整合临床、MRI和关节镜发现的诊断模型可准确识别过度活动外侧半月板：开发与内部验证
简介
回顾性纳入100例疑似外侧半月板病变患者，以动态关节镜下活动度为参考标准，仅用术前临床和MRI变量构建Hypermobile Meniscus Index。关节镜确诊外侧半月板过度活动占47%；最终模型含5个预测因子，AUC 0.84，最佳截断≥5分时敏感度81%、特异度79.2%，内部校正AUC 0.82。该模型可辅助术前识别并指导针对性关节镜评估，但仅内…
英文摘要
BACKGROUND: To develop and internally validate a diagnostic model for hypermobile lateral meniscus integrating clinical symptoms, magnetic resonance imaging findings, and dynamic arthroscopic assessment. METHODS: A retrospective diagnostic model study was performed including 100 consecutive patients evaluated for suspected lateral meniscal pathology at a tertiary knee surgery center. All patients underwent standardized clinical examination, preoperative magnetic resonance imaging (MRI), and arthroscopic assessment of lateral meniscal mobility, which served as the reference standard. Candidate predictors were restricted to preoperative clinical and magnetic resonance imaging variables, including intermittent locking, posterolateral McMurray test, meniscocapsular gap, and suspected meniscotibial discontinuity. Arthroscopic findings were used exclusively to establish the reference-standard diagnosis and were not entered into the diagnostic model. A penalized logistic regression model with LASSO selection was used to identify independent predictors and construct the Hypermobile Meniscus Index. Internal validation was performed using bootstrap resampling, and robustness was assessed with Monte Carlo simulation. RESULTS: Hypermobile lateral meniscus was confirmed arthroscopically in 47% of patients. The final model included five independent predictors and demonstrated excellent discriminative performance, with an area under the receiver operating characteristic curve of 0.84 (95% confidence interval (CI), 0.83-0.95). At the optimal cut-off (≥5 points), sensitivity was 81%, specificity 79.2%, positive predictive value 77.6%, and negative predictive value 82.4%. Calibration analysis showed good agreement between predicted and observed probabilities (calibration slope 0.94, intercept 0.03; Brier score 0.13). Bootstrap validation yielded an optimism-corrected AUC of 0.82. Monte Carlo simulation confirmed stability of model performance across varying prevalence scenarios, with consistent sensitivity and specificity ranges. CONCLUSION: A diagnostic model combining clinical and magnetic resonance imaging findings demonstrated internally validated performance for identifying hypermobile lateral meniscus as defined by dynamic arthroscopy. The Hypermobile Meniscus Index provides a structured framework to improve preoperative recognition and guide targeted arthroscopic assessment in patients with mechanical lateral compartment symptoms and inconclusive imaging. External validation is required prior to routine clinical implementation. LEVEL OF EVIDENCE: Level III, diagnostic study.