老年重症肺炎患者病情转归不良的风险预测Nomogram模型建立及验证  

Establishment and verification of a Nomogram model to predict the risk of adverse outcomes in elderly patients with severe pneumonia

作　　者：王静 孙振康[1] WANG Jing;SUN Zhenkang(Department of Critical Care Medicine,Fuyang People′s Hospital,Fuyang,Anhui 236000,China)

机构地区：[1]阜阳市人民医院重症医学科,安徽阜阳236000

出　　处：《临床肺科杂志》2024年第7期1024-1029,1048,共7页Journal of Clinical Pulmonary Medicine

基　　金：阜阳市卫生健康委科研项目(No.FY2021-131)。

摘　　要：目的构建老年重症肺炎(SP)患者病情转归不良的风险预测Nomogram模型,并验证其预测效能。方法回顾性分析2022年1月~2023年6月本院重症监护病房收治的432例老年SP患者,随机分为训练集(n=288)和验证集(n=144)。根据患者入住ICU后28天病情转归将其分为转归不良组和转归良好组。采用Logistic回归分析筛选患者病情转归不良的影响因素,构建其风险预测Nomogram模型,以受试者工作特征(ROC)曲线和校准曲线评估模型的区分度和准确性,以决策曲线分析(DCA)评估模型的临床净获益。结果老年SP患者病情转归不良发生率为24.77%;肺炎严重程度(PSI)评分、白细胞计数(WBC)、纤维蛋白原(FIB)、降钙素原(PCT)是患者病情转归不良的影响因素(OR=1.677、1.619、1.435、1.935,P均<0.05);训练集与验证集的一致性指数(C-index)分别为0.874、0.839;ROC曲线下面积(AUC)分别为0.929、0.891;决策曲线(DCA)显示在训练集及验证集中模型预测病情转归不良的最大净获益值阈值概率区间分别在0~0.91和0~0.90。结论PSI评分、WBC、FIB、PCT是老年SP患者病情转归不良的影响因素,以此为基础构建的风险预测Nomogram模型具有良好预测效能与临床应用价值。Objective To construct a Nomogram model for predicting the risk of poor prognosis in elderly patients with severe pneumonia(SP)and verify its predictive efficacy.Methods A retrospective analysis was conducted on 432 elderly SP patients admitted to the intensive care unit of our hospital from January 2022 to June 2023.They were randomly divided into a training set(n=288)and a validation set(n=144)in a 2:1 ratio.According to the outcome of the patient′s condition 28 days after admission to the ICU,they will be divided into a poor outcome group and a good outcome group.Logistic regression analysis was used to screen for the influencing factors of poor patient outcomes,and a Nomogram model for risk prediction was constructed.The discrimination and accuracy of the model were evaluated using receiver operating characteristic(ROC)curves and calibration curves,and the clinical net benefit of the model was evaluated using decision curve analysis(DCA).Results The incidence of poor prognosis in elderly SP patients was 24.77%(107/432);Pneumonia severity(PSI)score,white blood cell count(WBC),fibrinogen(FIB),and procalcitonin(PCT)were the influencing factors for adverse disease outcome(OR=1.677,1.619,1.435,1.935,all P<0.05).The consistency index(C-index)of the training set and verification set were 0.874 and 0.839,respectively.The areas under ROC curve(AUC)were 0.929 and 0.891,respectively.The DCA showed that the probability interval of the maximum net benefit value for predicting adverse disease outcomes in the training set and validation set was 0-0.91 and 0-0.90,respectively.Conclusion PSI score,WBC,FIB,and PCT are the influencing factors for poor prognosis in elderly SP patients.The risk prediction Nomogram model constructed based on these factors has good predictive efficacy and clinical application value.

关 键 词：老年人 重症肺炎 转归 Nomogram模型 

分 类 号：R563.1[医药卫生—呼吸系统]