ICU获得性衰弱预测模型的研究进展

Research Progress on Prediction Models for ICU-Acquired Weakness

摘要: 重症监护病房获得性肌无力(Intensive Care Unit Acquired Weakness, ICU-AW)是危重患者常见的神经肌肉系统严重并发症，具有发病率高、预后不良的特点，可显著影响患者近期康复进程与远期生存质量。风险预测模型是实现ICU-AW早期识别、高危人群筛查与预防性干预的核心工具，目前国内外已研发多款ICU-AW风险预测模型，但缺乏对其系统的梳理与对比分析。本综述系统梳理了全球范围内ICU-AW风险预测工具的发展现状，涵盖传统统计学模型与前沿人工智能算法在该领域的应用，深度对比分析了不同模型的构建思路、核心参数、预测效能、临床适用性及核心优缺点，明确了当前该领域研究存在的空白与局限性。本综述旨在为临床医疗工作者科学选择适配的ICU-AW预测工具提供系统性参考，同时为未来构建符合我国临床实际、兼具高预测效能与临床可操作性的ICU-AW风险预测模型提供理论支撑与方向性建议。

Abstract: Intensive Care Unit Acquired Weakness (ICU-AW) is a common and serious neuromuscular complication in critically ill patients. Characterized by high incidence and poor prognosis, it significantly impacts patients’ short-term recovery and long-term quality of life. Risk prediction models are essential tools for the early identification of ICU-AW, screening high-risk populations, and enabling preventive interventions. Although various risk prediction models for ICU-AW have been developed globally, a systematic review and comparative analysis of these models is currently lacking. This review systematically summarizes the current state of development of ICU-AW risk prediction tools worldwide, covering both traditional statistical models and advanced artificial intelligence algorithms. It provides an in-depth comparative analysis of the construction methodologies, core parameters, predictive performance, clinical applicability, and key advantages and disadvantages of different models, identifying existing gaps and limitations in this field. This review aims to offer a systematic reference for clinical practitioners to select appropriate ICU-AW prediction tools and to provide theoretical support and directional recommendations for the future development of ICU-AW risk prediction models that align with China’s clinical practice, offering both high predictive accuracy and clinical feasibility.