187E Clinical Epidemiology of Carbapenem-Resistant Enterobacteriaceae in Community Hospitals: A Case-Case-Control Study

Wednesday, October 24, 2012
Westin Diplomat Resort
Grace C. Lee, Pharm.D, BCPS1, Donna R. Burgess, RPH2, Kurt R. Winkler, Pharm.D., MHA, BCPS3 and David S. Burgess, Pharm.D., FCCP1
1University of Texas at Austin College of Pharmacy and University of Texas Health Science Center School of Medicine, San Antonio, TX
2Methodist Hospital Department of Pharmacy and University of Texas at Austin College of Pharmacy, San Antonio, TX
3Methodist Hospital Department of Pharmacy, San Antonio, TX

Purpose: Despite the increasing rate of carbapenem-resistant Enterobacteriaceae (CRE), there are limited data identifying risk factors.  This study evaluated risk factors associated with the acquisition of CRE among hospitalized patients.

Methods: We performed a retrospective matched case-case-control study in four community hospitals from June 2007 through November 2011.  Case Group (CG) 1 comprised of patients with CRE. CG 2 comprised of patients with carbapenem-susceptible Enterobacteriaceae (CSE). CG 2 patients were matched to CG 1 patients by site of infection and specific species of Enterobacteriaceae. Hospitalized controls were matched 2:1 by date of admission and hospital location to patients in CG 1.  Two sets of analyses were conducted comparing demographics, comorbidities and antibiotic exposures of CG 1 and CG 2 to controls then contrasted to identify unique risk factors associated with CRE.

Results: Overall, 104 patients (CG 1 – 25 patients; CG 2 – 29 patients, Control Group – 50 patients) were evaluated.  CRE and CSE comprised mostly of Klebsiella species (64%) from a urinary source (28%). In univariate analysis, renal failure (p<0.001), exposure to fluroquinolones (p<0.01), carbapenems (p<0.001), aminoglycosides (p<0.01), poor functional status (p<0.01), ICU stay (p < 0.01), cumulative number of antibiotics exposures (p <0.001), and cumulative number of antimicrobial days by “time at risk” ratio (p<0.001) were significantly higher in CG 1 than controls. In multivariable analysis, poor functional status (OR 4.43, CI 2.25-8.73; p<.01), ICU stay (8.9, CI 3.49-23.03 p < 0.01), and cumulative number of antibiotics exposures (p <0.01) were distinct independent predictors of CRE isolation whereas cumulative healthcare exposures (p<0.01) and vancomycin exposure (OR 2.5, CI 1.63-3.84; p<0.01) were predictors for CSE.    

Conclusion: CRE should be considered in patients with poor functional status requiring ICU admission, particularly those who have received multiple antibiotics.