Department of Critical Care Medicine, Xiangya Hospital Central South University, Changsha, China.
School of Computer Science and Engineering, Central South University, Changsha, China.
Ren Fail. 2022 Dec;44(1):1886-1896. doi: 10.1080/0886022X.2022.2142139.
Acute kidney injury (AKI) is more likely to develop in the elderly admitted to the intensive care unit (ICU). Acute kidney disease (AKD) affects ∼45% of patients with AKI and increases short-term mortality. However, there are no studies on the prognosis of AKD in the elderly.
Data from 2666 elderly patients with AKD in the Medical Information Mart for Intensive Care IV were used for model development and 535 in the eICU Collaborative Research Database for external validation. Based on 5 machine learning algorithms, 33 noninvasive parameters were extracted as features for modeling.
In-hospital mortality of AKD in the elderly was 29.6% and 31.8% in development and validation cohorts, respectively. The comprehensive best-performing algorithm was the support vector machine (SVM), and a simplified online application included only 10 features employing SVM (AUC: 0.810 and 0.776 in the training and external validation cohorts, respectively) was deployed. Model interpretation by SHapley Additive exPlanation (SHAP) values revealed that the difference (AKD day - ICU day) in sequential organ failure assessment (delta SOFA), Glasgow coma scale (GCS), delta GCS, delta peripheral oxygen saturation (SpO2), and SOFA were the top five features associated with prognosis. The optimal target was determined by SHAP values from partial dependence plots.
A web-based tool was externally validated and deployed to predict the early prognosis of AKD in the elderly based on readily available noninvasive parameters, assisting clinicians in intervening with precision and purpose to save lives to the greatest extent.
急性肾损伤（AKI）在入住重症监护病房（ICU）的老年人中更易发生。急性肾疾病（AKD）影响约 45%的 AKI 患者，并增加短期死亡率。然而，目前尚无关于老年人 AKD 预后的研究。
使用 Medical Information Mart for Intensive Care IV 中 2666 例老年 AKD 患者的数据进行模型开发，以及 eICU Collaborative Research Database 中的 535 例数据进行外部验证。基于 5 种机器学习算法，提取 33 个非侵入性参数作为建模特征。
老年 AKD 的院内死亡率分别为发展队列和验证队列的 29.6%和 31.8%。表现最佳的综合算法是支持向量机（SVM），一个简化的在线应用程序仅包括 10 个采用 SVM 的特征（在训练和外部验证队列中的 AUC 分别为 0.810 和 0.776）。通过 SHapley Additive exPlanation（SHAP）值进行模型解释表明，序贯器官衰竭评估（SOFA）的差值（AKD 天数-ICU 天数）、格拉斯哥昏迷评分（GCS）、GCS 差值、外周血氧饱和度（SpO2）差值和 SOFA 是与预后相关的五个最重要特征。通过偏依赖图的 SHAP 值确定最佳目标。
基于可获得的非侵入性参数，开发并外部验证了一个基于网络的工具，以预测老年 AKD 的早期预后，帮助临床医生有针对性地进行干预，最大限度地拯救生命。
