| Objective: Analyze the risk factors for Catheter-related blood stream infection(CRBSI)in patients undergoing deep vein indwelling hemodialysis,establish a risk prediction model for CRBSI,and conduct internal and external validation to provide reference for effective prediction and reduction of CRBSI.Methods:(1)Through a systematic review of literature,literature screening was conducted according to inclusion and exclusion criteria.After quality evaluation of the screened literature,the risk factors of CRBSI were summarized.Combined with expert group discussions,the risk prediction factors of CRBSI were determined and a risk factor questionnaire was developed.(2)Patients who met the inclusion and exclusion criteria at a tertiary A hospital in Changsha from January 2017 to December 2021 were selected as the research subjects of the modeling group.They were randomly divided into a training set and a validation set in a 7:3 ratio.The training set was used for model construction,while the validation set was used for internal validation of the model.Patients who met the inclusion criteria in the same hospital from February 2022 to November 2022 were selected as the validation group research subjects for external validation of the model.(3)SPSS25.0 and R-4.1.0 were used for data analysis.Measurement data conforming to normal distribution were expressed by(`x ± s),and the difference between groups was compared by t test;The measurement data of non normal distribution is represented by Median(P25,P 75),and the difference between groups is compared by Mann Whitney U test.The counting data is represented by frequency and percentage,and the differences between groups are compared using chi square test.Single factor analysis showed statistically significant differences with P<0.1.Significant factors were included in multivariate logistic regression analysis,and variables were screened using the forward backward method based on AIC criteria.A logistic regression model was established and presented in the form of a column chart.The area under the receiver operating characteristic(ROC)curve(AUC)was used to evaluate the discrimination of the model,and Hosmer-Lemeshow Goodness of fit was used to test the calibration of the evaluation model.Results:(1)A total of 30 articles were screened(7 guides,1 expert consensus,1 systematic evaluation,14 case control studies,and 7 cohort study);After quality evaluation of the included literature,combined with expert group discussions,a total of 56 risk prediction factors were formed from 3 aspects.(2)A total of 756 patients were included in the modeling group,including 485 males(64.2%)and 271 females(35.8%).Among them,64 patients developed CRBSI,with an incidence rate of8.5%.Among the 64 patients,60 were positive for catheter blood culture,including 47 cases of Gram positive bacteria(78.3%),12 cases of Gram negative bacteria(20.0%),and 1 case of fungi(1.7%).Among Gram positive bacteria,Staphylococcus aureus and Staphylococcus epidermidis were the most common,with 24 strains(40.0%)and 16 strains(26.7%),respectively;Escherichia coli is the most common Gram negative bacteria,with a total of 5 strains(8.3%);Only one fungus is a nearly smooth Candida.(3)The results of univariate analysis showed that there were statistically significant differences in medical expense payment mode,diabetes,coronary heart disease,catheter infection history,dialysis age,catheter retention time,white blood cell count,neutrophil count,neutrophil percentage,globulin,high-density lipoprotein,C-reactive protein,pro Calcitonin,sodium,chlorine,and phosphorus(P<0.1).(4)Multivariate logistic regression analysis showed that diabetes(OR=8.60,95% CI:3.94-20.77),dialysis age(OR=2.47,95% CI: 1.16-5.38),catheter indwelling time(OR=5.19,95% CI: 2.44-11.63),procalcitonin(OR=1.02,95% CI: 1.01-1.03),C-reactive protein(OR=1.01,95% CI: 1.00-1.01)were independent risk factors for CRBSI in patients with deep vein indwelling hemodialysis catheter(P<0.05).(5)CRBSI risk prediction model: Logit(P)=-5.39 + 2.15 × Diabetes + 0.90 × Dialysis age + 1.65× Catheter retention time + 0.02 × Calcitonin + 0.01 × C-reactive protein.The AUC of the model is 0.88(95% CI: 0.83~0.93),and the results of the Hosmer-Lemeshow showed that χ~2=5,P=0.7,P>0.05,indicating that the model has good discrimination and calibration.The internal validation of the model was conducted using validation set data,and the results showed an AUC of 0.82(95% CI: 0.73~0.90).The Hosmer-Lemeshow results showed that χ~2=11,P=0.2,P>0.05,indicating that the model had a good predictive effect in the validation set.(6)A total of 247 patients were included in the validation group,including 159 males(64.4%)and 88 females(35.6%).Among them,7 patients developed CRBSI,with a CRBSI incidence rate of 2.8%.All 7 patients showed positive results in catheter blood culture,with 5 cases(71.4%)showing Gram positive bacteria,with Staphylococcus aureus accounting for the largest proportion,totaling 4 strains(57.1%);Gram negative bacteria were found in 2 cases(28.6%),all of which were Stenotrophomonas maltophilia.(7)The validation group data was used for external validation of the model,and the results showed an AUC of 0.80(95% CI: 0.58-1.00).The Hosmer-Lemeshow results showed that χ~2=13,P=0.1,P>0.05,indicating that the model had better predictive performance in the validation group.Conclusion:(1)In this study,there were 56 risk predictors for bloodstream infection related to deep vein indwelling hemodialysis catheters,including general information,hemodialysis related information,and laboratory indicators.(2)The incidence of CRBSI in the modeling group of this study was 8.5%,while the incidence of CRBSI in the validation group was 2.8%.Complicated diabetes,dialysis age ≥ 1 year,catheter retention time ≥ 14 days,elevated procalcitonin levels,and elevated C-reactive protein levels are independent risk factors for CRBSI in patients with deep vein indwelling hemodialysis catheters.(3)The CRBSI risk prediction model constructed in this study has good predictive performance,providing a simple,intuitive,and highly operable predictive tool for clinical prediction of CRBSI in patients with deep vein indwelling hemodialysis catheters. |
PDF Full Text Request