Construction And Verification Of Risk Nomogram Model For Ureteral Calculi Complicated With Urogenic Sepsis

Objective : To analyze the relevant risk factors of ureteral calculi complicated with urogenic sepsis.The risk nomogram model of the disease was constructed and verified to provide a theoretical basis for identifying high-risk groups and taking clinical measures in time.Methods:The clinical data of 395 patients with ureteral calculi admitted to the 940 th Hospital of the Joint Logistics Support Force of the Chinese People’s Liberation Army from January 2016 to August 2021 were collected.According to the presence or absence of urogenic sepsis,the patients were divided into sepsis group and non-sepsis group.The data of part of the patients were used to construct the model(the modeling Group),and the other data of patients were used to verify the model(the validation group)based on the duration of hospital stay.Univariate analysis on the modeling group was used to find statistically different factors,and multivariate Logistic regression analysis on these factors was used to identify the predictive factors and construct the logistic regression predictive model.R statistical software was used to visualize and verify the model for the next step.Finally,the receiver operating characteristic curve(ROC),Gi Vi TI calibration belt and decision curve were used to evaluate the discrimination,calibration and clinical validity of the risk nomogram model,respectively.Results: 1.The univariate analysis showed that factors such as gender,age,history of diabetes,transverse diameter of ureteral calculi,CT value,perirenal exudation,urine white blood cell count,creatinine,platelet count,fibrinogen and other factors were significantly different between the sepsis group and the non-septic group(P<0.05).2.Multivariate Logistic regression analysis showed 6 independent risk Factors of ureteral calculi complicated with urinary sepsis: age(OR=3.787,95%CI:1.155-12.421,P=0.028),gender(OR=9.999,95%CI: 2.900-34.476,P<0.001),stone transverse diameter(OR=4.340,95%CI: 1.371-13.739,P=0.013),platelet count(OR=6.400,95%CI: 1.994-20.546,P=0.002),FIB(OR =12.403,95%CI: 3.66-42.039,P<0.001),urine white blood cell count(OR=8.221,95%CI: 2.487-27.171,P=0.001).3.Logistic regression prediction model was successfully constructed and visualized by the risk nomogram model.4.Evaluation of the risk nomogram model: Discrimination in the modeling group and the validation group: area under curve(AUC)=0.963(95%CI:0.941-0.985),AUC=0.894(95%CI:0.823-0.956);Calibration degree: The Gi Vi TI calibration belt based on the predictors in two groups did not cross the 45° diagonal bisector in the 80% and 95%confidence interval regions.P values of two groups were 0.946 and 0.224,respectively;Clinical validity: The decision curve based on the predictive factors was within the threshold probability ranges of 1%-98% and 3%-79%,respectively.The net benefit of patients with ureteral calculi was always higher than the other two extreme curves.It showed that the established risk nomogram model has good clinical validity.Conclusion: 1.Female,age≥60 years old,urine white blood cell count≥2+,Platelet count<150×109/L,FIB≥4.04g/L and ureteral calculus transverse diameter≥7mm were the risk factors of the ureteral calculi complicated with urogenic sepsis which were required more attention in clinical work.2.A risk nomogram prediction model for ureteral calculi complicated with urinay septicemia was successfully constructed.All evaluations of the model showed good results.The operation of the model was convenient and could be helpful for the early detection of high-risk individuals.