818. Risk Factors for and Clinical Outcomes of Multi-Drug Resistant Gram Negative Bacterial Bloodstream Infections: Initial Results from a 12-year Prospective Cohort Study
Session: Poster Abstract Session: Bacteremia and Endocarditis
Friday, October 9, 2015
Room: Poster Hall
Background: The clinical impact of and risk factors for bloodstream infections (BSI) due to multi-drug resistant (MDR) Gram negative bacteria (GNB) are incompletely understood.

Methods: From 2002-2015, all adult, non-neutropenic inpatients with monomicrobial GNB BSI were prospectively enrolled at Duke University Hospital. MDR was defined as resistant to ≥3 antibiotic classes. Risk factors and clinical outcomes associated with MDR status were identified.

Results: 1543 unique patients were prospectively enrolled during the study period.  Of these, 422 (27%) had MDR bacteria. The most common causes of BSI were Escherichia coli (38%), Klebsiella species (21%), Pseudomonas aeruginosa (10%), and Enterobacter species (7%). MDR organisms were more common in patients with hematological or solid-organ transplants (70/205 [34%] vs. 352/1338 [26%]; P = 0.02). In a multivariable logistic regression analysis, MDR phenotype was associated with dialysis dependence (odds ratio 1.57; 95% confidence interval 1.07 – 1.32), transplant (1.50; 1.00 – 1.23), and malignancy (1.39; 1.08 – 1.79). In-hospital mortality (19% vs 21%, P = 0.79) and presence of complications (acute kidney injury, septic shock, acute respiratory distress syndrome, shock liver, disseminated intravascular coagulation, stroke, and/or death) (48% vs 50%, P = 0.30) did not differ between MDR and non-MDR groups, though hospital length of stay (LOS) was longer in the MDR group (mean 20.8 days [Standard deviation 32.7] vs 15.7 [21.8]; P = 0.01). Negative binomial regression analysis revealed that the MDR phenotype (1.20; 95% CI 1.09 – 1.32), dialysis-dependence (1.53; 1.32 – 1.77), recent glucocorticoid use (1.23; 1.09 – 1.40), history of GNB infections (1.13; 1.03 – 1.24), recent surgery (1.15; 1.04 – 1.28), and hospital-acquisition of BSI (3.30; 2.99 – 3.64) were independently associated with longer hospital LOS. In transplant recipients there was a significant association between MDR BSI and BSI recurrence (MDR, 4/70 [5.7%]; Non-MDR, 1/135 [0.7%]; P = 0.05).

Conclusion: MDR GNB BSI are associated with immunosuppressed patients and longer hospitalizations. In transplant recipients, MDR GNB BSI were associated with higher BSI recurrence.

Joshua T. Thaden, MD, PhD1, Yanhong Li, MD2, Felicia Ruffin, RN, MSN3, Shelby Reed, PhD2 and Vance Fowler Jr, MD4, (1)Division of Infectious Diseases, Duke University Medical Center, Durham, NC, (2)Duke Clinical Research Institute, Durham, NC, (3)Division of Infectious Diseases and International Health, Department of Medicine, Duke University, Durham, NC, (4)Duke University Medical Center, Durham, NC


J. T. Thaden, Cubist: Investigator , Grant recipient

Y. Li, Cubist: Investigator , Grant recipient

F. Ruffin, None

S. Reed, Cubist: Investigator , Grant recipient

V. Fowler Jr, Cubist: Grant Investigator , Grant recipient

Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.