胸腹联合腔镜下食管癌根治术后肺部并发症的危险因素分析及预测模型建立  

Risk factors and predictive model for postoperative pulmonary complications in patients after laparoscopic and thoracoscopic esophagectomy

作　　者：杨唯奕 刘冉 刘海林[1] 张志捷[1] 苏珍[1] YANG Weiyi;LIU Ran;LIU Hailin;ZHANG Zhijie;SU Zhen(Department of Anesthesiology,the Affiliated Huai'an No.1 People's Hospital of Nanjing Medical University,Jiangsu Huai'an 223300,China;Department of Anesthesiology,Zhuzhou Central Hospital,Hunan Zhuzhou 412000,China)

机构地区：[1]南京医科大学附属淮安第一医院麻醉科,江苏淮安223300 [2]株洲市中心医院麻醉科,湖南株洲412000

出　　处：《现代肿瘤医学》2024年第20期3886-3890,共5页Journal of Modern Oncology

基　　金：江苏省淮安市基础研究计划(联合专项)卫生健康类科研项目(编号:HABL2023067);南京医科大学科技发展基金项目(编号:NMUB2020139)。

摘　　要：目的:筛选胸腹联合腔镜下食管癌根治术术后肺部并发症(postoperative pulmonary complications,PPCs)发生的危险因素,建立PPCs列线图预测模型,为该类患者PPCs的早期识别和防治提供临床指导。方法:本研究回顾性收集2022年03月至2022年08月在南京医科大学附属淮安第一医院行胸腹联合腔镜下食管癌根治术患者的临床资料。根据术后7天内是否发生PPCs,将患者分为PPCs组与非PPCs组。运用SPSS 26.0软件行logistic单因素和多因素回归分析,筛选出PPCs的危险因素。运用R语言建立Nomogram预测模型;并绘制受试者工作特征(ROC)曲线、校准曲线以及临床决策(DCA)曲线分别评价模型的区分度、一致性和临床实用性。结果:本研究共纳入165例患者,发生PPCs 79例。逻辑回归分析显示:男性、肺年龄、术前PEF百分比、术前WBC、术后24 h血糖变异系数是胸腹联合腔镜下食管癌根治术发生PPCs的独立危险因素(P<0.05)。根据这些危险因素构建Nomogram列线图预测模型,ROC曲线下面积0.821,特异性73.3%,敏感性79.7%,校准曲线一致性良好,DCA决策曲线净收益良好。结论:男性、肺年龄、术前PEF百分比、术前WBC、术后24 h血糖变异系数是胸腹联合腔镜下食管癌根治术PPCs发生的独立危险因素。建立的Nomogram列线图模型具有较高的PPCs预测效能,能帮助早期识别PPCs高风险人群。Objective:To screen the risk factors for the occurrence of pulmonary complications(PPCs)in patients after laparoscopic and thoracoscopic esophagectomy,and to establish a nomogram prediction model for PPCs,so as to provide clinical guidance for the early identification and prevention of PPCs in such patients.Methods:This study retrospectively collected the clinical data of patients who underwent laparoscopic and thoracoscopic esophagectomy in Huai'an No.1 People's Hospital,Nanjing Medical University from March 2022 to August 2022.Patients were divided into PPCs group and non-PPCs group according to whether PPCs occurred within 7 days after surgery.SPSS 26.0 software was used for logistic univariate and multivariate regression analysis to screen out the risk factors of PPCs.Nomogram prediction model was established using R language.The receiver operating characteristic(ROC)curve,calibration curve and clinical decision making(DCA)curve were plotted to evaluate the discrimination,consistency and clinical practicability of the model.Results:A total of 165 patients were included in this study,and 79 patients developed PPCs.Logistic regression analysis showed that male,lung age,percentage of preoperative PEF,preoperative WBC,and 24 h postoperative blood glucose coefficient of variation were independent risk factors for PPCs after laparoscopic and thoracoscopic esophagectomy(P<0.05).Based on these risk factors,the Nomogram nomogram prediction model was constructed,and the area under the ROC curve was 0.821,the specificity was 73.3%,the sensitivity was 79.7%,the calibration curve was consistent,and the DCA decision curve had a good net benefit.Conclusion:Male,lung age,percentage of preoperative PEF,preoperative WBC,and 24-hour postoperative blood glucose coefficient of variation are independent risk factors for the occurrence of PPCs after laparoscopic and thoracoscopic esophagectomy.The established Nomogram nomogram model has high predictive performance of PPCs and can help identify high-risk groups of PPCs at an early s

关 键 词：胸腹联合腔镜下食管癌根治术 肺部并发症 危险因素 预测模型 

分 类 号：R735.1[医药卫生—肿瘤]