胸腹腔引流管拔除后手术切口愈合不良风险预测模型构建

Construction of a risk prediction model for poor healing of surgical incisions after removal of thoracic and abdominal drainage tubes

目的探讨行胸腹部手术后引流管拔除患者手术切口愈合不良发生的独立危险因素,建立手术切口愈合不良的风险预测模型。方法采用便利抽样法选取2020年7—12月青岛大学附属医院的545例胸腹部手术患者,根据患者是否发生切口愈合不良将其分为术后切口愈合不良组(n=87)和无切口愈合不良组(n=458)。采用Logistic回归分析对手术切口愈合不良的危险因素进行分析并构建风险预测模型,应用受试者工作特征(ROC)曲线的曲线下面积检验模型预测效果,并选取230例患者进行模型预测效果验证。结果本研究最终纳入渗液持续时间、血清白蛋白、有无切口感染、置管时渗液量和置管时间5个影响因素构建风险预测模型,模型公式为Z=4.608+4.855×渗液持续时间+3.173×血清白蛋白+3.739×切口感染+2.271×置管时渗液量+0.466×置管时间。本模型ROC曲线的曲线下面积为0.773(95%置信区间:0.678~0.868),Youden指数最大值为0.549,灵敏度为0.742,特异度为0.807。结论胸腹腔手术患者引流管拔除后切口愈合不良风险预测模型能较好地预测切口愈合不良的发生风险,可以为临床医护人员及时对高危患者采取预防性管理措施提供依据。

Objective To explore the independent risk factors of poor healing of surgical incisions in patients with drainage tube removal after thoracic and abdominal surgery and establish a risk prediction model for poor healing of surgical incisions.Methods Using the convenient sampling method,a total of 545 patients who underwent thoracic and abdominal surgery in the Affiliated Hospital of Qingdao University were selected from July to December 2020.The patients were divided into the poor wound healing group(n=87)and the non-incision poor healing group(n=458)according to whether they had poor wound healing.Logistic regression analysis was used to analyze the risk factors of poor healing of surgical incisions and build a risk prediction model.The receiver operating characteristic(ROC)area under the curve was used to test the model to predict the effect and 230 patients were selected to verify the model prediction effect.Results In this study,5 factors including duration of exudation,serum albumin,incision infection,the volume of exudation during catheterization and catheterization time were finally included to construct a risk prediction model.The model formula was Z=4.608+4.855×duration of exudation+3.173×serum albumin+3.739×infection of the incision+2.271×the volume of exudation during catheterization+0.466×catheterization time.The area under ROC curve of this model was 0.773(95%CI:0.678-0.868).The maximum value of Youden index was 0.549,the sensitivity was 0.742 and the specificity was 0.807.Conclusions The risk prediction model of poor incision healing after drainage tube removal for patients undergoing thoracic and abdominal surgery can better predict the risk of poor incision healing and provide a basis for clinical medical staff to take preventive management measures for high-risk patients in time.