闽西地区感染性结石的危险因素分析及预测模型构建  

Analysis of risk factors and construction of prediction model for infectious stones in western Fujian Province

作　　者：何泉芳 王沛昌 杨华旭 唐华[2] 陈启富 王德明 林金松 林叶福 HE Quanfang;WANG Peichang;YANG Huaxu;TANG Hua;CHEN Qifu;WANG Deming;LIN Jinsong;LIN Yefu(Department of Urological Surgery,Longyan People′s Hospital Affiliated to Xiamen Medical University,Longyan,Fujian,364000;Department of Urological Surgery,Sanming Second Hospital Affiliated to Fujian University of Traditional Chinese Medicine,Sanming,Fujian,365000)

机构地区：[1]厦门医学院附属龙岩人民医院泌尿外科,福建龙岩364000 [2]福建中医药大学附属三明市第二医院泌尿外科,福建三明365000

出　　处：《实用临床医药杂志》2024年第10期24-28,34,共6页Journal of Clinical Medicine in Practice

基　　金：福建省自然科学基金项目(2021J01537);福建省龙岩市科技计划项目(2022LYF17005)。

摘　　要：目的探讨闽西地区居民感染性结石形成的危险因素,据此构建术前预测感染性结石风险的列线图模型。方法分析2021年10月—2023年11月在厦门医学院附属龙岩人民医院接受泌尿系结石相关治疗的204例患者的临床资料,患者均接受结石成分分析。采用单因素和多因素Logistic回归分析筛选感染性结石的独立危险因素,构建预测感染性结石风险的列线图模型,并评估模型的区分度和准确性。结果根据结石成分分析结果,204例患者分为感染性结石组56例和非感染性结石组148例。单因素和多因素Logistic回归分析结果显示,女性(OR=2.602,95%CI:0.766~8.842,P=0.039)、糖尿病(OR=9.751,95%CI:2.547~17.332,P=0.001)、结石长径长(OR=1.123,95%CI:1.046~4.207,P=0.031)、肾积水程度为中重度(OR=4.173,95%CI:1.256~13.865,P=0.020)、尿pH值高(OR=6.078,95%CI:1.922~19.216,P=0.002)、尿培养阳性(OR=6.060,95%CI:1.398~16.262,P=0.016)是感染性结石的独立危险因素。基于独立危险因素构建预测感染性结石风险的列线图模型,该模型预测感染性结石的曲线下面积为0.860,显著大于性别、糖尿病、结石长径、肾积水程度、尿pH值、尿培养结果预测的曲线下面积(P<0.05);该模型经1000次内部重抽样验证,平均绝对误差为1.8%,提示其预测的感染性结石风险与实际情况一致性较高。结论女性、糖尿病、结石长径长、肾积水程度为中重度、尿pH值高、尿培养阳性是闽西地区居民发生感染性结石的独立危险因素,据此构建的列线图模型可在术前预测感染性结石风险,且具有较高的预测效能。Objective To investigate the risk factors for the formation of infectious stones among residents in western Fujian Province and construct a nomogram model for preoperative prediction of the risk of infectious stones.Methods Clinical data of 204 patients who received treatment for urinary tract stones at Longyan People′s Hospital Affiliated to Xiamen Medical University from October 2021 to November 2023 were analyzed.All patients underwent stone composition analysis.Univariate and multivariate Logistic regression analysis were used to screen independent risk factors for infectious stones,construct a nomogram model for predicting the risk of infectious stones,and the discriminative power and accuracy of the model was evaluated.Results Based on the results of stone composition analysis,204 patients were divided into infectious stone group(56 cases)and non-infectious stone group(148 cases).Univariate and multivariate Logistic regression analysis showed that female(OR=2.602,95%CI,0.766 to 8.842,P=0.039),diabetes(OR=9.751,95%CI,2.547 to 17.332,P=0.001),long diameter of stones(OR=1.123,95%CI,1.046 to 4.207,P=0.031),moderate to severe hydronephrosis(0R=4.173,95%CI,1.256 to 13.865,P=0.020),high urine pH value(0R=6.078,95%CI,1.922 to 19.216,P=0.002),and positive urine culture(0R=6.060,95%Cl,1.398 to 16.262,P=0.016)were independent risk factors for infectious stones.A nomogram model for predicting the risk of infectious stones was constructed based on independent risk factors.The area under the curve of this model for predicting infectious stones was 0.860,which was significantly larger than the area under the curve predicted by gender,diabetes,stone long diameter,degree of hydronephrosis,urine pH value,and urine culture results(P<0.05).The model was validated by 1000 internal resampling,with an average absolute error of 1.8%,indicating a high consistency between the predicted risk of infectious stones and the actual situation.Conclusion Female,diabetes,long diameter of stones,moderate to severe hydronephrosis,high urine

关 键 词：感染性结石 列线图模型 危险因素 肾积水 糖尿病 

分 类 号：R692.4[医药卫生—泌尿科学] R44[医药卫生—外科学] R319[医药卫生—临床医学]