Culture And Identification And Observation Of Blood Pathogenic Bacteria And Construction Of A Prognostic Nomogram Model For Elderly Patients With Bloodstream Infection

Objective: To analyze the pathogenic characteristics,establish and verify a nomogram model for predicting the risk of poor prognosis in elderly patients with bloodstream infection.Methods: From December 2018 to December 2021,the date of 656 elderly inpatients(≥ 65 years old)with bloodstream infection in the Affiliated Hospital of Southwest Medical University were retrospectively analyzed,including the results of blood culture and clinical characteristics.The subjects were randomly divided into the training cohort(n=457)and validation cohort(n=199)at a ratio of 7:3.The training cohort was divided into poor prognosis group and favorable prognosis group according to the prognosis.Adverse events nomogram prediction model was constructed by univariate analysis and multivariate regression model analysis.The receiver operating characteristic curve(ROC),calibration curve,and decision curve analysis(DCA)were adopted to evaluate the discrimination,calibration,and clinical validity of the nomogram.Results: A total of 656 elderly patients with bloodstream infection were included in the study.Men were slightly more than women,accounting for 55.8%.A total of 682 non-repetitive pathogenic bacteria were detected,including 437 gram-negative bacteria(64.1%,437/682),221 gram-positive bacteria(32.4%,221/682)and 26 fungi(3.8%,26/682).The common pathogenic bacteria were Escherichia coli(37.4%,255/682),Klebsiella pneumoniae(12.0%,82/682)and Staphylococcus aureus(5.1%,35/682).Multivariate regression analysis showed that infectious shock(OR=3.705,P<0.001),admission to intensive care unit(ICU)(OR=5.343,P<0.001),elevated blood urea nitrogen(OR=1.014,P=0.007),elevated D-dimer(OR=1.041,P=0.024),elevated NT-pro BNP(OR=1.000,P=0.023),and decreased plasma albumen(OR=0.900,P=0.043)were independent predictors for adverse prognosis in elderly patients with bloodstream infection.The visual nomogram was drawn according to the six variables.The AUC,Brier score,and probability interval of net benefit threshold of the nomogram prediction model in the training set and validation set were 0.879(95% CI: 0.844～0.915),0.909(95% CI:0.864～0.955);0.119,0.114;and 2.2% ～ 98.1%,8.8% ～ 97.8%,respectively.Conclusion: Gram-negative bacteria play a significant role in the bloodstream infections of elderly patients.Escherichia coli,Klebsiella pneumoniae and Staphylococcus aureus were the most common pathogenic bacteria.The nomogram model containing six predictors(septic shock,admission to intensive care unit,blood urea nitrogen,D-dimer,NT-pro BNP,and albumin)could be used to predict the risk of adverse prognostic events in elderly patients with bloodstream infection.The nomogram prediction model in this study has certain application value and contributes to recognizing high-risk elderly inpatients with bloodstream infection and assist medical staff to make clinical strategies.