629 Molecular Anatomy of a Plasmid-Mediated Outbreak of KPC-producing CRE

Sunday, April 3, 2011: 2:45 PM
Coronado A (Hilton Anatole)
Costi D. Sifri, MD , University of Virginia Health System, Charlottesville, VA
Amy J. Mathers, MD , University of Virginia Health System, Charlottesville, VA
Heather L. Cox, PharmD , University of Virginia Health System, Charlottesville, VA
Hugo Bonatti, MD , University of Virginia Health System, Charlottesville, VA
Ann Karen C. Brassinga, PhD , University of Virginia Health System, Charlottesville, VA
Joanne L. Carroll, BS MT , University of Virginia Health System, Charlottesville, VA
W. Michael Scheld, MD , University of Virginia Health System, Charlottesville, VA
Kevin C. Hazen, PhD , University of Virginia Health System, Charlottesville, VA
Background: The spread of carbapenem-resistant Enterobacteriaceae (CRE) in the United States and globally has largely been attributed to a dominant strain of Klebsiella pneumoniae carbapenemase (KPC)-producing K. pneumoniae. Institutional outbreaks of CRE due to spread of a clonal blaKPC-carrying strain have also been reported for other enteric pathogens. While these reports highlight the clonal nature of CRE dissemination, several cases of horizontal transmission of blaKPC through plasmids have also been described. We report the unique experience of a heterogeneous outbreak of CRE at a university hospital due primarily to horizontal transfer of a blaKPC-carrying plasmid. 

Objective: To describe the molecular epidemiological characteristics of a heterogeneous outbreak of KPC-producing CRE. 

Methods: Following identification of the first CRE isolate in our institution, all clinical enteric Gram-negative isolates identified as possible ESBL-producers by automated testing were evaluated for carbapenem resistance (CR) by in vitro phenotypic susceptibility testing. CR isolates were examined for presence of blaKPC and Tn4401 by PCR. Pulsed-field gel electrophoresis (PFGE) evaluated relatedness of CRE isolates. Nested arbitrary PCR, a technique previously used to rapidly identify chromosomal transposon insertion sites, was adapted to assess plasmid relatedness. Plasmid extraction, Southern hybridization, and generation of transformants confirmed KPC-plasmid profiles. Clinical data were gathered from medical records.

Results: Sixteen isolates from 14 hospitalized patients were identified from 8/07-4/08, representing 6 species: Enterobacter cloacae (6), K. pneumoniae (4), Klebsiella oxytoca (3), Escherichia coli (1), Enterobacter asburiae (1), and Citrobacter freundii (1). PCR confirmed blaKPC and other coding regions of Tn4401 in all isolates, and sequencing demonstrated only the blaKPC-2 allele. PFGE identified multiple strains within species. Arbitrary PCR revealed 3 plasmid types; a unique E. coli plasmid (pUVA03), a plasmid shared within a K. oxytoca strain (pUVA02), and a plasmid present in all other isolates (pUVA01). Spread of blaKPC correlated well with epidemiologic risk factors including the patient location, healthcare team, and time period of potential exposure.

Conclusions: This outbreak is unusual in the diversity of species and strains carrying blaKPC-2, all epidemiologically linked to the index patient. Transmission of CR occurred primarily through horizontal interstrain or interspecies transfer of the Tn4401-bearing plasmid pUVA01 and, to a lesser extent, transmission of CRE strains between patients. Understanding the dynamics of blaKPC horizontal gene transfer will be crucial to limiting spread of CRE as epidemiologic and containment effort must focus not only on the spread of multidrug resistant strains but also on the associated resistance plasmids and transposons.