目的 探究老年社区获得性肺炎相关脓毒症患者预后影响因素并构建预测模型.方法 回顾性纳入宁夏医科大学总医院2020年10月至2022年10月老年社区获得性肺炎相关脓毒症患者的临床资料,通过随机数字表法按7∶3比例分为建模人群和验证人群,并比较两人群的临床资料有无可比性.根据入院30 d结局分为生存组和死亡组,通过LASSO回归、多因素Logistic回归分析筛选出老年社区获得性肺炎相关脓毒症患者预后的独立危险因素,使用R软件构建列线图预测模型.采用受试者工作特征曲线的曲线下面积(area under the curve,AUC)、校准曲线和决策曲线分别在建模人群和验证人群中对列线图预测模型进行验证以判断其区分度、校准度和临床实用性.结果 共纳入472例患者,建模与验证人群分别为331例、141例,建模人群与验证人群间临床资料具有可比性.LASSO回归、多因素Logistic回归分析显示肺炎严重程度(pneumonia severity index,PSI)评分及序贯器官衰竭(sequential organ failure assessment,SOFA)评分是老年社区获得性肺炎相关脓毒症患者预后的独立危险因素.建模人群预测模型的AUC为0.984(95％CI:0.975～0.994),验证人群预测模型的AUC为0.961(95％CI:0.926～0.996).列线图预测模型在建模人群和验证人群中均具有良好的区分度、校准度和临床实用性.结论 本研究建立的列线图预测模型对老年社区获得性肺炎相关脓毒症患者预后的早期识别和风险有较高的准确性,能够为临床医生制定个性化干预措施提供指导.
A prediction model to predict the prognosis of elderly patients with community-acquired pneumonia-associated sepsis
Objective To explore the prognostic factors of elderly patients with community-acquired pneumonia-related sepsis and to construct a prediction model.Methods The clinical data of elderly patients with community-acquired pneumonia-associated sepsis from October 2020 to October 2022 in the General Hospital of Ningxia Medical University from October 2020 to October 2022 were retrospectively included,and the clinical data of the two groups were divided into the modeling population and the validation population in the ratio of 7∶3 by random number table method,and the clinical data of the two groups were compared.According to the 30-day outcomes of admission,the patients were divided into survival group and death group,and the independent risk factors for the prognosis of elderly patients with community-acquired pneumonia-related sepsis were screened out by LASSO regression and multivariate logistic regression analysis,and the nomogram prediction model was constructed by R software.The area under the curve(AUC),calibration curve and decision curve of the receiver operating characteristic curve were used to validate the nomogram prediction model in the modeling population and the validation population to judge its discrimination,calibration and clinical practicability.Results A total of 472 patients were included,with 331 and 141 models and validations,respectively,indicating that the clinical data were comparable between the modeled and validated populations.LASSO regression and multivariate logistic regression analysis showed that pneumonia severity index(PSI)score and sequential organ failure assessment(SOFA)score were independent risk factors for the prognosis of elderly patients with community-acquired pneumonia-associated sepsis.The AUC of the modeled population prediction model was 0.984(95％CI:0.975-0.994),and the AUC of the validated population prediction model was 0.961(95％CI:0.926-0.996).The nomogram prediction model has good discrimination,calibration and clinical practicability in both the modeled and validated populations.Conclusions The nomogram prediction model established in the study has high accuracy for early identification and risk of sepsis in elderly patients with CAP and can guide for clinicians to formulate personalized interventions.
NETL
NSTL
万方数据
