人工全髋关节置换术后关节脱位的危险因素分析及预测模型构建  

Analysis of Risk Factors and Forecasting Model Construction of Joint Dislocation after Total Hip Arthroplasty

作　　者：林永飞 刘立林 叶桂秀 LIN Yongfei;LIU Lilin;YE Guixiu(Department of Orthopedics,Ganzhou Hospital of Traditional Chinese Medicine,Ganzhou 341000,China;不详)

机构地区：[1]赣州市中医院骨科,江西赣州341000 [2]赣州市南康区第一人民医院骨科,江西赣州341000

出　　处：《中国医学创新》2024年第1期123-127,共5页Medical Innovation of China

基　　金：江西省卫生健康委科技计划项目(202312111)。

摘　　要：目的:分析人工全髋关节置换术(THA)后关节脱位的危险因素并构建列线图预测模型。方法:以2021年1月—2022年7月于赣州市中医院行THA的184例患者为研究对象,术后随访3个月,根据患者是否发生髋关节脱位分为脱位组(n=26)和未脱位组(n=158)。比较患者的一般资料,其中有统计学意义的计量资料的最佳截断值通过受试者操作特征(ROC)曲线获取;logistic多元回归模型分析THA后关节脱位的独立危险因素并构建预测THA后关节脱位的预测模型,预测模型的内部验证采用校正曲线验证,列线图模型的预测效能采用决策曲线进行评估。结果:与未脱位组相比,脱位组患者年龄更大,体重指数(BMI)更高,假体直径<30 mm、后外侧入路、有髋关节手术史、术后体位控制不严格、假体放置在安全区外的比例均更高(P<0.05)。ROC曲线分析结果显示:年龄、BMI的AUC值分别为0.819、0.729,最佳截断值分别为68岁、25.1 kg/m^(2)。年龄>68岁、有髋关节手术史、假体直径≤30 mm、后外侧入路、术后体位控制不严格、假体放置在安全区外是THA后关节脱位的危险因素(P<0.05)。对于THA后关节脱位的预测,列线图模型的C-index为0.732[95%CI(0.663,0.846)],观测值与预测值间有着较好的一致性;列线图模型的阈值>0.19,且该模型所供给的临床净收益均高于各项危险因素。结论:本研究基于年龄、髋关节手术史、假体直径、手术入路、术后体位控制、假体放置位置构建了预测THA后关节脱位的列线图模型,该模型对于THA后关节脱位的预测价值较好,可识辨THA后关节脱位的高危患者。Objective:To analyze the risk factors of joint dislocation after total hip arthroplasty(THA)and construct a nomograms forecasting model.Method:A total of 184 patients who underwent THA in Ganzhou Hospital of Traditional Chinese Medicine from January 2021 to July 2022 were selected as the study objects.Patients were followed up for 3 months after surgery and were divided into dislocation group(n=26)and non-dislocation group(n=158)according to whether they had hip dislocation.The general data of patients were compared,and the best cut-off value of statistically significant measurement data were obtained by receiver-operating characteristic(ROC)curve.The independent risk factors of joint dislocation after THA were analyzed by logistic multiple regression model and a forecasting model for joint dislocation after THA was constructed.The internal verification of the forecasting model was verified by calibration curve,and the prediction efficiency of the nomograms model was evaluated by decision curve.Result:Compared with the non-dislocation group,patients in the dislocation group were older,had higher body mass index(BMI)and the proportions of prosthesis diameter<30 mm,postlateral approach,had hip joint surgical history,less strict postoperative position control and prosthesis placed outside of safety zone were higher(P<0.05).ROC curve analysis results showed that the AUC values of age and BMI respectively was 0.819 and 0.729,and the best cut-off values respectively was 68 years old and 25.1 kg/m^(2).Age>68 years old,had hip joint surgical history,prosthesis diameter<30 mm,postlateral approach,less strict postoperative position control and prosthesis placed outside of safety zone were the risk factors for joint dislocation after THA(P<0.05).For the prediction of joint dislocation after THA,the C-index of the nomogram model was 0.732[95%CI(0.663,0.846)],has a good agreement between the observed and predicted values.The threshold>0.19 of the nomogram model,and the clinical net benefits provided by the model were higher t

关 键 词：人工全髋关节置换术 关节脱位 预测模型 危险因素 

分 类 号：R687.4[医药卫生—骨科学]