Program Schedule

Predictors of Hospital Mortality Among Septic ICU Patients With Acinetobacter spp. Bacteremia: A Cohort Study

Session: Poster Abstract Session: Bacteremia: Gram-Negative Bacteremia
Friday, October 10, 2014
Room: The Pennsylvania Convention Center: IDExpo Hall BC

Inappropriate empiric antibiotic therapy (IEA) increases mortality in severe infections. Carbapenem resistance among Acinetobacter spp. challenges a clinician’s ability to choose appropriate coverage. We hypothesized that among septic ICU patients with Acinetobacter spp. bacteremia (Ac-BSI), patients with carbapenem-resistant Acinetobacter spp. (CRAc) are at high risk for IEA, and that IEA is a predictor of hospital death. 


We conducted a single-center retrospective cohort study, 2002-2012, of adult septic ICU patients with Ac-BSI at Barnes-Jewish Hospital, a 1200-bed urban teaching hospital. IEA was defined as exposure to initially prescribed antibiotics not active against the pathogen based on in vitro susceptibility testing, and having no exposure to appropriate antimicrobial treatment within 24 hours of drawing positive culture. We compared patients who died to those who survived, and derived logistic regression models to identify predictors of hospital mortality and of IEA. Proportions and odds ratios (OR) with 95% confidence intervals (CI) are presented for categorical variables, and median values with interquartile ranges (IQR) for continuous variables. 

Results: Out of 131 patients with Ac-BSI, 65 (49.6%) died (non-survivors, NS) before discharge. NS were older than survivors (S) (63 [51, 76] vs. 56 [45, 66] years, p=0.014), and had a higher APACHE II (24 [19, 31] vs. 18 [13, 22], p<0.001) and Charlson (5 [2, 8] vs. 3 [1, 6], p=0.009) scores. NS were also more likely than S to have hospital-acquired Ac-BSI (72.3% vs. 50.0%, p=0.009), and prior antibiotics (75.4% vs. 57.6%, p=0.031) as risk factors for a healthcare-associated infection, as well as to require pressors (75.4% vs. 42.4%, p<0.001) and mechanical ventilation (75.4% vs. 53.0%, p=0.008). Both CRAc (69.2% vs. 47.0%, p=0.010) and IEA (83.1% vs. 59.1%, p=0.002) were more frequent among NS than S. In logistic regressions, IEA emerged as an independent predictor of hospital death (OR 3.58, 95% CI 1.28-9.98), while CRAc was the single strongest predictor of IEA (OR 50.38, 95% CI 13.49-188.17).    


Among septic ICU patients with Ac-BSI, IEA predicts mortality. Carbapenem resistance appears to mediate the relationship between IEA and mortality.

Andrew F. Shorr, MD, MPH, Pulmonary and Critical Care Medicine, Washington Hospital Center, Washington, DC, Marya D. Zilberberg, MD, MPH, University of Massachusetts and Evimed Research Group, LLC, Goshen, MA, Scott Micek, PharmD, Pharmacy, Barnes-Jewish Hospital, St. Louis, MO and Marin Kollef, MD, Pulmonary and Critical Care Medicine, Washington University School of Medicine, St. Louis, MO


A. F. Shorr, Tetraphase: Consultant and Grant Investigator, Research support

M. D. Zilberberg, Tetraphase: Grant Investigator, Research support

S. Micek, Tetraphase: Grant Investigator, Research support

M. Kollef, Tetraphase: Grant Investigator, Research support

Findings in the abstracts are embargoed until 12:01 a.m. EDT, Oct. 8th with the exception of research findings presented at the IDWeek press conferences.

Sponsoring Societies:

© 2014, All Rights Reserved.

Follow IDWeek