Department of Anesthesiology and Pain Medicine, Seoul National University College of Medicine, Seoul National University Hospital, Seoul, South Korea.
Department of Anesthesiology and Pain Medicine, Anesthesia and Pain Research Institute, Yonsei University College of Medicine, Seoul, South Korea.
Br J Anaesth. 2024 Jun;132(6):1304-1314. doi: 10.1016/j.bja.2024.01.030. Epub 2024 Feb 26.
Postoperative respiratory failure is a serious complication that could benefit from early accurate identification of high-risk patients. We developed and validated a machine learning model to predict postoperative respiratory failure, defined as prolonged (>48 h) mechanical ventilation or reintubation after surgery.
Easily extractable electronic health record (EHR) variables that do not require subjective assessment by clinicians were used. From EHR data of 307,333 noncardiac surgical cases, the model, trained with a gradient boosting algorithm, utilised a derivation cohort of 99,025 cases from Seoul National University Hospital (2013-9). External validation was performed using three separate cohorts A-C from different hospitals comprising 208,308 cases. Model performance was assessed by area under the receiver operating characteristic (AUROC) curve and area under the precision-recall curve (AUPRC), a measure of sensitivity and precision at different thresholds.
The model included eight variables: serum albumin, age, duration of anaesthesia, serum glucose, prothrombin time, serum creatinine, white blood cell count, and body mass index. Internally, the model achieved an AUROC of 0.912 (95% confidence interval [CI], 0.908-0.915) and AUPRC of 0.113. In external validation cohorts A, B, and C, the model achieved AUROCs of 0.879 (95% CI, 0.876-0.882), 0.872 (95% CI, 0.870-0.874), and 0.931 (95% CI, 0.925-0.936), and AUPRCs of 0.029, 0.083, and 0.124, respectively.
Utilising just eight easily extractable variables, this machine learning model demonstrated excellent discrimination in both internal and external validation for predicting postoperative respiratory failure. The model enables personalised risk stratification and facilitates data-driven clinical decision-making.
术后呼吸衰竭是一种严重的并发症，可以通过早期准确识别高危患者获益。我们开发并验证了一种机器学习模型，用于预测术后呼吸衰竭，定义为手术后长时间（>48 小时）机械通气或重新插管。
使用不需要临床医生主观评估的易于提取的电子健康记录（EHR）变量。从首尔国立大学医院（2013-9 年）的 307333 例非心脏手术病例的 EHR 数据中，使用梯度提升算法训练模型，利用 derivation 队列 99025 例进行训练。使用来自三家不同医院的三个独立队列 A-C 进行外部验证，共 208308 例。通过接收者操作特征曲线（AUROC）下面积和精度-召回曲线（AUPRC）下面积评估模型性能，这是一种在不同阈值下衡量敏感性和精度的指标。
该模型包括 8 个变量：血清白蛋白、年龄、麻醉持续时间、血糖、凝血酶原时间、血清肌酐、白细胞计数和体重指数。内部模型的 AUROC 为 0.912（95%置信区间 [CI]，0.908-0.915），AUPRC 为 0.113。在外部验证队列 A、B 和 C 中，模型的 AUROC 分别为 0.879（95%CI，0.876-0.882）、0.872（95%CI，0.870-0.874）和 0.931（95%CI，0.925-0.936），AUPRC 分别为 0.029、0.083 和 0.124。
该机器学习模型仅使用 8 个易于提取的变量，在内部和外部验证中均表现出出色的鉴别能力，可用于预测术后呼吸衰竭。该模型能够实现个性化风险分层，并有助于数据驱动的临床决策。
