术前尿培养阴性上尿路结石患者逆行肾内手术后尿脓毒症的列线图模型构建及验证  

Construction and verification of the nomogram model of urinary sepsis in patients with preoperative urine culture negative upper urinary tract stones after retrograde intrarenal surgery

作　　者：彭坚 罗贝 徐英峰 张新乐 马效伦 PENG Jian;LUO Bei;XU Yingfeng;ZHANG Xinle;MA Xiaolun(Department of Urology,Kunshan Hospital of Integrated Traditional Chinese and Western Medicine,Suzhou Jiangsu 215331,China)

机构地区：[1]昆山市中西医结合医院泌尿外科,江苏苏州215331

出　　处：《中国急救复苏与灾害医学杂志》2025年第3期348-352,共5页China Journal of Emergency Resuscitation and Disaster Medicine

基　　金：2019年高层次卫生人才“六个一工程”拔尖人才科研项目(LGY201934);2024年昆山市级科技专项立项项目(KS2449)。

摘　　要：目的建立和验证一个术前列线图,用于预测逆行肾内手术(RIRS)术前尿液培养阴性患者经RIRS治疗上尿路结石术后发生尿脓毒症的风险。方法回顾性招募2022年1月—2024年3月于昆山市中西医结合医院接受RIRS治疗上尿路结石的700例术前尿培养阴性患者,根据术后是否发生尿脓毒症分为阳性组42例与阴性组658例。比较两组患者的一般临床资料、结石特征、尿培养和手术情况,多因素Logistic回归模型筛选术后尿脓毒症的危险因素,R软件建立列线图模型,采用1000次自我重复采样进行内部验证,包括受试者工作特征(ROC)曲线计算曲线下面积(AUC)、校准曲线和决策曲线。结果单因素比较发现,与阴性组相比,阳性组糖尿病、复发性尿路感染史和结石手术史、孤立性肾结石、术前尿亚硝酸盐阳性和手术时间≥75 min患者比例明显增多,结石体积增大,术后住院时间延长(P<0.05)。Logistic回归显示,孤立性肾结石(OR=1.896,95%CI=1.235~2.326,P=0.003)、复发性尿路感染史(OR=2.526,95%CI=2.001~3.123,P<0.001)、尿亚硝酸盐阳性(OR=3.326,95%CI=2.758~3.659,P<0.001)、糖尿病(OR=1.426,95%CI=1.123~1.958,P=0.006)和手术时间≥75 min(OR=2.102,95%CI=1.659~2.595,P=0.001)是术后发生尿脓毒症的危险因素。列线图总分220分,ROC显示列线图预测术后尿脓毒症的AUC为0.876(95%CI=0.822~0.935,P<0.001),提示模型的预测区分度较好。校正曲线和决策曲线也显示模型有较好的吻合度和临床净获益比。结论结合术前独立风险因素的列线图可以预测术前尿培养阴性的上尿路结石患者RIRS术后尿脓毒症的发生风险,可以帮助泌尿科医生在手术后提前采取预防措施,避免更严重的感染并发症。Objective To establish and validate a nomogram for predicting the risk of urinary sepsis in patients upper urinary tract stones with preoperative urine culture negative treated by RIRS.Methods 700 patients with preoperative negative urine culture who underwent RIRS treatment for upper urinary tract stones in Kunshan Hospital of Integrated Traditional Chinese and Western Medicine from January 2022 to March 2024 were retrospectively recruited.Based on the occurrence of postoperative urinary sepsis,they were divided into positive group of 42 cases and negative group of 658 cases.The general clinical data,stone characteristics,urine culture,and surgical conditions between the two groups of patients,then multivariate Logistic regression model was used to screen risk factors for postoperative urinary sepsis were compared.A nomogram model was established using R software,and internal validation was conducted using 1000 self repeated sampling,including receiver operating characteristic(ROC)curve for area under curve(AUC),calibration curve and decision curve.Results Single factor comparison showed that diabetes,recurrent urinary tract infection and stone surgery history,solitary kidney stones,preoperative urinary nitrite positive and operation time≥75 min in positive group significantly increased,stone volume increased,and postoperative hospital stay prolonged(P<0.05).Logistic regression showed that solitary kidney stones(OR=1.896,95%CI=1.235-2.326,P=0.003),history of recurrent urinary tract infection(OR=2.526,95%CI=2.001-3.123,P<0.001),positive urine nitrite(OR=3.326,95%CI=2.758-3.659,P<0.001),diabetes(OR=1.426,95%CI=1.123-1.958,P=0.006),and operation time≥75 min(OR=2.102,95%CI=1.659-2.595,P=0.001)were the risk factors to postoperative urinary sepsis.The total score of nomogram was 220 points,ROC showed that AUC of the model for predicting postoperative urinary sepsis was 0.876(95%CI=0.822-0.935,P<0.001),indicating good predictive discrimination of the model.The calibration curve and decision curve also showed that t

关 键 词：逆行肾内手术 上尿路结石 尿脓毒症 列线图 亚硝酸盐 

分 类 号：R691.4[医药卫生—泌尿科学]