严重创伤患者并发肺部感染的风险预测列线图模型构建  

Construction of a nomogram model for risk prediction of pulmonary infection in patients with severe trauma

作　　者：乔梦梦 张斌[2] 巩月英[2] 施昊旻 张欣妍 王璐璐 杨惠莲[1] QIAO Mengmeng;ZHANG Bin;GONG Yueying;SHI Haomin;ZHANG Xinyan;WANG Lulu;YANG Huilian(Department of Public Health of Medical College,Qinghai University,Qinghai Province,Xining810016,China;Emergency Intensive Care Unit,Qinghai Province People’s Hospital,Qinghai Province,Xining810001,China)

机构地区：[1]青海大学医学部公共卫生系,青海西宁810016 [2]青海省人民医院急诊重症监护室,青海西宁810001

出　　处：《中国医药导报》2024年第28期124-129,共6页China Medical Herald

基　　金：青海省科技计划项目(2021-QY-206)。

摘　　要：目的探讨严重创伤患者并发肺部感染的影响因素,并利用列线图模型预测严重创伤患者发生肺部感染的风险。方法选取青海省人民医院急诊重症监护病房(ICU)于2016年1月至2023年9月收治住院的严重创伤患者作为研究对象,根据是否发生肺部感染将其分为感染组(163例)和未感染组(339例)。采用二元logistic回归对并发肺部感染的危险因素进行分析,在此基础上构建列线图模型,绘制受试者操作特征曲线对模型进行评估。结果两组性别、严重不良事件、并发重症感染、机械通气、气管切开、ICU住院时间、GCS评分、收缩压比较,差异有统计学意义(P<0.05)。多因素logistic回归分析结果显示,男性(OR=2.340,95%CI:1.309~4.183)、并发重症感染(OR=2.706,95%CI:1.455~5.036)、ICU住院时间(OR=1.038,95%CI:1.016~1.061)、气管切开(OR=3.121,95%CI:1.799~5.413)为肺部感染的独立危险因素(P<0.05)。基于这4项独立危险因素构建的列线图模型,曲线下面积为0.805(95%CI:0.757~0.852)。校准曲线显示,预测概率与实际概率之间具有较好的一致性。临床决策曲线分析显示,阈值概率为10%~70%时模型具有临床有效性。结论性别、并发重症感染、ICU住院时间、气管切开是严重创伤患者并发肺部感染的独立危险因素,基于此构建的列线图模型具有良好的风险预测能力。Objective To explore the influencing factors of pulmonary infection in patients with severe trauma,and to predict the risk of pulmonary infection in patients with severe trauma by using a nomogram model.Methods Patients with severe trauma admitted to the emergency intensive care unit(ICU),Qinghai Province People’s Hospital from January 2016 to September 2023 were selected as the study objects.They were divided into infection group(163 cases)and non-infection group(339 cases)based on whether lung infection occurred.Binary logistic regression was used to analyze the risk factors for concurrent lung infection.On this based,a nomogram model was constructed,and receiver operating characteristic curves and other methods were used to evaluate the model.Results There were statistically significant differences in gender,serious adverse events,concurrent severe infection,mechanical ventilation,tracheotomy,length of ICU stay,GCS score and systolic blood pressure between the two groups(P<0.05).Multivariate logistic regression analysis showed male(OR=2.340,95%CI:1.309-4.183),concurrent severe infection(OR=2.706,95%CI:1.455-5.036),length of stay in ICU(OR=1.038,95%CI:1.016-1.061),and tracheotomy(OR=3.121,95%CI:1.799-5.413)were independent risk factors for pulmonary infection(P<0.05).The area under the curve was 0.805(95%CI:0.757-0.852)in the nomogram model based on these four independent risk factors.The calibration curve showed a good agreement between the predicted probability and the actual probability,and the clinical decision curve analysis showed that the model was clinically valid when the threshold probability was 10%to 70%.Conclusion Gender,concurrent severe infection,length of stay in ICU,and tracheotomy are independent risk factors for concurrent pulmonary infection in patients with severe trauma.The nomogram model based on these factors has good risk prediction ability.

关 键 词：严重创伤 肺部感染 LOGISTIC 列线图 

分 类 号：R641[医药卫生—外科学] R563.1[医药卫生—临床医学]