Stieger Andrea, Schober Patrick, Venetz Philipp, Andereggen Lukas, Bello Corina, Filipovic Mark G, Luedi Markus M, Huber Markus
Department of Anaesthesiology and Pain Medicine, Cantonal Hospital of St. Gallen, St. Gallen, Switzerland.
Department of Anaesthesiology, Amsterdam University Medical Centres, Vrije Universiteit Amsterdam, Amsterdam, the Netherlands.
J Clin Anesth. 2025 Apr;103:111810. doi: 10.1016/j.jclinane.2025.111810. Epub 2025 Mar 9.
Accurate prediction of intensive care unit (ICU) admission and length of stay (LOS) after major surgery is essential for optimizing patient outcomes and healthcare resources. Factors such as age, BMI, comorbidities, and perioperative complications significantly influence ICU admissions and LOS. Machine learning methods have been increasingly utilized to predict these outcomes, but their clinical utility beyond traditional metrics remains underexplored.
This study examined a sub-cohort of 6043 patients who underwent general anesthesia at Seoul National University Hospital from August 2016 to June 2017. Various prediction models, including logistic regression and random forest, were developed for ICU admission and different LOS thresholds, e.g., a LOS of more than a week. Clinical utility was evaluated using decision curve analysis (DCA) across predefined risk preferences.
Among patients studied, 19.8 % were admitted to the ICU, with 1.4 % staying longer than a week. Prediction models demonstrated high discrimination (AUROC 0.93 to 0.96) and good calibration for ICU admission and short LOS. DCA revealed that intraoperative data provided the greatest decision-related benefit for predicting ICU admission, while preoperative data became more important for predicting longer LOS.
Intraoperative data are crucial for immediate postoperative decisions, while preoperative data are essential for extended LOS predictions. These findings highlight the need for a comprehensive risk assessment approach in perioperative care, utilizing both preoperative and intraoperative information to enhance clinical decision-making and resource allocation.
准确预测重症监护病房（ICU）收治情况以及大手术后的住院时长（LOS）对于优化患者治疗效果和医疗资源至关重要。年龄、体重指数（BMI）、合并症以及围手术期并发症等因素会显著影响ICU收治情况和住院时长。机器学习方法已越来越多地用于预测这些结果，但其在传统指标之外的临床效用仍未得到充分探索。
本研究对2016年8月至2017年6月在首尔国立大学医院接受全身麻醉的6043例患者的一个亚队列进行了研究。针对ICU收治情况以及不同的住院时长阈值（例如住院时长超过一周），开发了包括逻辑回归和随机森林在内的各种预测模型。使用决策曲线分析（DCA）对预定义风险偏好下的临床效用进行了评估。
在研究的患者中，19.8%被收治入ICU，其中1.4%的住院时间超过一周。预测模型在ICU收治情况和较短住院时长方面显示出较高的区分度（曲线下面积[AUC]为0.93至0.96）和良好的校准度。DCA显示，术中数据在预测ICU收治情况时提供了最大的决策相关益处，而术前数据在预测较长住院时长时变得更为重要。
术中数据对于术后即刻决策至关重要，而术前数据对于预测较长住院时长必不可少。这些发现凸显了在围手术期护理中采用综合风险评估方法的必要性，利用术前和术中信息来加强临床决策和资源分配。
