Outbreak of KPC-3 producing carbapenem- resistant Klebsiella pneumoniae in a US pediatric hospital

Abstract

Background. The increase in carbapenem-resistant Enterobacteriaceae (CRE) infections is a critical public health issue. We recently experienced the largest single-center pediatric outbreak of carbapenem-resistant Klebsiella pneumoniae (CRKP) at our hospital. The objective of this study was to describe the molecular epidemiology of this outbreak before and after infection-prevention interventions. Methods. All positive cultures and associated clinical conditions were reviewed to determine whether health care-associated infections (HAIs) exist. HAIs were defined using Centers for Disease Control and Prevention guidelines. CRKP isolates were collected and screened for the presence of β-lactamase genes. Strain relatedness of CRKP isolates was determined by field-inversion gel electrophoresis (FIGE) and multilocus sequence typing (MLST). Polymerase chain reaction (PCR) amplification and sequencing of blaTEM, blaSHV, and blaKPC genes were performed on representative isolates. Results. During March-July 2010, 18 CRKP isolates were recovered from 15 unique patients. Six isolates were considered HAIs; all were central-line-associated bloodstream infections. All isolates testing positive by PCR for blaKPC were found to carry KPC-3 in transposon Tn4401 isotype "b." FIGE revealed 2 prevalent patterns (accounting for 10 and 3 CRKP isolates, respectively) thatMLST demonstrated to consist entirely of strains from ST730; the remaining FIGE types corresponded to ST14, ST15, and ST1559 (a single-locus variant of ST730), with these alternate backgrounds appearing later in the outbreak. New CRKP cases decreased after the implementation of infection-control interventions. All isolates were ciprofloxacin sensitive. Conclusions. Molecular analyses document the introduction of a KPC-3-producing CRKP clone into our hospital setting, though some isolates appear to have other mechanisms of carbapenem resistance. The transition to a polyclonal epidemiology suggests that the initial outbreak was due to nosocomial spread of a single ST730 clone, while latter isolates may have been secondary to the introduction of a blaKPC-3/Tn4401 isotype "b"-containing plasmid into other K pneumoniae strain backgrounds versus new carbapenemase-producing bacteria.