神经源性膀胱上尿路损害风险评估工具的研制

Development of the risk evaluation tool for upper urinary tract damage of neurogenic bladder

目的研制神经源性膀胱上尿路损害风险评估工具,以早期发现危险人群。方法调查112例神经源性膀胱患者的一般资料、病情资料、尿动力学和泌尿系影像学资料,分别建立Logistic回归和决策树模型,根据准确度和受试者工作特征曲线下面积较高的模型形成风险评估工具。结果决策树模型的准确度和受试者工作特征曲线下面积(84.8%,0.909)高于Logistic回归模型(73.2%,0.842),据此形成了包含尿道功能、性别、最大腹压、最大膀胱内压4个指标的风险评估表。结论本研究研制的神经源性膀胱上尿路损害风险评估表,为临床筛选危险人群提供了一种简便易行的工具。

Objective To develop a risk evaluation tool for upper urinary tract damage （UUTD） of neurogenic bladder （NGB） to identify population at risk in the early stage. Methods The general,elinical,urodynamie,urinary imaging data of 112 NGB patients were retrospectively investigated. Logistic regression and decision tree were used to establish the risk early warning model respectively. The risk evaluation tool was formed according to the model with a higher accuracy and area under curve. Results The accuracy and area under curve of decision tree model （84.8%,0.909）was higher than that of logistic regression model （73.2%,0.842）. A risk evaluation scale including urinary tract funetion,gender,Pabd max,Pves max was formed. Conclusion We developed a risk evaluation scale to predict the risk of UUTD in patients with NGB. It might provide a convenient way to screen UUTD in NGB patients.