Department of Hematology, Fifth Medical Center of Chinese PLA General Hospital, Beijing, 100853, China; National Clinical Research Center of Geriatric Diseases, Chinese PLA General Hospital, Beijing, 100853, China.
Department of Emergency Medicine, General Hospital of Northern Theatre Command, 83 Wenhua Road, Shenyang 110016, China.
Aust Crit Care. 2023 Jul;36(4):604-612. doi: 10.1016/j.aucc.2022.06.001. Epub 2022 Jul 14.
Currently, very few preventive or therapeutic strategies are used for mechanical ventilation (MV)-associated severe acute kidney injury (AKI).
We developed clinical prediction models to detect the onset of severe AKI in the first week of intensive care unit (ICU) stay during the initiation of MV.
A large ICU database Medical Information Mart for Intensive Care IV (MIMIC-IV) was analysed retrospectively. Data were collected from the clinical information recorded at the time of ICU admission and during the initial 12 h of MV. Using univariate and multivariate analyses, the predictors were selected successively. For model development, two machine learning algorithms were compared. The primary goal was to predict the development of AKI stage 2 or 3 (AKI-23) and AKI stage 3 (AKI-3) in the first week of patients' ICU stay after initial 12 h of MV. The developed models were externally validated using another multicentre ICU database (eICU Collaborative Research Database, eICU) and evaluated in various patient subpopulations.
Models were developed using data from the development cohort (MIMIC-IV: 2008-2016; n = 3986); the random forest algorithm outperformed the logistic regression algorithm. In the internal (MIMIC-IV: 2017-2019; n = 1210) and external (eICU; n = 1494) validation cohorts, the incidences of AKI-23 were 154 (12.7%) and 119 (8.0%), respectively, with areas under the receiver operator characteristic curve of 0.78 (95% confidence interval [CI]: 0.74-0.82) and 0.80 (95% CI: 0.76-0.84); the incidences of AKI-3 were 81 (6.7%) and 67 (4.5%), with areas under the receiver operator characteristic curve of 0.81 (95% CI: 0.76-0.87) and 0.80 (95% CI: 0.73-0.86), respectively.
Models driven by machine learning and based on routine clinical data may facilitate the early prediction of MV-associated severe AKI. The validated models can be found at: https://apoet.shinyapps.io/mv_aki_2021_v2/.
目前，用于机械通气（MV）相关严重急性肾损伤（AKI）的预防或治疗策略非常有限。
我们开发了临床预测模型，以检测 MV 开始后 ICU 入住第一周内严重 AKI 的发生。
回顾性分析大型 ICU 数据库医学信息集市重症监护 IV（MIMIC-IV）。数据来自 ICU 入院时和 MV 最初 12 小时内记录的临床信息。使用单变量和多变量分析，依次选择预测因子。为了模型开发，比较了两种机器学习算法。主要目标是预测 MV 后初始 12 小时内患者 ICU 入住第一周内 AKI 第 2 或 3 期（AKI-23）和 AKI 第 3 期（AKI-3）的发展。使用另一个多中心 ICU 数据库（eICU 协作研究数据库，eICU）对开发的模型进行外部验证，并在各种患者亚群中进行评估。
模型使用发展队列（MIMIC-IV：2008-2016；n=3986）的数据进行开发；随机森林算法优于逻辑回归算法。在内部（MIMIC-IV：2017-2019；n=1210）和外部（eICU；n=1494）验证队列中，AKI-23 的发生率分别为 154（12.7%）和 119（8.0%），受试者工作特征曲线下面积分别为 0.78（95%置信区间 [CI]：0.74-0.82）和 0.80（95% CI：0.76-0.84）；AKI-3 的发生率分别为 81（6.7%）和 67（4.5%），受试者工作特征曲线下面积分别为 0.81（95% CI：0.76-0.87）和 0.80（95% CI：0.73-0.86）。
基于机器学习并基于常规临床数据的模型可能有助于早期预测 MV 相关严重 AKI。验证后的模型可以在以下网址找到：https://apoet.shinyapps.io/mv_aki_2021_v2/。
