关节镜下前交叉韧带重建术后对侧韧带损伤的风险预警可视化模型的研究及其应用价值

Research and application of risk warning visualization model for contralateral ligament injury after arthroscopic anterior cruciate ligament reconstruction surgery

摘要:

目的 构建关节镜下前交叉韧带（ACL）重建术后对侧韧带损伤的风险预警可视化模型，并对预警模型行内部验证，以期为医护人员评估ACL损伤患者术后对侧韧带损伤风险提供简洁、直观和高效的个体化评估工具。方法 选取2021年1月－2024年1月该院所有行关节镜下ACL重建术治疗的ACL损伤患者583例，排除失访和资料不完整的51例患者后，最终纳入532例患者。根据是否发生关节镜下ACL重建术后对侧韧带损伤，将患者分为发生组（n = 42）与未发生组（n = 490），采用单因素分析比较两组患者临床资料，采用R软件，通过多因素二分类Logistic回归分析，获取预测因子和回归系数，建立可视化列线图预警模型，并采用校准曲线、C指数和决策曲线，进行一致性、区分度和临床适用性等内部验证和外部验证。结果 多因素Logistic回归分析显示，运动员、女性、膝关节内扣、术后6周Lachman试验阳性、术后3个月国际膝关节文献委员会（IKDC）评分、年龄 < 20岁和术前胫骨平台后倾角（PTS）是关节镜下ACL重建术后对侧韧带损伤的影响因素（P < 0.05）；在此基础上，应用R软件建立可视化列线图预警模型，经Bootstrap检验后显示，C指数为0.892；绘制受试者操作特征曲线（ROC curve），该模型在内部验证集和外部验证集中的曲线下面积（AUC）分别为：0.820（95%CI：0.751～0.890）和0.799（95%CI：0.712～0.886）；绘制校准曲线，该模型内部验证集和外部验证集中的斜率均近似为1；决策曲线中，该模型内部验证集和外部验证集均具有明显净获益。结论 关节镜下ACL重建术后，对侧韧带损伤受年龄、性别、术后6周Lachman试验阳性、膝关节内扣、术后3个月IKDC评分、术前PTS和运动员的影响，据此构建的可视化预警模型，具有良好的临床应用价值，可指导临床工作者开展针对性干预。

Abstract:

Objective To construct a visualized model for early warning of contralateral ligament injury risk after arthroscopic anterior cruciate ligament (ACL) reconstruction, and to conduct internal validation of the warning model, with a view to providing a concise, intuitive, and efficient individualized assessment tool for medical staff to evaluate the risk of contralateral ligament injury in patients with ACL injury after surgery.Methods 583 patients with ACL injuries who underwent arthroscopic ACL reconstruction surgery from January 2021 to January 2024 were selected. Excluding 51 cases those were lost to follow-up or had incomplete data, clinical data of 532 patients were finally included. According to whether there was a contralateral ligament injury, patients were categorized into an occurrence group (n = 42) and a non-occurrence group (n = 490). Data from the two groups of patients were compared using single-factor analysis. R software was employed to obtain predictive factors and regression coefficients through multivariate binary Logistic regression analysis. A visual nomogram early warning model was established, and internal and external validations were conducted using calibration curves, C index, and decision curves to assess consistency, discrimination, and clinical utility.Results Multivariate Logistic regression analysis revealed that athletes, females, knee joint buckling, positive Lachman test results at 6 weeks post-surgery, International Knee Documentation Committee (IKDC) score at 3 months post-surgery, age < 20 years, and preoperative posterior tibial slope (PTS) were influencing factors for contralateral ligament injury after arthroscopic ACL reconstruction (P < 0.05). Based on this, a visual nomogram warning model was established using R software. After Bootstrap testing, the model demonstrated a C index of 0.892. Draw the receiver operating characteristic curve (ROC curve), the model's area under the curve (AUC) values in the internal and external validation sets were 0.820 (95%CI: 0.751 ~ 0.890) and 0.799 (95%CI: 0.712 ~ 0.886), respectively. The calibration curves indicated that the slopes in both the internal and external validation sets were approximately 1. In the decision curve, the model demonstrated significant net benefit in both the internal and external validation sets.Conclusion The contralateral ligament injury after arthroscopic ACL reconstruction is influenced by age, gender, positive Lachman test at 6 weeks postoperatively, knee joint buckling, IKDC score at 3 months postoperatively, preoperative PTS, and athlete status. The visualized early warning model constructed based on these factors has good clinical application value and can guide clinicians to carry out targeted interventions.