肺癌放化疗后患者发生肺部感染的危险因素及其风险预测列线图模型构建  

Analysis of risk factors of pulmonary infection in patients with lung cancer after chemoradiotherapy and construction of risk prediction nomograph model

作　　者：付金平 陈惠 赵洁 晋发 张文君 张秀伟 FU Jin-ping;CHEN Hui;ZHAO Jie;JIN Fa;ZHANG Wen-jun;ZHANG Xiu-wei(The Affiliated Jiangning Hospital with Nanjing Medical University,Nanjing,Jiangsu 21l100,China)

机构地区：[1]南京医科大学附属江宁医院放疗科,江苏南京211100 [2]南京医科大学附属江宁医院肿瘤科,江苏南京211100 [3]南京医科大学附属江宁医院呼吸科,江苏南京211100

出　　处：《中华医院感染学杂志》2024年第8期1145-1149,共5页Chinese Journal of Nosocomiology

基　　金：国家自然科学基金青年基金资助项目(82002881);江苏省社发临床重点专项(EB2016721);江苏省自然科学基金青年基金资助项目(BK20170142)。

摘　　要：目的探究肺癌放化疗后患者发生肺部感染的影响因素,并构建其风险预测列线图模型。方法研究对象为2019年5月-2022年10月南京医科大学附属江宁医院收治的198例肺癌且接受放化疗治疗的患者,根据放化疗期间有无发生肺部感染病例分为感染组73例和非感染组125例。采用LASSO回归分析筛选潜在影响因素;采用Logistic回归分析探究患者发生肺部感染的影响因素;列线图模型采用R4.2.6语言“rms”包构建,并通过绘制受试者工作特征(ROC)曲线、校准曲线、决策曲线评价模型的区分度、校准度及有效性。结果采用LASSO回归模型筛选出潜在的影响因素,糖尿病病史、血白蛋白(ALB)、血红蛋白(HGB)、化疗周期>4次、放化疗药物>2种、中性粒细胞和淋巴细胞比值(NLR)、血小板和淋巴细胞比值(PLR)、使用抗菌药物、肺不张、肿瘤患者功能状态评分<80分,多因素Logistic回归分析结果显示,ALB、HGB、化疗周期>4次、使用抗菌药物、NLR、PLR是肺癌放化疗患者发生肺部感染的影响因素(P<0.05)。基于6个影响因素构建列线图模型,ROC曲线显示该模型的曲线下面积(AUC)为0.904(95%CI:0.865~0.997),一致性指数为0.963;模型的校准曲线与理想曲线接近;决策曲线提示,当该模型预测的概率为0.381~0.922时,模型预测价值较优越。结论肺癌放化疗患者发生肺部感染的影响因素有ALB、HGB、化疗周期>4次、使用抗菌药物、NLR、PLR等,构建的列线图模型对预测肺癌放化疗患者发生肺部感染具有较好的效能。OBJECTIVE To explore the influencing factors of pulmonary infection in patients with lung cancer after radiotherapy and chemotherapy,and to construct a risk prediction column chart model.METHODS A total of 198 patients with lung cancer admitted to the Affiliated Jiangning Hospital of Nanjing Medical University from May 2019 to Oct 2022 received radiotherapy and chemotherapy treatment were recruited in the study.They were divid-ed into the infection group(73 cases)and non-infection group(125 cases)according to whether pulmonary infec-tion occurred during the radiotherapy and chemotherapy period.Potential influencing factors were screened by LASSO regression analysis,and the influencing factors of pulmonary infection in patients were analyzed by logistic regression analysis.The column chart model is constructed using the"rms"package of R4.2.6 language,and the discrimination,calibration,and effectiveness of the model are evaluated by receiver operating characteristic(ROC)curves,calibration curves,and decision curves.RESULTS Ten potential influencing factors was analyzed by LASSO regression model,including the history of diabetes,serum albumin(ALB),hemoglobin(HGB),more than four cycles of radiotherapy and chemotherapy,more than two kinds of radiotherapy and chemotherapy drugs,neutrophil to lymphocyte ratio(NLR),platelet to lymphocyte ratio(PLR),use of antibiotics,atelectasis and functional status score of tumor patients<80 points.Multivariate logistic regression analysis showed that ALB,HGB,radiotherapy and chemotherapy cycles>4 times,use of antibiotics,NLR,and PLR were independ-ent influencing factors for the occurrence of pulmonary infection in lung cancer patients undergoing radiotherapy and chemotherapy(P<0.05).The column chart model was built based on six influencing factors and the ROC curve showed that the AUC value of the model was 0.904(95%CI:0.865-0.997),with the consistency index of 0.963.The calibration curve of the model was close to the ideal curve.The decision curve indicated that the model had superio

关 键 词：肺癌放化疗 肺部感染 影响因素 LOGISTIC回归分析 列线图模型 

分 类 号：R563.1[医药卫生—呼吸系统]