摘要
目的探究连续性肾脏替代治疗(CRRT)重症肺部感染相关脓毒症患者预后影响因素,并构建可视化的风险列线图模型。方法选取2017年1月-2020年1月桂东人民医院重症科收治的130例重症肺部感染合并脓毒症并接受CRTT治疗的患者作为研究对象,收集其临床资料,根据患者30d内的生存将其分为死亡组和存活组,采用logistic回归筛选患者预后不良的独立危险因素,R软件建立列线图预测模型,并对模型的预测准确性进行内外部验证。结果Logistic回归分析结果显示,APACHEⅡ评分、降钙素原(PCT)、白蛋白(Alb)、血乳酸(Lac)、内毒素(ET)水平为CRRT治疗重症肺部感染相关脓毒症患者死亡的独立危险因素(P<0.05)。基于以上独立危险因素建立列线图模型,模型区分度:AUC=0.975(95%CI:0.954~0.995),敏感度和特异度分别为87.8%、95.8%;校准曲线贴近于理想曲线。对模型进行外部验证,AUC=0.965(95%CI:0.921~1.000),敏感度和特异度分别为93.8%、85.3%。结论CRRT治疗重症肺部感染合并脓毒症患者预后不良的危险因素,主要有高水平的APACHEⅡ、PCT、Lac、ET及低水平的Alb等,基于危险因素构建列线图模型具有良好的预测准确性,可为临床人员筛查高危人群和制定相关防治措施提供参考。
Objective To explore the prognostic factors of patients with severe pulmonary infection-related sepsis treated with continuous renal replacement therapy(CRRT),and to construct a visual risk nomogram model.Methods From January 2017 to January 2020,130 patients with severe pulmonary infection complicated with sepsis and treated with CRTT in the intensive care unit of Guidong People′s Hospital were selected as the research subjects,and their clinical data were collected.According to whether the patients were survived within 30 days,they were divided into the death group and survival group.Logistic regression was used to screen independent risk factors for the poor prognosis of patients.The nomogram prediction model was established by R software,and the prediction accuracy of the model was verified internally and externally.Results Logistic regression analysis showed that APACHEⅡscore,procalcitonin(PCT),albumin(Alb),blood lactate(Lac),and endotoxin(ET)levels were independent risk factors for death in patients with severe lung infection-related sepsis treated with CRRT(P<0.05).A nomogram model was established based on the above 5 independent risk factors.The model discrimination was AUC=0.975(95%CI:0.954~0.995),and the sensitivity and specificity were 87.8%and 95.8%,respectively.The calibration curve was close to the ideal curve.The model was further verified externally,AUC=0.965(95%CI:0.921~1.000),and the sensitivity and specificity were 93.8%and 85.3%,respectively.Conclusion The main risk factors for poor prognosis of patients with severe lung infection combined with sepsis and treated with CRRT mainly include high levels of APACHEⅡ,PCT,Lac,ET,and low levels of Alb.The construction of a nomogram model based on risk factors has good predictive accuracy,which can provide a reference for clinical staff to screen high-risk groups and formulate relevant preventive measures.
作者
黄泽波
曾凡乔
江祚龙
吴传黠
李健
韦妍飞
HUANG Ze-bo;ZENG Fan-qiao;JIANG Zuo-long;WU Chuan-xian;LI Jian;WEI Yan-fei(Department of Critical Care Medicine,Guidong people′s Hospital of Guangxi Zhuang Autonomous Region,Wuzhou,Guangxi 543000,China;Department of Integrative Critical Care Medicine,nanxishan Hospital of Guangxi Zhuang Autonomous Region,Guilin,Guangxi 541002,China)
出处
《临床肺科杂志》
2022年第7期1030-1035,1041,共7页
Journal of Clinical Pulmonary Medicine
基金
广西壮族自治区卫生健康委员会自筹经费科研课题(No.Z20190379)。
关键词
连续性肾脏替代治疗
重症肺部感染
脓毒症
预后
列线图
continuous renal replacement therapy
severe pulmonary infection
sepsis
prognosis
nomogram
