Multiresistent Klebsiella pneumoniae hos nyfødte.

The increasing level of antimicrobial resistance among clinical relevant bacteria, especially the increasing prevalence of extended-spectrum ß-lactamase (ESBL) producing Enterobacteriaceae, is of great concern in most parts of the world including Norway. During 2008-09 an outbreak caused by ESBL-producing (type CTX-M-15) Klebsiella pneumoniae occurred in the neonatal intensive care unit (NICU) at Stavanger University Hospital. This was the first reported NICU outbreak caused by ESBL-producing Enterobacteriaceae in Scandinavia. The objectives of this study were: (i) to describe the characteristics of the NICU outbreak and the K. pneumoniae outbreak strain(s), (ii) to investigate the duration of intestinal ESBL-carriage in the children affected by the outbreak, risk factors for prolonged carriage and intra-household transmission during colonization, and (iii) to characterize the ESBL-encoding plasmid of the outbreak strain, and examine the plasmid’s dynamics in relation to its K. pneumoniae host strain. In total 58 children were affected by the NICU outbreak, caused by multidrug resistant CTX-M-15-producing K. pneumoniae (CTX-M-15-Kp). Of these, 56 were colonized by the ST17 outbreak strain (two children were colonized by an unrelated strain). No commonly encountered virulence factors were detected in the outbreak strain. One child suffered a blood stream infection caused by CTX-M-15-Kp, but no children died of reasons related to the outbreak. A probable index case was identified as one child and its mother, from which several CTX-M-15-Kp positive breast milk samples were obtained. The outbreak was contained by enforced infection control measures and strict cohorting of patients and health care workers. In a follow-up study of colonized children and their families, CTX-M-15-Kp carriage for up to two years was confirmed in some of the children. Median carriage time was 12.5 months. Risk factors for prolonged carriage were delivery by caesarean section and treatment with antibiotics (ampicillin and gentamicin) during the NICU stay. Intra-household transmission of CTX-M-15-Kp was documented in 33% of the households. CTX-M-15 was encoded by a ~180 kb IncFIIK pKPN3-like plasmid (pKp848CTX) in the ST17 outbreak strain. pKp848CTX encoded multiple antibiotic, heavy metal and thermoresistance determinants. The plasmid could not be transferred from its K. pneumoniae host to Escherichia coli in vitro, consistent with in vivo observations. pKp848CTX was stably maintained by its host strain during intestinal colonization for up to two years. Segregational plasmid loss ranging from 0% to 17.5% in evolved populations was detected in vitro, but only negligible fitness costs associated with plasmid carriage were uncovered. In conclusion, the NICU outbreak was caused by a CTX-M-15-producing ST17 K. pneumoniae strain, which spread efficiently among patients. The virulence properties of the outbreak strain seem to be low. Vertical mother-to-neonate transmission due to contaminated breast milk was the most likely the source of the outbreak. Children may become long-term faecal ESBL-carriers after colonization during a NICU outbreak, and represent a reservoir for intra-household ESBL-spread after hospital discharge. BlaCTX-M-15 was encoded by a pKPN3-like IncFIIK plasmid, pKp848CTX, which seems to be well adapted to CTX-M-15-Kp. The plasmid’s inability of self-transfer when hosted by this strain may have limited the extent of the outbreak and the duration of ESBL-colonization. It is important to investigate, characterize and report outbreaks. Such information may be useful in the prevention and containment of future outbreaks. In our hospital, procedures have been changed according to our findings. Infection control measures, such as enhanced hand hygiene and more efficient hygienic barriers between patients, were implemented in the NICU during the outbreak, and have been maintained since. The empirical treatment for systemic infections in neonates was changed from the combination of ampicillin and gentamicin to penicillin and gentamicin, to prevent future selection of K. pneumoniae in the NICU environment. Furthermore, Enterobacteriaceae isolated from breast milk samples are now always screened for ESBL-production to prevent ESBL-transmission from mothers to infants via contaminated breast milk. Regular ESBL-screening of children hospitalized in the NICU or at admission has been discussed, but has until present not been implemented. We documented that children colonized during a NICU outbreak are at risk for prolonged ESBL-carriage, which has implications for infection control measures during NICU stays and at re-admissions. Due to our results, the use of alert codes to identify previously colonized patients and screening at re-admission might be considered for 1-2 years after ESBL-colonization in the perinatal period to prevent ESBL-spread within NICUs or other pediatric wards. Plasmid stability, plasmid-host adaptation and the magnitude of plasmid-conferred fitness costs are of importance for the persistence and spread of antibiotic resistance determinants. Biological costs of an antibiotic resistance plasmid and plasmid maintenance by the host strain during long-term colonization of a human host has until present not been investigated. By combining in vivo and in vitro studies, we gained new insight into the dynamics between a plasmid encoding antibiotic resistance determinants and its host strain. Our results suggest that multidrug resistance plasmids may be well adapted to successful host strains, and that antibiotic resistance determinants may be maintained both on the plasmid and the bacterial strain level, without antibiotic selection, for long periods of time, also in vivo. The combat against AMR is a multifaceted global task. To limit further spread of ESBL-producing and other MDR bacteria, resistance surveillance programs, adequate empirical treatment recommendations, antibiotic stewardship, increased focus on infection control and rapid detection of antibiotic resistance are of great importance. Increased understanding of the dynamics of spread and maintenance of resistance determinants in hospitals, in the environment, in animals, in food and in the human community is needed. The interactions between plasmids encoding multidrug resistance and their host strains need to be further studied to better understand the mechanisms for the spread and maintenance of MDR bacteria also as a potential target for interventions.