59. Risk Factors for Recurrent Gram-Negative Bacterial Bloodstream Infections

Abstract Background Gram-negative bacteria bloodstream infections (GNB-BSI) are a significant cause of morbidity and mortality. Recurrent GNB-BSI is an incompletely understood phenomenon. In this study we identify risk factors for recurrent GNB-BSI. Methods Patients with GNB-BSI have been prospectively enrolled into the Bloodstream Infection Biorepository (BSIB) since 2002. From the BSIB, patients with >1 episode of GNB-BSI with the same bacterial species were identified. Chi-Square, Fisher Exact, and a multivariate linear regression models were used to identify clinical risk factors for recurrent GNB-BSI. Paired isolate samples from the initial and the recurrent episode of GNB-BSI in same patient underwent Pulsed Field Gel Electrophoresis (PFGE) to differentiate Relapse (paired isolates identical) from Reinfection (paired isolates different). Results Among the 1,423 unique patients with GNB-BSI enrolled from 2002- 2015, 60 (4.2%) experienced recurrent GNB-BSI with the same bacterial species. Median time to recurrent GNB-BSI was 133 d (IQR: 40-284.75 days). Causes of recurrent-GNB-BSI included Escherichia coli (38%), Klebsiella species (30%), Pseudomonas aeruginosa (12%), and Serratia marcescens (5%) and did not differ from causes of non-recurrent GNB-BSI (Figure 1). Risk factors for recurrent GNB-BSI included Black race (OR: 2.45 [CI: 1.43-4.20]), implanted cardiac device (OR: 2.39 [CI: 1.00-5.07]), and admission to surgical service (OR: 2.16 [CI 1.24-3.75]). Forty-eight isolate-pairs from 43 patients with recurrent GNB-BSI underwent PFGE, relapse occurred in 31 (65%) and reinfection occurred in 17 (35%). Risk factors for GNB-BSI relapse included cardiac device (OR: 3.7 [CI: 1.7-8.3]), and admission to surgical service (OR: 3.7 [CI:1.3-9.4]). Figure 1: Species Breakdown Proportional comparison of the Gram-negative bacterial species identified in patients with recurrent and non-recurrent bloodstream infections. Conclusion Recurrent GNB-BSI is an uncommon complication of GNB-BSI. Recurrent GNB-BSI is most often driven by relapse, as opposed to reinfection, and is associated with associated with black race, implanted cardiac devices and admission to surgical service. Disclosures Vance G. Fowler, Jr., MD, MHS, Achaogen (Consultant)Advanced Liquid Logics (Grant/Research Support)Affinergy (Consultant, Grant/Research Support)Affinium (Consultant)Akagera (Consultant)Allergan (Grant/Research Support)Amphliphi Biosciences (Consultant)Aridis (Consultant)Armata (Consultant)Basilea (Consultant, Grant/Research Support)Bayer (Consultant)C3J (Consultant)Cerexa (Consultant, Other Financial or Material Support, Educational fees)Contrafect (Consultant, Grant/Research Support)Debiopharm (Consultant, Other Financial or Material Support, Educational fees)Destiny (Consultant)Durata (Consultant, Other Financial or Material Support, educational fees)Genentech (Consultant, Grant/Research Support)Green Cross (Other Financial or Material Support, Educational fees)Integrated Biotherapeutics (Consultant)Janssen (Consultant, Grant/Research Support)Karius (Grant/Research Support)Locus (Grant/Research Support)Medical Biosurfaces (Grant/Research Support)Medicines Co. (Consultant)MedImmune (Consultant, Grant/Research Support)Merck (Grant/Research Support)NIH (Grant/Research Support)Novadigm (Consultant)Novartis (Consultant, Grant/Research Support)Pfizer (Grant/Research Support)Regeneron (Consultant, Grant/Research Support)sepsis diagnostics (Other Financial or Material Support, Pending patent for host gene expression signature diagnostic for sepsis.)Tetraphase (Consultant)Theravance (Consultant, Grant/Research Support, Other Financial or Material Support, Educational fees)Trius (Consultant)UpToDate (Other Financial or Material Support, Royalties)Valanbio (Consultant, Other Financial or Material Support, Stock options)xBiotech (Consultant)


