Identification Of Risk Factors,Pathogenic Bacteria And Predictive Model For Catheter Related Blood Stream Infection In Maintenance Hemodialysis Patients

Objective:In this study,the purpose of this study is to analyze the risk factors and causative organisms for the occurrence of CRBSI in patients on maintenance hemodialysis with CVC,and to build an infection risk prediction model to provide new ideas for clinical diagnosis and treatment and prevention of CRBSI.Methods:For the retrospective study of all patients who underwent MHD with CVC while hospitalized in the Department of Nephrology at the First Affiliated Hospital of Gannan Medical College between January 2019 and December 2021.260 patients were included according to inclusion,exclusion and CRBSI diagnostic criteria,of which 107 patients developed infection as the infected group.Other 153 patients did not develop infection and were the control group.In each patient,general data such as age,dialysis age and placement site were collected,and laboratory data such as WBC,HDL,CRP,blood culture results and drug sensitivity were collected.All the above data were statistically analyzed for pathogenic bacteria causing CRBSI and their drug sensitivity,screening risk factors for the occurrence of CRBSI,and constructing the prediction model for infection.Results:Of 107 patients in the infection group,pathogenic microorganisms were cultured in 79 patients,with positive rate of 73.83%.The causal microorganisms included 43Gram-negative bacterium（54.43%）,34 Gram-positive bacterium（43.04%）,and 2fungi（2.53%）.Gram-positive bacterium was dominated by Staphylococcus aureus（24.05%）and Gram-negative bacterium was dominated by Escherichia coli（10.13%）.All Gram-positive bacteria were not resistant to VA,TGC,TEC,LZD,and the sensitivity rate was 100%.Gram-negative bacteria with high susceptibility to antibiotics were SCF,IPM,CAZ,FEP and TZP,with 83.72%for all the top 4antibiotics and 81.40%for TZP.Fungi were all sensitive to FLU and VOR with 100%sensitivity rate.There were 124 antibiotics used in the clinical treatment of Gram pathogenic bacteria,with the top 3 commonly used antibiotics being TZP（41.94%）,VA（22.58%）,and SCF（7.26%）.Both groups of patients were statistically analyzed for clinical data,and the statistically significant indicators of univariate analysis results were diabetes,dialysis age,WBC,NEUT,NEUT%,ALB,CHO,HDL,LDL,FER,VB₁₂,and CRP（P<0.05）.Multifactorial Logjstic regression analysis revealed that diabetes,dialysis age,WBC,HDL,VB₁₂,and CRP were independent risk factors for the development of CRBSI（P<0.05）.The predictive model was constructed based on the results of the multifactorial analysis.the ROC curve indicated that the AUC was 0.854 and the 95%CI was 0.807to 0.900.Mapped nomograms of risk of CRBSI occurrence,with calibration results showing good agreement.Clinical decision curves indicated net benefits of up to 0.30,and the use of the model to predict the risk of infection occurrence might add additional clinical benefit if the probability threshold for the occurrence of CRBSI was not less than 2%.Conclusion:1.Most common causative organism of CRBSI in patients with MHD using CVC was Staphylococcus aureus,followed by Escherichia coli,with an increasing trend of pathogenicity of Gram-negative bacteria.Gram-positive bacteria were sensitive to VA,TGC,TEC and LZD,Gram-negative bacteria were sensitive to SCF,IPM,CAZ,FEP and TZP,and fungi were sensitive to FLU and VOR.The antibiotics most commonly used for clinical treatment were TZP,followed by VA.2.The combination of diabetes,older age on dialysis,elevated WBC,elevated CRP,lower HDL,and lower VB₁₂were independent risk factors for the development of CRBSI in patients with MHD using CVC.3.In the current study,an innovative and effective nomogram of the risk of CRBSI was developed,combining six predictors of WBC,HDL,VB₁₂,CRP,diabetes,and age on dialysis,which could rapidly predict the risk of CRBSI in patients undergoing MHD with CVC.