule nature, nodule number, nodule boundary, burr, leaf segmentation, vascular collection, bronchial resection, pleural depression, and calcification were all factors affecting the benign and malignant nature of the main lesions (P < 0.05). Multivariate Logistic regression analysis indicated that age (OR = 1.086, P = 0.042, 95%CI:1.003~1.176), subsolid nodules (OR = 7.307, P = 0.046, 95%CI: 1.032~51.719), and burr sign (OR = 5.860, P = 0.048, 95%CI: 1.019~33.685) were identified as independent risk factors affecting the benign and malignant nature of the major lesions. Further establishment of prediction model showed its predictive efficacy in the malignant group of the major lesions (AUC: 0.885, P = 0.000, 95%CI: 0.795~0.978, Sensitivity: 77.8%, Specificity: 93.3%) was better than age (AUC: 0.644, P = 0.032, 95%CI: 0.518~0.771, Sensitivity: 77.8%, Specificity: 51.7%), burr sign (AUC: 0.529, P = 0.070, 95%CI: 0.396~0.661, Sensitivity: 40.7%, Specific degree: 65.0%), and subsolid nodules (AUC: 0.826, P = 0.000, 95%CI: 0.733~0.919, Sensitivity: 85.2%, Specific degree: 78.3%), showing better predictive efficacy. Conclusion: Age, nodule nature, and burr signs are closely related to the benign or malignant nature of major lesions. Older age, subsolid nodule, and burr signs are more likely to be malignant in pulmonary nodules of major lesions, which are 1.086,7.037 and 5.860 times that of benign pulmonary nodules, respectively.