Session: O-12. Endocarditis
Background. Drug use-associated infective endocarditis (DUA-IE) is typically treated with 4-6 weeks of in hospital intravenous antibiotics (IVA). Outpatient parenteral antimicrobial therapy (OPAT) and partial oral antibiotics (PO) may be as effective as IVA, though long-term outcomes and costs remain unknown. We evaluated the clinical outcomes and cost-effectiveness of four antibiotic treatment strategies for DUA-IE.
Methods. We used a validated microsimulation model to compare: 1) 4-6 weeks of inpatient IVA along with opioid detoxification, status quo (SQ); 2) 4-6 weeks of inpatient IVA along with inpatient addiction care services (ACS) which offers medications for opioid use disorder (SQ with ACS); 3) 3 weeks of inpatient IVA with ACS followed by OPAT (OPAT); and 4) 3 weeks of IVA with ACS followed by PO antibiotics (PO). We derived model inputs from clinical trials and observational cohorts. All patients were eligible for either in-home or post-acute care OPAT. Outcomes included life years (LYs), discounted costs, incremental cost-effectiveness ratios (ICERs), proportion of DUA-IE cured, and mortality attributable to DUA-IE. Costs ($US) were annually discounted at 3%. We performed probabilistic sensitivity analyses (PSA) to address uncertainty.
Results. The SQ scenario resulted in 18.64 LY at a cost of $416,800/person with 77.4% hospitalized DUA-IE patients cured and 5% of deaths in the population were attributable to DUA-IE. Life expectancy was extended by each strategy: 0.017y in SQ with ACS, 0.011 in OPAT, and 0.024 in PO. The PO strategy provided the highest cure rate (80.2%), compared to 77.9% in SQ with ACS and 78.5% in OPAT and X in SQ. OPAT was the least expensive strategy at $412,300/person, Compared to OPAT, PO had an ICER of $141,500/LY. Both SQ strategies provided worse clinical outcomes for money invested than either OPAT or PO (dominated). All scenarios decreased deaths attributable to DUA-IE compared to SQ. Findings were robust in PSA. Table 1 Selected cost and clinical outcomes comparing treatment strategies for drug-use associated infective endocarditis including the status quo, status quo with addiction care services, outpatient parenteral antimicrobial therapy, and partial oral antibiotics.
Conclusion. Treating DUA-IE with OPAT along with ACS increases the number of people completing treatment, decreases DUA-IE mortality, and is cost-saving compared to the status quo. The PO strategy also improves clinical outcomes, but may not be cost-effective at the willingness-to-pay threshold of $100,000. Background. Gram-negative bacteria bloodstream infections (GNB-BSI) are a significant cause of morbidity and mortality. Recurrent GNB-BSI is an incompletely understood phenomenon. In this study we identify risk factors for recurrent GNB-BSI.
Methods. Patients with GNB-BSI have been prospectively enrolled into the Bloodstream Infection Biorepository (BSIB) since 2002. From the BSIB, patients with >1 episode of GNB-BSI with the same bacterial species were identified. Chi-Square, Fisher Exact, and a multivariate linear regression models were used to identify clinical risk factors for recurrent GNB-BSI. Paired isolate samples from the initial and the recurrent episode of GNB-BSI in same patient underwent Pulsed Field Gel Electrophoresis (PFGE) to differentiate Relapse (paired isolates identical) from Reinfection (paired isolates different).
Results  Conclusion. Recurrent GNB-BSI is an uncommon complication of GNB-BSI. Recurrent GNB-BSI is most often driven by relapse, as opposed to reinfection, and is associated with associated with black race, implanted cardiac devices and admission to surgical service.  Background. There are several clinical tools that have been developed to predict the likelihood of extended-spectrum β-lactamase producing Enterobacterales; however, the creation of these tools included few patients with cancer or otherwise immunosuppressed. The objectives of this retrospective cohort study were to develop a decision tree and traditional risk score to predict ceftriaxone resistance in cancer patients with Escherichia coli (E. coli) bacteremia as well as to compare the predictive accuracy between the tools.

Creation and Comparison of a Machine Learning Decision Tree and Traditional Risk Score to Predict Ceftriaxone Resistance in Cancer Patients with E. coli Bacteremia
Methods. Adults age ≥ 18 years old with E. coli bacteremia at The University of Texas MD Anderson Cancer Center from 1/2018 to 12/2019 were included. Isolates recovered within 1 week from the same patient were excluded. The decision tree was constructed using classification and regression tree analysis, with a minimum node size of 10. The risk score was created using a multivariable logistic regression model derived by using stepwise variable selection with backward elimination at level 0.2. The decision tree and risk score statistical metrics were compared.
Results. A total of 629 E. coli isolates were screened, of which 580 isolates met criteria. Ceftriaxone-resistant (CRO-R) E. coli accounted for 36% of isolates. The machine learning-derived decision tree included 5 predictors whereas the logistic regression-derived risk score included 7 predictors. The risk score cutoff point of ≥ 5 points demonstrated the most optimized overall classification accuracy. The positive predictive value of the decision tree was higher than that of the risk score (88% vs 74%, respectively), but the area under the receiver operating characteristic curve and model accuracy of the risk score was higher than that of the decision tree (0.85 vs 0.73 and 82% vs 74%, respectively).  Conclusion. The decision tree and risk score can be used to determine the likelihood of whether a cancer patient with E. coli bacteremia has a CRO-R infection. In both clinical tools, the strongest predictor was a history of CRO-R E. coli colonization or infection in the last 6 months. The decision tree was more user-friendly, has fewer variables, and has a better positive predictive value in comparison to the risk score. However, the risk score has a significantly better discrimination and model accuracy than that of the decision tree. Background. The optimal duration of antimicrobial therapy for uncomplicated Pseudomonas aeruginosa bloodstream infection (BSI) is unknown. We compared the outcomes of short and prolonged courses of antimicrobial therapy in adults with uncomplicated pseudomonal BSI.
Methods. All patients with uncomplicated P. aeruginosa BSI admitted at a tertiary-care hospital from May 2016 to September 2020 were included. We compared the rate of recurrent P. aeruginosa infection and 30-day mortality among patients who underwent short (7-11 days) and prolonged (12-21 days) courses of antimicrobial therapy using propensity score analysis with the inverse probability of treatment weighting (IPTW) method.