超重/肥胖患者并发冠心病的风险预警模型的中西医研究进展

Research Progress of Integrated Traditional Chinese and Western Medicine in Risk Early Warning Models for Coronary Artery Disease Complicated with Overweight/Obesity

摘要: 随着全球超重/肥胖流行率攀升，其引发冠心病的风险及致死负担日益严峻，风险预警模型成为精准防控的关键工具。本文系统梳理相关研究进展：西医模型从传统回归模型演进至机器学习模型，通过整合临床指标、生物标志物及多组学数据，预测精度持续提升，但存在“黑箱”局限与人群适配性问题；中医模型以证素评分和中西医结合量表为核心，紧扣“痰浊”“血瘀”等核心证素，体现整体辨证特色，却面临量化标准不统一的挑战；中西医结合模型通过“指标叠加”与“机制融合”两种模式，实现微观指标与宏观辨证互补，展现最优预测效能。现有研究存在样本代表性不足、验证体系不完善等问题，未来需依托多中心大样本队列，推动指标创新、技术突破与临床转化，构建精准化、可解释、实用性强的新一代预警模型。

Abstract: With the global rise in the prevalence of overweight/obesity, the risk of coronary heart disease and the associated mortality burden it poses have become increasingly severe, making risk prediction models a key tool for precision prevention and control. This article systematically reviews the relevant research progress: Western medicine models have evolved from traditional regression models to machine learning models, with continuously improved prediction accuracy through the integration of clinical indicators, biomarkers, and multi-omics data. However, they face limitations such as the “black box” problem and population adaptability issues. Traditional Chinese medicine (TCM) models, centered on syndrome element scoring and integrated TCM-Western medicine scales, focus on core syndrome elements like phlegm turbidity and blood stasis, reflecting the characteristic of holistic syndrome differentiation, but encounter challenges such as inconsistent quantification standards. Integrated TCM-Western medicine models, through two modes—“index superposition” and “mechanism integration”—achieve the complementarity of microcosmic indicators and macrocosmic syndrome differentiation, demonstrating the optimal predictive performance. Existing studies suffer from insufficient sample representativeness and imperfect validation systems. Future research should rely on multi-center, large-sample prospective cohorts to promote indicator innovation, technological breakthroughs, and clinical translation, thereby constructing a new generation of prediction models that are precise, interpretable, and highly practical.