业余运动员半月板损伤预测的多域整合模型：个体化预防算法的开发
简介
该研究纳入620名业余运动员，整合骨骼形态、功能生物力学、训练负荷、生物生活方式及主观症状五个领域，开发并内部验证了半月板损伤多域预测模型。模型筛选出7个独立预测因子，区分度良好，校准度优异，并据此衍生出半月板损伤风险指数，将运动员分为四个递增风险等级。该模型为基于个体风险分层的个性化预防策略提供了实用框架。
英文摘要
BACKGROUND: Meniscal injuries are common among amateur athletes and are a major cause of knee dysfunction and long-term joint degeneration. Current strategies for identifying athletes at risk rely mainly on isolated structural or functional factors and remain largely reactive. An integrative predictive approach combining multiple risk domains to guide individualized prevention is lacking. PURPOSE: To develop and internally validate a multidomain predictive model for meniscal injury in amateur athletes and to derive a personalized prevention algorithm based on individual risk profiles. METHODS: A combined prospective-retrospective cohort study included 620 amateur athletes aged 15-50 years, followed for 24-36 months. Baseline assessment integrated five domains: osseous and meniscal morphology, functional biomechanics, training load and sport exposure, biological and lifestyle factors, and subjective symptoms. Predictor selection was performed using least absolute shrinkage and selection operator regression, followed by multivariable logistic regression. Model performance was evaluated using the area under the receiver operating characteristic curve, calibration analysis, and Decision Curve Analysis. A clinically interpretable risk score, the Meniscal Injury Risk Index, was derived to stratify athletes and guide prevention strategies. RESULTS: During follow-up, 92 athletes (14.8%) sustained a meniscal injury. Seven independent predictors were retained. The model demonstrated strong discrimination, excellent calibration, and consistent net clinical benefit. The risk score stratified athletes into four categories with progressively increasing injury incidence. CONCLUSION: This multidomain model accurately identifies amateur athletes at risk for meniscal injury and provides a practical framework for personalized, risk-based prevention. LEVEL OF EVIDENCE: II-III. STUDY DESIGN: Prognostic cohort study.