断指再植患者再植成活率的风险预测模型构建

Construction of Risk Prediction Model for Replantation Survival Rate in Patients with Severed Finger Replantation

目的:构建断指再植患者再植成活率的风险预测模型,为断指再植患者术后再植成活率提升提供依据。方法:采用便利抽样法收集2015年12月—2021年12月期间在我院接受断指再植术治疗患者330例,将其中230例患者据患者术后再植是否成活分组,成活组(n=198)、未成活组(n=32)。并以表格方式记录影响断指再植患者再植成活率的相关因子,如年龄、性别、受伤类型、离断平面、离断程度等因子。首先行单因子分析,再行多因子logistic回归分析,以构建风险预测模型。采用该模型对剩余100例患者评估,验证风险预测模型的效能。结果:最终得到风险预测模型公式:Z=1.625×年龄40岁~60岁赋值+0.525×末节指骨赋值+0.925×完全离断赋值+1.258×热缺血时间≥6h赋值+0.856×出现血管危象赋值+1.156×术后指腹弹性差赋值+1.769×术后皮肤温度低赋值-0.452。通过ROC曲线评价预测效能,AUC为0.870,95%CI[0.789,0.952],P<0.001;灵敏度为0.852,特异度为0.863;且本次模型的预测准确率为96.00%。结论:断指再植成活率受多种因子影响,通过纳入影响因子构建风险预测模型,且该模型具有良好的预测效能,准确率高,值得临床应用。

Objective:To establish a risk prediction model for the survival rate of replantation in patients with severed finger replantation,and to provide evidence for the improvement of the survival rate after replantation.Methods:A total of 330 patients who received replantation of severed finger in our hospital from December 2015 to December 2021 were collected by convenience sampling method.Among them,230 patients were divided into survival group(n=198)and non-survival group(n=32)according to whether the patients survived replantation.Related factors affecting the replantation survival rate of patients with replantation were recorded in table form,such as age,gender,injury type,dissociation plane,dissociation degree,etc.Firstly,single factor analysis was performed,followed by multiple logistic regression analysis to construct the risk prediction model.The model was used to evaluate the remaining 100 patients to verify the efficacy of the risk prediction model.Results:The formula of risk prediction model was finally obtained:Z=1.625×age 40~60 years old+0.525×terminal phalangeal bone+0.925×complete detachment+1.258×heat ischemia time≥6h+0.856×vascular crisis+1.156×postoperative elasticity difference of finger abdomen+1.769×postoperative low skin temperature-0.452.The ROC curve was used to evaluate the prediction efficiency,and the AUC was 0.870,95%CI[0.789,0.952],P<0.001.The sensitivity and specificity were 0.852 and 0.863 respectively.And the prediction accuracy of this model is 96.00%.Conclusion:The survival rate of replantation of severed finger is affected by various factors.The risk prediction model was built by incorporating the influencing factors,and the model has good prediction efficiency and high accuracy,and is worthy of clinical application.