【摘要】 目的 通过分析Ⅰ期肺腺癌中气道播散(STAS)的高分辨率CT特征及临床资料，建立个体化预测I期肺腺癌患者气道播散的Nomogram模型。方法 本研究回顾性评估2018年12月到2023年2月共421例Ⅰ期肺腺癌手术切除患者STAS表达状态，分为STAS阴性组和STAS阳性组。采用单因素、多因素logistic回归分析CT特征及临床因素(性别、年龄、大小、CT值、位置、分叶征、毛刺征、胸膜牵拉征、肿瘤成分、支气管充气征、微血管征、月牙征、空泡征、瘤肺边界)，筛选出I期肺腺癌患者STAS表达的重要危险因素，并建立Nomogram预测模型，并应用ROC曲线评估Nomogram模型的预测效能。结果 最终纳入421例患者，STAS阴性225例，男79例，女146例，年龄60(50.5～70岁)，STAS阳性166例，男89例，女77例，年龄62(55.0～69.75岁)。经单因素及多因素logistic回归分析发现，CT值、分叶征、胸膜牵拉征、月牙征、空泡征是预测STAS阳性的独立危险因素，其OR值(95%CI)分别为1.00(1.00～1.01)、24.81(7.80～78.9)、0.29(0.10～0.84)、0.18(0.06～0.51)、0.17(0.07～0.44)，根据上述5个独立危险因素构建Nomogram模型的曲线下面积为0.918(95%CI 0.895～0.948)，灵敏度为86.1%，特异性为85.3%。结论 本研究基于高分辨率CT特征及临床因素构建的Nomogram模型，可以辅助临床对Ⅰ期肺癌STAS患者的精准诊断及选择合适的治疗、管理策略。更多还原

【Abstract】 Objective By analyzing the high-resolution CT features and clinical data of STAS in stageⅠlung adenocarcinoma, a Nomogram model for individualized prediction of STAS in stage I lung adenocarcinoma patients was established. Methods This study retrospectively evaluated the STAS expression status of 421 patients with stageⅠlung adenocarcinoma who were surgically resected from December 2018 to February 2023, and were divided into STAS negative group and STAS positive group. Unifactorial and multivariate logistic regression were used to analyze CT features and clinical factors(gender, age, size, CT value, location, leaf sign, burr sign, pleural stretch sign, tumor component, air bronchogram, tumor vessel sign, crescent sign, vacuolar sign, tumor lung boundary), and screen out important risk factors for STAS expression in patients with stageⅠlung adenocarcinoma. Then a Nomogram prediction model is established and the ROC curve is used to evaluate the prediction efficiency of the Nomogram model. Results A total of 421 patients were included, including 225 STAS-negative patients, 79 males and 146 females, aged 60(50.5～70 years), and 166 STas-positive patients, 89 males and 77 females, aged 62(55.0～69.75 years). Univariate and multivariate logistic regression analysis showed that CT value, leaf sign, pleural stretch sign, crescent sign and vacuolar sign were independent risk factors for predicting STAS positive. The OR values(95%CI) were 1.00(1.00～1.01), 24.81(7.80～78.9), 0.29(0.10～0.84), 0.18(0.06～0.51) and 0.17(0.07～0.44), respectively. The area under the curve 0.918(95%CI 0.895～0.948) sensitivity 86.1%, and specificity 85.3% of the Nomogram model were established based on these five independent risk factors. Conclusion This study establishes a Nomogram model based on high-resolution CT features and clinical factors to assist in the accurate diagnosis of stageⅠlung cancer and the selection of appropriate treatment and management strategies.更多还原