脑出血手术患者肺部感染风险预测评分模型的构建与验证  

Establishment and validation of risk prediction scoring model for pulmonary infection in cerebral hemorrhage patients undergoing surgery

作　　者：宋甜田 李亚婷[1,2] 宋明 刘雪燕 王书会 SONG Tian-tian;LI Ya-ting;SONG Ming;LIU Xue-yan;WANG Shu-hui(Department of Infection Management,Qilu Hospital of Shandong University,Jinan 250012,China;School of Nursing,Shandong University,Jinan 250012,China)

机构地区：[1]山东大学齐鲁医院感染管理处,山东济南250012 [2]山东大学护理学院,山东济南250012

出　　处：《中国感染控制杂志》2021年第4期299-303,共5页Chinese Journal of Infection Control

基　　金：山东省自然科学基金(ZR2018MG015)。

摘　　要：目的构建脑出血手术患者肺部感染风险预测评分模型,识别肺部感染的高危人群,为临床医务人员早期采取有效预防与控制措施提供依据。方法前瞻性收集山东省某医院2016—2018年脑出血手术患者的临床资料,将患者按照7∶3的比例随机分为建模组和验证组,利用建模组数据建立logistic回归模型,依据β值对危险因素进行赋分,构建肺部感染风险预测评分模型。应用建立的评分模型对建模组和验证组患者进行风险评分,通过受试者工作特征(ROC)曲线和Hosmer-Lemeshow(H-L)检验判断模型的预测能力。结果logistic回归分析显示,气管插管、留置胃管、静脉血栓、ASA分级是脑出血手术患者发生肺部感染的独立危险因素,风险评分模型中相应地分值分别为5、6、4、2分(每个等级)。风险评分模型在建模组中的ROC曲线下面积(AUC)为0.864(95%CI:0.825~0.904,P<0.001),H-L检验结果为P=0.404;在验证组中的AUC为0.861(95%CI:0.800~0.921,P<0.001),H-L检验结果为P=0.452。结论构建的风险预测评分模型能够较好地预测脑出血手术患者发生肺部感染的风险,早期识别高风险人群,优化患者预后。Objective To establish the risk prediction scoring model of pulmonary infection(PI)in cerebral hemo-rrhage patients undergoing surgery,identify the high-risk population of PI,and provide basis for health care wor-kers(HCWs)to take early effective prevention and control measures.Methods Clinical data of cerebral hemorrhage patients undergoing surgery in a hospital in Shandong Province in 2016-2018 were collected prospectively,patients were randomly divided into modeling group and validation group based on a ratio of 7∶3,logistic regression model was established based on data of modeling group,factors were assigned according toβvalue,risk prediction scoring model of PI was established.The established risk scoring model was used to score the risk of patients in modeling group and validation group,and the predictive ability of the model was judged by receiver operating characteristic(ROC)curve and Hosmer-Lemeshow(H-L)test.Results Logistic regression analysis showed that endotracheal intubation,indwelling gastric tube,venous thrombosis and ASA were independent risk factors for PI in cerebral hemorrhage patients undergoing surgery,and corresponding scores in the risk scoring model were 5,6,4,and 2 points(per stratification)respectively.AUC under ROC curve of the risk scoring model in the modeling group was 0.864(95%CI:0.825-0.904,P<0.001),H-L test was P=0.404;AUC under ROC curve in validation group was 0.861(95%CI:0.800-0.921,P<0.001),H-L test was P=0.452.Conclusion The established risk prediction scoring model can better predict the risk of PI in cerebral hemorrhage patients undergoing surgery,which can early identify high-risk population and optimize the prognosis of patients.

关 键 词：脑出血 肺部感染 危险因素 风险预测 评分模型 

分 类 号：R181.32[医药卫生—流行病学]