RESUMO
An infection control programme was implemented in a 21,000-bed multihospital institution for controlling the spread of carbapenemase-producing Enterobacteriaceae (CPE) and glycopeptide-resistant Enterococcus faecium (GRE), classified as 'emergent extensively drug-resistant bacteria' (eXDR) in France. We evaluated factors associated with outbreaks occurrence (n = 103), which followed 901 eXDR introductions (index case followed or not by secondary cases) from 2010 to 2015. In univariate analysis, knowing that patients had been hospitalised abroad, bacterial species (GRE vs CPE, as well as the CPE Klebsiella pneumoniae compared with the other Enterobacteriaceae species) and type of measures implemented within the first 2 days of hospitalisation were associated with outbreaks occurrence, but not the type of wards where carriers were hospitalised, nor the eXDR colonisation or infection status. In multivariate analysis, occurrence of outbreaks was significantly lower when contact precautions (odds ratio (OR): 0.34; 95% confidence interval (CI): 0.22-0.54) and even more when dedicated nursing staff (OR: 0.09; 95% CI: 0.02-0.39) were implemented around eXDR index cases within the first 2 days of hospitalisation (p < 10 - 3). GRE introductions were more frequently associated with occurrence of outbreaks than CPE (OR: 3.58; 95% CI: 2.32-5.51, p < 10 - 3). A sustained and coordinated strategy is efficient to limit the spread of eXDR at the scale of a large health institution.