实用老年医学

实用老年医学 ›› 2022, Vol. 36 ›› Issue (9): 892-896.doi: 10.3969/j.issn.1003-9198.2022.09.007

• 临床研究 • 上一篇    下一篇

老年脓毒症病人28 d死亡的危险因素及预测模型建立

简万均, 蒋昌华, 符宜龙, 舒艾娅, 郑春   

  1. 408000 重庆市,重庆大学附属涪陵医院重症医学科(简万均,蒋昌华,符宜龙,舒艾娅);
    641000 四川省内江市,内江市第一人民医院重症医学科(郑春)
  • 收稿日期:2021-09-04 出版日期:2022-09-20 发布日期:2022-09-21
  • 通讯作者: 蒋昌华,Email: 727709712@qq.com
  • 基金资助:
    重庆市卫生计生委医学科研计划项目(2016ZDXM012)

Risk factors and establishment of predictive model of 28-day death in elderly patients with sepsis

JIAN Wan-jun, JIANG Chang-hua, FU Yi-long, SHU Ai-ya, ZHENG Chun   

  1. JIAN Wan-jun, JIANG Chang-hua, FU Yi-long, SHU Ai-ya. Department of Critical Care Medicine, Fuling Hospital Affiliated to Chongqing University, Chongqing 408000, China;
    ZHENG Chun. Department of Critical Care Medicine, the First People's Hospital of Neijiang, Neijiang 641000, China
  • Received:2021-09-04 Online:2022-09-20 Published:2022-09-21

PDF (PC)

1094

摘要： 目的 探讨老年脓毒症病人28 d死亡的危险因素,并进一步构建预后预测模型。 方法 回顾性分析2016年10月至2020年12月我院ICU诊治的637例老年脓毒症病人的临床资料。通过电子病历系统采集病人入院时临床资料,并记录28 d预后。采用多因素Logistic回归分析脓毒症病人28 d死亡的危险因素,并进一步构建预测模型,采用ROC曲线对模型进行验证。 结果 637例老年脓毒症病人,28 d死亡130例,病死率为20.41%。多因素Logistic回归分析结果显示,入院时中性粒细胞/淋巴细胞(NLR)＞17.50、APACHEⅡ评分＞22.75、中性粒细胞明胶酶相关脂质运载蛋白(NGAL)＞535.29 ng/mL、活化部分凝血活酶时间(APTT)＞39.05 s是老年脓毒症病人28 d死亡的危险因素,乳酸清除率(LCR)＞25.36%是其保护因素。根据Logistic回归分析结果,构建脓毒症病人28 d死亡的预测模型,ROC曲线分析显示,内部验证预测模型预测脓毒症病人死亡的AUC为0.878(95%CI:0.802~0.935),准确率为80.85%,外部验证时,AUC为0.856(95%CI:0.793~0.921),准确率为80.22%。 结论 NLR、APACHEⅡ评分、NGAL水平升高,APTT延长是老年脓毒症28 d死亡的危险因素,LCR升高是其保护因素,据此构建的风险模型可能预测病人的短期死亡风险。

关键词: 脓毒症, 老年人, 预后, 预测模型

Abstract: Objective To explore the risk factors of 28-day death in elderly patients with sepsis, and to further construct the prediction model. Methods A retrospective analysis of the clinical data of 637 elderly patients with sepsis from the ICU of our hospital from October 2016 to December 2020 was carried out. The clinical data of the patients were collected through the electronic medical record system, and the 28-day prognosis was recorded. Multivariate Logistic regression was used to analyze the risk factors of 28-day death in the elderly patients with sepsis, and the selected influencing factors were used to construct a predictive model, and the receive operating characteristic (ROC) curve was used to verify the model. Results Among 637 elderly patients with sepsis, 130 died in 28 days, with a mortality rate of 20.41%. Multivariate Logistic regression analysis showed that neutrophil-to-lymphocyte ratio (NLR)>17.50, APACHE Ⅱ score>22.75, neutrophil gelatinase-associated lipid carrier protein (NGAL)>535.29 ng/mL, activated partial thrombin time (APTT) >39.05 s at admission were the risk factors of death in 28 days in the elderly patients with sepsis. Lactic acid clearance (LCR) >25.36% was the protective factor. According to the results of Logistic regression analysis, a prediction model for death of sepsis patients was established. ROC curve analysis showed that the area under the curve (AUC) of the model predicting poor prognosis of sepsis was 0.878 (95%CI:0.802-0.935, P<0.001) in internal verification, with an accuracy of 80.85%. AUC was 0.856(95%CI:0.793-0.921, P<0.001) and the accuracy was 80.22% in external validation. Conclusions Elevated NLR, APACHE Ⅱ score, NGAL, APTT are risk factors, and elevated LCR is a protective factor for 28-day death due to sepsis in the elderly. The risk model constructed on this basis may predict the short-term mortality of the patients with sepsis.

Key words: sepsis, aged, prognosis, prediction model

中图分类号: