文献速递重症综合管理与镇静镇痛复苏、休克与脓毒症

基于随机森林的重症脓毒症/脓毒性休克28天预后预测模型

[A model based on random forests in prediction of 28-day prognosis in patients with severe sepsis/septic shock].
PubMed 文献 · Zhonghua Wei Zhong Bing Ji Jiu Yi Xue · 2017-01-01
一句话结论在重症脓毒症/脓毒性休克患者中,基于随机森林构建的预后模型预测28天死亡的AUC为0.836,优于SOFA和APACHE II评分。

该研究纳入2013年9月至2017年5月浙江医院ICU收治的497例重症脓毒症/脓毒性休克患者,收集第1、3、5天基本资料、生命体征、生化及血常规指标,按28天预后分为死亡组与生存组,筛选特异性指标构建随机森林预后模型,并用ROC曲线比较该模型与SOFA、APACHE II评分的评估效能。497例中201例死亡,28天死亡率为40.4%。随机森林模型筛选出19个特异性参数,预测28天死亡的AUC为0.836,高于第1、3、5天的SOFA(0.643、0.554、0.766)和APACHE II(0.590、0.670、0.758),敏感度86.1%、特异度77.0%、准确率80.7%。

要点

  • 497例重症脓毒症/脓毒性休克患者28天死亡201例,死亡率40.4%。
  • 随机森林模型预测28天死亡的AUC为0.836,敏感度86.1%、特异度77.0%、准确率80.7%。
  • 随机森林模型AUC高于第1、3、5天SOFA评分(0.643、0.554、0.766)和APACHE II评分(0.590、0.670、0.758)。
  • 模型入选19个参数,包括年龄、第1天24小时尿量、BUN、SCr、PLT,第3天HR、MAP、发绀、皮肤湿冷,第5天体温、HR、MAP、24小时尿量、PLT、发热、发绀、呼吸困难、皮肤湿冷、花斑。
  • 作者认为基于随机森林的评估模型能有效预测重症脓毒症/脓毒性休克患者28天死亡风险,预测效能优于SOFA和APACHE II评分。

适用人群与场景

  • 2013年9月至2017年5月浙江医院ICU收治的重症脓毒症/脓毒性休克患者,共497例。

研究设计与关键结果

  • 研究类型为预后预测模型构建与效能评估,纳入497例重症脓毒症/脓毒性休克患者,201例死亡,28天死亡率40.4%。
  • 按死亡组与生存组指标差异筛选出19个随机森林模型特异性参数:年龄;第1天24小时尿量、尿素氮、血清肌酐、血小板计数;第3天心率、平均动脉压、发绀、皮肤湿冷;第5天体温、心率、平均动脉压、24小时尿量、血小板计数、发热、发绀、呼吸困难、皮肤湿冷、花斑。
  • ROC曲线分析显示,随机森林模型预测28天死亡的AUC高于第1、3、5天SOFA评分和APACHE II评分(AUC:0.836 vs. 0.643、0.554、0.766和0.590、0.670、0.758)。
  • 随机森林模型预测28天死亡的敏感度86.1%、特异度77.0%、准确率80.7%。

证据类型、等级与局限

  • 证据类型:预后预测模型构建与效能评估研究。
  • 局限:原文未报告局限性。

对床旁实践的改变

  • 作者认为基于随机森林的评估模型能有效预测重症脓毒症/脓毒性休克患者28天死亡风险,其预测效能优于SOFA和APACHE II评分。

提炼说明

  • 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开

OBJECTIVE: To establish a severe sepsis/septic shock prognosis prediction model based on randomize forest law (RF model), and to evaluate the prognostic value of this model for patients with severe sepsis/septic shock.

METHODS: 497 patients with severe sepsis/septic shock admitted to intensive care unit (ICU) of Zhejiang Hospital from September 2013 to May 2017 were enrolled. The basic data, vital signs and symptoms, biochemical indexes and blood routine indexes on the 1st, 3rd, 5th day and prognosis were collected. According to the 28-day prognosis, the patients were divided into death group and survival group, and the specific indicators about the prognosis of severe sepsis/septic shock were screened. A RF model was constructed by using the specificity indicators. The assessment effectiveness of RF model, sequential organ failure assessment (SOFA), acute physiology and chronic health evaluation II (APACHE II) were evaluated by receiver operating characteristic (ROC) curve analysis.

RESULTS: In 497 cases of severe sepsis/septic shock, 201 cases died, 28-day mortality was 40.4%. (1) According to the index difference of death group and survival group, 19 specific parameters of the RF model were selected, which included the age; 24-hour urine output, urea nitrogen (BUN), serum creatinine (SCr), platelet count (PLT) on the 1st day; heart rate (HR), mean arterial pressure (MAP), cyanosis and clammy skin on the 3rd day; temperature, HR, MAP, 24-hour urine output, PLT, fever, cyanosis, dyspneic, clammy skin, piebald on the 5th day. (2) ROC curve analysis showed that the area under the ROC curve (AUC) of RF model predicting 28-day mortality was higher than that of SOFA and APACHE II score on the 1st, 3rd, 5th day (AUC: 0.836 vs. 0.643, 0.554, 0.766 and 0.590, 0.670, 0.758). The sensitivity of RF model to predict the 28-day mortality was 86.1%, the specificity was 77.0%, the accuracy was 80.7%.

CONCLUSIONS: The evaluation model based on random forest can effectively predict the death risk of 28-day in patients with severe sepsis/septic shock, and its predictive efficiency is better than that of the SOFA and APACHE II score.

来源:Zhonghua Wei Zhong Bing Ji Jiu Yi Xue,查看原文。版权归原作者所有。

参考来源

  1. Wang Y, Chen S, Hu C, et al. [A model based on random forests in prediction of 28-day prognosis in patients with severe sepsis/septic shock]. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2017. PMID: 29216938. doi:10.3760/cma.j.issn.2095-4352.2017.12.004

供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。