胸腔镜下肺癌根治术患者术后喉返神经麻痹的影响因素分析及预测模型构建  

Analysis of influencing factors and construction of predictive models of recurrent laryngeal nerve paralysis in patients undergoing thoracoscopic radical resection for lung cancer

作　　者：唐璐 周丽 唐梅 薛萍 钱蓉[1] TANG Lu;ZHOU Li;TANG Mei;XUE Ping;QIAN Rong(Operating Room,Inpatient Department,Chengdu Third People's Hospital,Chengdu,Sichuan 610031,China)

机构地区：[1]成都市第三人民医院住院部手术室,成都610031

出　　处：《新疆医科大学学报》2025年第3期328-333,共6页Journal of Xinjiang Medical University

基　　金：四川省医学青年创新科研项目(S2219)。

摘　　要：目的探讨胸腔镜下肺癌根治术患者术后喉返神经麻痹的影响因素,并构建预测模型。方法选取2022年2月—2023年12月在成都市第三人民医院进行胸腔镜下肺癌根治术治疗的300例肺癌患者作为研究对象,根据患者术后是否发生喉返神经麻痹将其分为麻痹组和无麻痹组。采用单因素及多因素Logistic回归分析,筛选胸腔镜下肺癌根治术患者术后喉返神经麻痹的影响因素,构建预测模型并评估其预测效能。结果300例肺癌患者中30例患者术后发生喉返神经麻痹,喉返神经麻痹发生率为10.00%。麻痹组肿瘤最大直径≥5 cm比率、胸腔镜全肺切除术比率、淋巴结清扫个数、淋巴结大小、术后肺部感染比率均高于无麻痹组,喉返神经监测比率低于无麻痹组(P<0.05)。多因素Logistic回归分析结果显示,肿瘤最大直径、淋巴结清扫个数、淋巴结大小、术后肺部感染是肺癌患者术后发生喉返神经麻痹的独立危险因素,喉返神经监测是肺癌患者术后发生喉返神经麻痹的独立保护因素(P<0.05);Hosmer-lemeshowχ^(2)=6.820,P=0.556,提示模型拟合度佳。该模型预测肺癌患者术后喉返神经麻痹的受试者工作特征(ROC)曲线下面积(AUC)为0.860,灵敏度和特异度分别为80.00%、72.59%。外部验证显示,模型预测肺癌患者术后喉返神经麻痹发生的AUC值为0.866(95%CI:0.801~0.916),敏感度为86.84%,特异性为73.68%,Hosmer-lemeshowχ^(2)=5.978,P=0.645,提示预测曲线与实际曲线贴合度良好。结论肿瘤最大直径、淋巴结清扫个数、淋巴结大小、术后肺部感染、喉返神经监测是肺癌患者术后喉返神经麻痹的影响因素,利用各独立影响因素构建的列线图模型对患者的生存预测具有良好的效能。Objective To explore the influencing factors of recurrent laryngeal nerve paralysis in patients undergoing thoracoscopic radical resection of lung cancer and to construct a prediction model.Methods 300 lung cancer patients who underwent thoracoscopic radical resection of lung cancer in the Chengdu Third People's Hospital from February 2022 to December 2023 were selected as the research subjects.The patients were divided into recurrent laryngeal nerve paralysis group and non recurrent laryngeal nerve paralysis group based on whether they experienced postoperative recurrent laryngeal nerve paralysis.Single factor and multiple factor Logistic regression analysis were used to screen the influencing factors of recurrent laryngeal nerve palsy in patients with thoracoscopic radical lung cancer surgery,build a prediction model and evaluate its prediction efficiency.Results Among 300 lung cancer patients,30 patients experienced recurrent laryngeal nerve paralysis(RLNP)postoperatively,with an incidence rate of 10.00%.The ratio of tumors with a maximum diameter≥5 cm,the rate of thoracoscopic pneumonectomy,the number of lymph nodes dissected,lymph node size and the rate of postoperative lung infection were higher in the RLNP group compared to the non-RLNP group.Additionally,the rate of recurrent laryngeal nerve monitoring was lower in the RLNP group(P<0.05).Multivariate Logistic regression analysis showed that maximum tumor diameter,number of lymph node dissection,lymph node size and postoperative pulmonary infection were independent risk factors for postoperative recurrent nerve palsy in lung cancer patients and recurrent nerve monitoring was independent protective factor for postoperative recurrent nerve palsy in lung cancer patients(P<0.05).The Hosmer-lemeshowχ^(2)=6.820,P=0.556,indicating a good fit of the model.The model predicted that the area under receiver Operating characteristic(ROC)curve(AUC)of postoperative recurrent laryngeal nerve palsy in lung cancer patients was 0.860,and the sensitivity and specificity we

关 键 词：肺癌 胸腔镜 喉返神经 影响因素 预测模型 

分 类 号：R734.2[医药卫生—肿瘤]