膀胱癌患者灌注治疗过程中尿路感染危险因素分析及预测模型构建  

Risk factors of urinary tract infection during perfusion therapy for bladder cancer patients and construction of a prediction model

作　　者：符海能 梁培育[1] 王声兴[1] 符津山 FU Haineng;LIANG Peiyu;WANG Shengxing;FU Jinshan(Department of Urology,The First Affiliated Hospital of Hainan Medical College,Haikou 570102,China)

机构地区：[1]海南医学院第一附属医院泌尿外科,海南海口570102

出　　处：《现代泌尿外科杂志》2022年第10期821-826,共6页Journal of Modern Urology

基　　金：海南省重点研发计划项目(No.ZDYF2019122);海南省自然科学基金项目(No.20158298)。

摘　　要：目的 分析膀胱癌患者灌注治疗过程中尿路感染发生的危险因素并构建危险因素的预测模型。方法 收集2020年2-12月于海南医学院第一附属医院接受灌注治疗的160例膀胱癌患者的临床资料。多因素logistic回归模型分析尿路感染发生的有关因素;受试者工作特征(ROC)曲线分析各有关因素预测尿路感染的曲线下面积和最佳截断值;R语言软件4.0“RMS”包构建预测尿路感染的列线图模型。结果 尿路感染组患者年龄、糖尿病比例、灌注治疗次数、留置导尿管比例、导尿管留置时间、多个肿瘤位点比例、手术时间均大于无尿路感染组;尿路感染组患者白蛋白水平低于无尿路感染组。年龄、白蛋白水平、灌注治疗次数、导尿管留置时间、手术时间的AUC分别为0.878、0.873、0.802、0.864、0.836;最佳截断值分别为66.57岁、37.60 g/L、7.26次、8.31 d、95.87 min。糖尿病、白蛋白水平、灌注治疗次数、留置导尿管、肿瘤位点是尿路感染的独立危险因素。Nomogram模型预测尿路感染发生风险C-index为0.723(95%CI:0.686~0.770),风险阈值>0.10,可提供临床净收益;此外,Nomogram模型临床净收益均高于糖尿病、白蛋白水平、灌注治疗次数、留置导尿管、肿瘤位点。结论 本研究基于糖尿病、白蛋白水平、灌注治疗次数、留置导尿管、肿瘤位点构建了预测膀胱癌患者灌注治疗过程中尿路感染发生风险的Nomogram模型,对尿路感染发生的预测具有重要意义。Objective To analyze the risk factors of urinary tract infection(UTI) during perfusion therapy for bladder cancer patients and to construct a prediction model.Methods A total of 160 patients who received perfusion therapy in our hospital during Feb.2020 and Dec.2020 were selected as the research subjects.The clinical data were collected, and related factors of UTI were analyzed with Logistic regression model.Receiver operating characteristic(ROC) curve was used to analyze the area under the curve(AUC) and the optimum cut-off value for predicting UTI.R language software 4.0 “RMS” package was used to build the nomogram.Results Age, incidence of diabetes mellitus, times of perfusion therapy, use of indwelling catheter, catheter indwelling time, multiple tumor sites and operation time of UTI patients were significantly higher than those of non-UTI patients, while the albumin level of UTI patients was lower.The AUC of age, albumin level, times of perfusion therapy, catheter indwelling time and operation time were 0.878, 0.873, 0.802, 0.864 and 0.836, respectively.The optimum cut-off values were 66.57 years, 37.60 g/L, 7.26 times, 8.31 days and 95.87 min, respectively.Diabetes mellitus, albumin level, frequency of perfusion therapy, indwelling catheter and tumor site were independent risk factors for UTI.The nomogram predicted a C-index of 0.723(95%CI: 0.686-0.770) for UTI, with a risk threshold of >0.10, providing a net clinical benefit.In addition, the net clinical benefit of the nomogram was higher than that of diabetes mellitus, albumin level, number of perfusion therapy, indwelling catheter and tumor site.Conclusion This study constructed a nomogram to predict the risks of UTI during perfusion therapy for bladder cancer patients based on diabetes mellitus, albumin level, frequency of perfusion, indwelling catheter and tumor site, which would be of great significance.

关 键 词：膀胱癌 灌注治疗 尿路感染 危险因素 预测模型 

分 类 号：R737.14[医药卫生—肿瘤]