Risk factors for acquisition of carbapenemase-producing versus non-carbapenemase-producing enterobacterales: a case-control study.

OBJECTIVES: We aimed to assess the association between carbapenem-resistant Enterobacterales (CRE) colonization pressure and carbapenem exposure and acquisition of carbapenemase-producing Enterobacterales (CPE) and non-carbapenemase-producing carbapenem-resistant Enterobacterales (non-CP-CRE). METHODS: We conducted a parallel 1:2 matched case-control study at Rambam Health Care Campus, Israel, from January 2014 to June 2017. The cases included all adults who acquired CPE or non-CP-CRE in hospital. The controls were hospitalized patients who were negative for CRE on screening and matched by age, hospitalization division and the number of hospitalization days 90 days prior to CRE screening. The exposures of interest were high CRE colonization pressure, defined as a higher-than-median proportion of CRE carriers in the concurrent patient's department before acquisition, and carbapenem exposure, assessed as days of treatment. Conditional logistic regression was used for analyses of CPE and non-CP-CRE. RESULTS: In total, 1058 patients were included: 278 CPE and 75 non-CP-CRE cases, matched to 556 and 149 controls, respectively. High CRE colonization pressure was associated with CPE acquisition (adjusted odds ratio [aOR], 2.6; 95% CI, 1.69-4.02); however, the duration of carbapenem treatment was not (aOR, 1.004; 95% CI, 0.98-1.03; 1-day increment). The duration of carbapenem treatment was significantly associated with non-CP-CRE acquisition (aOR per day, 1.07; 95% CI, 1.03-1.11). A source patient was identified significantly more frequently in epidemiological acquisition investigations of CPE than in those of non-CP-CRE (107/240, 44.6% vs. 18/64, 28.1%, respectively; p 0.017). CONCLUSIONS: CPE acquisition was associated with horizontal transmission, whereas non-CP-CRE was associated with carbapenem exposure. Differences in the drivers of acquisition mandate tailored infection prevention efforts.