文献速递重症综合管理与镇静镇痛心脑血管急症复苏、休克与脓毒症

ECMO PAL:利用深度神经网络预测静脉-动脉体外膜氧合患者的生存

ECMO PAL: using deep neural networks for survival prediction in venoarterial extracorporeal membrane oxygenation.
PubMed 文献 · Intensive Care Med · 2023-08-07
一句话结论在ELSO国际注册数据库18,167例训练、5,015例外部验证中,ECMO PAL预测VA-ECMO住院死亡准确率72.7%,优于ECMO ACCEPTS(54.7%)、SAVE(61.1%)和Modified SAVE(62%)。

本研究基于国际体外生命支持组织(ELSO)注册数据库2017-2020年18,167例VA-ECMO患者,训练深度神经网络ECMO PAL预测住院死亡,采用五折交叉验证。外部验证纳入2021年全部成人注册患者(N=5015),并与SAVE、Modified SAVE、ECMO ACCEPTS评分比较。ECMO PAL验证准确率72.7%(99%CI 71.1-74.3%),仅比训练准确率75.5%低2.8%;交叉验证住院死亡敏感度82.1±0.2%、精确率77.6±0.2%。ECMO PAL准确率高于ECMO ACCEPTS(54.7%)、SAVE(61.1%)和Modified SAVE(62%)。

要点

  • ECMO PAL在2021年成人注册患者外部验证中预测住院死亡准确率72.7%(99%CI 71.1-74.3%),仅比训练准确率75.5%低2.8%。
  • 交叉验证显示ECMO PAL预测住院死亡敏感度82.1±0.2%、精确率77.6±0.2%。
  • ECMO PAL准确率72.7%高于ECMO ACCEPTS(54.7%)、SAVE(61.1%)和Modified SAVE(62%)。
  • 训练队列18,167例来自ELSO注册数据库(2017-2020),平均年龄56.8±15.1岁,男性66.7%,ECMO前心脏骤停50.2%。
  • 外部验证纳入2021年全部成人注册患者N=5015,提示模型在不同ECMO地区具有较高泛化性。

适用人群与场景

  • 训练队列:ELSO注册数据库中2017-2020年的18,167例VA-ECMO患者。
  • 外部验证队列:2021年ELSO注册数据库中全部成人VA-ECMO患者(N=5015)。
  • 训练队列平均年龄56.8±15.1岁,男性66.7%,ECMO前心脏骤停50.2%。

研究设计与关键结果

  • 研究类型:基于ELSO注册数据库的回顾性队列研究,训练深度神经网络ECMO PAL并用五折交叉验证评估性能,外部验证采用2021年全部成人注册患者(N=5015)。
  • 训练队列18,167例,平均年龄56.8±15.1岁,男性66.7%,ECMO前心脏骤停50.2%。
  • 交叉验证住院死亡敏感度82.1±0.2%,精确率77.6±0.2%。
  • 验证准确率72.7%(99%CI 71.1-74.3%),较训练准确率75.5%低2.8%。
  • ECMO PAL准确率72.7%高于ECMO ACCEPTS(54.7%)、SAVE(61.1%)和Modified SAVE(62%)。

证据类型、等级与局限

  • 证据类型:基于国际注册数据库的回顾性队列研究,含内部交叉验证与外部验证。
  • 局限:原文未报告局限性。

对床旁实践的改变

  • 作者认为ECMO PAL是首个基于大型国际患者队列训练和验证的人工智能ECMO生存评分,在ECMO地区间显示较高泛化性,并优于现有广泛使用的评分。
  • 作者认为该研究提示大型国际注册数据可用于复杂重症治疗的人工智能预后预测。

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

PURPOSE: Venoarterial extracorporeal membrane oxygenation (VA-ECMO) is a complex and high-risk life support modality used in severe cardiorespiratory failure. ECMO survival scores are used clinically for patient prognostication and outcomes risk adjustment. This study aims to create the first artificial intelligence (AI)-driven ECMO survival score to predict in-hospital mortality based on a large international patient cohort.

METHODS: A deep neural network, ECMO Predictive Algorithm (ECMO PAL) was trained on a retrospective cohort of 18,167 patients from the international Extracorporeal Life Support Organisation (ELSO) registry (2017-2020), and performance was measured using fivefold cross-validation. External validation was performed on all adult registry patients from 2021 (N = 5015) and compared against existing prognostication scores: SAVE, Modified SAVE, and ECMO ACCEPTS for predicting in-hospital mortality.

RESULTS: Mean age was 56.8 ± 15.1 years, with 66.7% of patients being male and 50.2% having a pre-ECMO cardiac arrest. Cross-validation demonstrated an inhospital mortality sensitivity and precision of 82.1 ± 0.2% and 77.6 ± 0.2%, respectively. Validation accuracy was only 2.8% lower than training accuracy, reducing from 75.5% to 72.7% [99% confidence interval (CI) 71.1-74.3%]. ECMO PAL accuracy outperformed the ECMO ACCEPTS (54.7%), SAVE (61.1%), and Modified SAVE (62%) scores.

CONCLUSIONS: ECMO PAL is the first AI-powered ECMO survival score trained and validated on large international patient cohorts. ECMO PAL demonstrated high generalisability across ECMO regions and outperformed existing, widely used scores. Beyond ECMO, this study highlights how large international registry data can be leveraged for AI prognostication for complex critical care therapies.

来源:Intensive Care Med,查看原文。版权归原作者所有。

参考来源

  1. Stephens AF, Šeman M, Diehl A, et al. ECMO PAL: using deep neural networks for survival prediction in venoarterial extracorporeal membrane oxygenation. Intensive Care Med. 2023. PMID: 37548758. doi:10.1007/s00134-023-07157-x

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。