缺血性脑卒中患者介入术后尿潴留预测模型的构建与验证  

Construction and validation of a predictive model for post-interventional urinary retention in ischaemic stroke patients

作　　者：范学宾 魏丽丽[2] 姜文彬[2] 于水[2] 陈浩[3] 陈素芳 刘丽君[2] 张艳[2] FAN Xuebin;WEI Lili;JIANG Wenbin;YU Shui;CHEN Hao;CHEN Sufang;LIU Lijun;ZHANG Yan(Qingdao University,Qingdao 266071,China;Affiliated Hospital of Qingdao University,Qingdao 266000,China;Affiliated Hospital of Qingdao University(Pingdu),Qingdao 266700,China)

机构地区：[1]青岛大学,山东省青岛市266071 [2]青岛大学附属医院,山东省青岛市266000 [3]青岛大学附属医院(平度),山东省青岛市266700

出　　处：《护理实践与研究》2024年第11期1620-1628,共9页Nursing Practice and Research

基　　金：青岛市科学技术局自然科学基金(编号:23-2-1-142-zyyd-jch)。

摘　　要：目的探讨缺血性脑卒中患者介入术后发生尿潴留的风险因素,构建并验证缺血性脑卒中患者介入术后尿潴留预测模型,为筛查高危人群,进行早期干预提供参考依据。方法选取2023年5—8月在青岛市某三甲医院行介入手术的653例缺血性脑卒中住院患者为研究对象,按照7:3的比例进行随机分配,其中建模组442例,验证组211例。采用多因素Logistics回归分析筛选分析风险因素,构建列线图预测模型,并通过绘制受试者工作特征(ROC)曲线和决策曲线分析(DCA)评价模型的区分度及临床实用性。结果多因素Logistic回归分析结果表明,BMI(OR=1.248,95%CI:1.130~1.378,P<0.05)、术前排尿测试结果(OR=0.271,95%CI:0.132~0.554,P<0.05)、泌尿系统疾病史(OR=0.376,95%CI:0.170~0.830,P<0.05)、腰背部疼痛评分(OR=3.097,95%CI:2.161~4.439,P<0.05)、焦虑评分(OR=1.315,95%CI:1.108~1.562,P<0.05)是脑卒中患者介入术后发生尿潴留的影响因素。计算结果显示,ROC曲线下面积(AUC)及其95%CI为0.847(0.790~0.904)。DCA曲线显示该列线图具有良好的临床实用性。结论缺血性脑卒中患者介入术后尿潴留预测模型有利于术前筛选高危人群,依据预测结果制订针对性术前干预方案,可降低缺血性脑卒中介入术后患者尿潴留的发生率。Objective To explore the risk factors for urinary retention after intervention in ischaemic stroke patients,construct and validate a prediction model for urinary retention after intervention in ischaemic stroke patients,to provide a reference basis for screening high-risk groups and early intervention.Methods To select 653 ischemic stroke inpatients who underwent interventional procedures in a Class Ⅲ Grade A hospital in Qingdao from May to August 2023 as the study subjects and randomly assigned according to the ratio of 7:3,of which 442 cases were in the modelling group and 211 cases were in the validation group.Multifactor Logistic regression analysis was used to screen and analyse the risk factors,construct a column-line graph prediction model,and evaluate the differentiation and clinical utility of the model by plotting the Receiver Operating Curve(ROC)and Decision Curve Analysis(DCA).Results The results of multifactorial Logistic regression analysis showed that BMI(OR=1.248,95%CI:1.130~1.378,P<0.05),preoperative voiding test results(OR=0.271,95%CI:0.132~0.554,P<0.05),and a history of urological disease(OR=0.376,95%CI:0.170~0.830,P<0.05),low back pain score(OR=3.097,95%CI:2.161~4.439,P<0.05),and anxiety score(OR=1.315,95%CI:1.108~1.562,P<0.05)were the influencing factors for the occurrence of urinary retention in post-intervention stroke patients.Calculations showed that the area under the ROC curve(AUC)and the 95%CI was 0.847(0.790~0.904).The DCA curve showed that this column chart has good clinical utility.Conclusion The prediction model of post-interventional urinary retention in patients with ischemic stroke is helpful for preoperative screening of high-risk groups,and based on the prediction results,a targeted preoperative intervention plan could be formulated,which can reduce the incidence of urinary retention in patients with ischemic stroke after interventional surgery.

关 键 词：缺血性脑卒中 介入手术 尿潴留 风险因素 预测模型 

分 类 号：R473.54[医药卫生—护理学]