Integrated genomic and interfacility patient-transfer data reveal the transmission pathways of multidrug-resistant Klebsiella pneumoniae in a regional outbreak

See allHide authors and affiliations

Science Translational Medicine  22 Nov 2017:
Vol. 9, Issue 417, eaan0093
DOI: 10.1126/scitranslmed.aan0093

Drug-resistant out-of-towners

With the rise of multidrug-resistant organisms, much attention has focused on preventing nosocomial infection within individual health care facilities. Transfer of patients between facilities, however, may also contribute to the proliferation of hard-to-treat bacterial pathogens. Snitkin et al. investigated the role of patient transfer between hospitals in a year-long regional outbreak of carbapenem-resistant Klebsiella pneumoniae in the United States. They traced the spread of the infection by combining genomic sequencing with information about the movement of patients across the local health care network. This analysis was able to identify where the outbreak began and pinpointed which facilities contributed to the transmission of the infection in the region. Such an approach may enable directed interventions to prevent the transfer of drug-resistant organisms among hospitals.


Development of effective strategies to limit the proliferation of multidrug-resistant organisms requires a thorough understanding of how such organisms spread among health care facilities. We sought to uncover the chains of transmission underlying a 2008 U.S. regional outbreak of carbapenem-resistant Klebsiella pneumoniae by performing an integrated analysis of genomic and interfacility patient-transfer data. Genomic analysis yielded a high-resolution transmission network that assigned directionality to regional transmission events and discriminated between intra- and interfacility transmission when epidemiologic data were ambiguous or misleading. Examining the genomic transmission network in the context of interfacility patient transfers (patient-sharing networks) supported the role of patient transfers in driving the outbreak, with genomic analysis revealing that a small subset of patient-transfer events was sufficient to explain regional spread. Further integration of the genomic and patient-sharing networks identified one nursing home as an important bridge facility early in the outbreak—a role that was not apparent from analysis of genomic or patient-transfer data alone. Last, we found that when simulating a real-time regional outbreak, our methodology was able to accurately infer the facility at which patients acquired their infections. This approach has the potential to identify facilities with high rates of intra- or interfacility transmission, data that will be useful for triggering targeted interventions to prevent further spread of multidrug-resistant organisms.

View Full Text

Stay Connected to Science Translational Medicine