摘要
目的分析胫骨平台骨折患者关节镜术后膝关节僵直的影响因素并构建预测模型。方法回顾性分析西安大兴医院2020年6月至2023年5月154例胫骨平台骨折患者的临床资料,患者均行关节镜辅助内固定术治疗。记录患者的性别、年龄、体质量、骨折原因、Schatzker分型、骨质疏松、半月板及韧带结构损伤、下肢力线不良、植骨、伸膝装置损伤、内固定方式、外固定时间、异位骨化、Tscherne分型、放置引流管、清创次数、手术时机和麻醉方式。患者随访1年,记录膝关节僵直发生情况,并将患者分为膝关节僵直组和膝关节正常组。采用多因素Logistic回归分析影响胫骨平台骨折患者关节镜术后发生膝关节僵直的独立危险因素。采用R语言安装包构建预测胫骨平台骨折患者关节镜术后膝关节僵直的列线图模型,绘制校准曲线并采用Bootstrap法验证模型区分度,绘制受试者工作特征(ROC)曲线验证模型的预测价值。结果154例胫骨平台骨折患者中,28例发生膝关节僵直(膝关节僵直组),发生率为18.18%;膝关节正常126例(膝关节正常组)。膝关节僵直组肥胖、半月板及韧带结构损伤、伸膝装置损伤、下肢力线不良、异位骨化比例明显高于膝关节正常组,差异有统计学意义(P<0.05或<0.01);两组性别构成、年龄、骨折原因、骨质疏松、Schatzker分型、植骨、内固定方式、外固定时间、Tscherne分型、放置引流管、清创次数、手术时机和麻醉方式比较差异无统计学意义(P>0.05)。多因素Logistic回归分析结果显示,肥胖、半月板及韧带结构损伤、伸膝装置损伤、下肢力线不良和异位骨化是影响胫骨平台骨折患者关节镜术后膝关节僵直的独立危险因素(OR=5.387、4.613、3.308、3.178和4.579,95%CI 1.207~24.034、1.447~14.709、1.063~10.291、1.155~8.745和1.540~13.613,P<0.05或<0.01)。以体质量、半月板及韧带结构损伤、伸膝装置损伤、下肢力线不良和异位骨化为预测因子构建预测胫骨平台骨折患者关节镜术后发生膝关节僵直的列线图模型。校准曲线分析结果显示,模型的理想曲线与实际曲线的走势轨迹基本一致(C-index=0.861);ROC曲线分析结果显示,该模型预测胫骨平台骨折患者关节镜术后发生膝关节僵直的精准度良好(曲线下面积为0.861),灵敏度为92.9%,特异度为73.0%。结论胫骨平台骨折患者关节镜术后有发生膝关节僵直的风险,可能与患者肥胖、半月板及韧带结构损伤、伸膝装置损伤、下肢力线不良、异位骨化有关,基于这些危险因素构建的列线图预测模型将风险可视化,对患者术后1年内发生膝关节僵直有一定预测价值。
Objective To analyze the influencing factors of knee joint stiffness in patients with tibial plateau fractures after arthroscopy and construct a predictive model.Methods The clinical data of 154 patients with tibial plateau fractures from June 2020 to May 2023 in Xi'an Daxing Hospital were retrospectively analyzed.All patients were treated with arthroscopic assisted internal fixation surgery.The gender,age,body weight,fracture cause,Schatzker classification,osteoporosis,meniscus and ligament structure injury,lower limb malalignment,bone graft,knee extension device injury,internal fixation method,external fixation time,heterotopic ossification,Tscherne classification,drainage tube placement,debridement times,operation opportunity and anesthesia method were recorded.The patients were followed up for 1 year,and the occurrence of knee joint stiffness was recorded.The patients were divided into knee joint stiffness group and normal knee joint group.Multivariate Logistic regression was used to analyze the independent risk factors of knee joint stiffness after arthroscopy in patients with tibial plateau fractures.R language software package was used to construct the nomogram model for predicting the knee joint stiffness after arthroscopy in patients with tibial plateau fractures,calibration curve was drawn,and the Bootstrap method was used to verify the model discrimination.The predictive value of model was evaluated by the receiver operating characteristics(ROC)curve.Results Among 154 patients with tibial plateau fractures,28 patients developed knee joint stiffness after arthroscopy(knee joint stiffness group),with an incidence rate of 18.18%;126 patients had normal knee joints(normal knee joint group).The proportions of obesity,meniscus and ligament structure injury,knee extension device injury,lower limb malalignment and heterotopic ossification in knee joint stiffness group were significantly higher than those in normal knee joint group,and there were statistical differences(P<0.05 or<0.01);there were no statistical differences in gender composition,age,fracture cause,osteoporosis,Schatzker classification,bone graft,internal fixation method,external fixation time,Tscherne classification,drainage tube placement,debridement times,operation opportunity and anesthesia method between the two groups(P>0.05).Multivariate Logistic regression analysis result showed that obesity,meniscus and ligament structure injury,knee extension device injury,lower limb malalignment and heterotopic ossification were independent risk factors for knee joint stiffness after arthroscopy in patients with tibial plateau fractures(OR=5.387,4.613,3.308,3.178 and 4.579;95%CI 1.207 to 24.034,1.447 to 14.709,1.063 to 10.291,1.155 to 8.745 and 1.540 to 13.613;P<0.05 or<0.01).The body weight,meniscus and ligament structure injury,knee extension device injury,lower limb malalignment and heterotopic ossification were used as predictors to construct a nomogram model for predicting the knee joint stiffness after arthroscopy in patients with tibial plateau fractures.The calibration curve analysis result showed that the theoretical curve was basically consistent with the actual curve trend trajectory(C-index=0.861).The ROC curve analysis result showed that the model for predicting the knee joint stiffness after arthroscopy in patients with tibial plateau fractures had good accuracy(area under curve was 0.861),with the sensitivity of 92.9%and specificity of 73.0%.Conclusions The patients with tibial plateau fractures have the risk of knee joint stiffness after arthroscopy,which may be related to patient obesity,meniscus and ligament structure injury,knee extension device injury,lower limb malalignment and heterotopic ossification.Based on these risk factors,a column chart risk prediction model can be constructed to visualize the risk and have certain predictive value for knee joint stiffness within 1 year after surgery.
作者
高帅帅
王延军
姚永峰
周群
Gao Shuaishuai;Wang Yanjun;Yao Yongfeng;Zhou Qun(Department of Orthopedics,Xi'an Daxing Hospital,Xi'an 710003,China)
出处
《中国医师进修杂志》
2025年第9期831-837,共7页
Chinese Journal of Postgraduates of Medicine
关键词
胫骨骨折
关节镜
列线图
膝关节
Tibial fractures
Arthroscopes
Nomograms
Knee joint
