断指再植患者再植指体功能恢复的临床预测模型  

Clinical Prediction Model of Replanted Finger Body Function Recovery in Patients with Severed Finger Replantation

作　　者：刘黎莉 成玲娜 曹燕河 吴艳丽 LIU Li-li;CHENG Ling-na;CAO Yan-he(Panyu TCM hospital of Guangzhou,Guangzhou,Guangdong 511400)

机构地区：[1]广州市番禺区中医院,广东广州511400

出　　处：《中国伤残医学》2022年第16期20-25,共6页Chinese Journal of Trauma and Disability Medicine

摘　　要：分析影响断指再植患者再植指体功能恢复的危险因素,并构建临床预测模型。 方法:2016年1月-2020年1月我院收治的断指病例331例,将其中231例病例根据患者术后12个月指体功能恢复情况分为中/差组(68 例)和优/良组(163 例) ,记录患者性别、年龄、受教育程度、职业类型、离断平面、受伤类型、内固定方式、拔除克氏针时间、术后心理状态、合理功能锻炼等因素,对相关因子进行单因素分析,并对有统计学意义的因子再进行多因素logistic回归分析,并建立风险预测模型。随后使用该模型对剩下的100例患者进行评估,结果采用受试者工作特征(receiver operating characteristic,ROC)曲线下面积(area under curve, AUC)进行统计分析,检验该风险预测模型的预测效能。结果:中/差组与优/良组相比,在性别、职业类型、受伤类型上不具有统计学意义(P>0.05);在年龄、受教育程度、离断平面、内固定方式、术后心理状态、合理功能锻炼这6项危险因素上具有统计学意义(P<0.05) ,对这些因素使用logistic回归分析构建断指再植患者再植指体功能恢复的临床预测模型:y=e^(Logit(P))/[1 +e^(Logit(P)) ]。使用模型对患者进行评估后显示模型总正确率为86.0%,通过ROC曲线评价该模型的鉴别效度,结果显示:AUC=0.870 ,敏感度为85.2%,特异度为86.3%。结论:断指再植患者再植指体功能恢复受多种因素影响,以此建立的临床预测模型对断指再植患者再植指体功能恢复预测具有良好的判别能力,值得临床应用与进一步研究。Objective:To analyze the risk factors that affect the functional recovery of replanted fingers in patients with severed finger replantation,and construct a clinical prediction model.Methods:From January 2016 to January 2020,331 cases of severed fingers were admitted to our hospital.Among them,231 cases were divided into moderate/poor group(68 cases)and excellent according to the recovery of finger body function at 12 months after operation.Good group(163 cases),record the patient's gender,age,education level,occupation type,fracture plane,injury type,intermal fixation method,time to remove Kirschner wire,postoperative mental state,reasonable functional exercise and other factors.A single factor analysis was performed on the relevant factors,and a multi-factor logistic regression analysis was performed on the statistically significant factors,and a risk prediction model was established.The model was then used to evaluate the remaining 100 patients,and the results were statistically analyzed using receiver operating characteristic(ROC)area under curve(AUC)to test the predictive power of the risk prediction model.Results:The medium/poor group was not statistically significant in gender,occupation type,and injury type compared with the excellent/good group( P>0.05);in terms of age,education level,separation plane,internal fixation method,and postoperative The six risk factors of mental state and reasonable functional exercise are statistically significant( P<0.05).Logistic regression analysis was used for these factors to construct a clinical predictive model for the functional recovery of replanted fingers in patients with severed finger replantation:y=eLogit( P)/[1+eLogit( P)].After evaluating patients with the model,the total accuracy of the model was shown to be 86.0%.The identification validity of the model was evaluated by the ROC curve.The results showed that:AUC=0.870,sensitivity of 85.2%,and specificity of 86.3%.Conclusion:The functional recovery of replanted fingers in patients with severed finger replantat

关 键 词：断指再植 指体功能恢复 危险因素 临床预测模型 延续护理 

分 类 号：R658.1[医药卫生—外科学]