Risk factors and outcomes of patients colonized with carbapenemase-producing and non-carbapenemase-producing carbapenem-resistant .

Objective: To compare risk factors and outcome of patients colonized with carbapenemase-producing (CP) carbapenem-resistant Enterobactereaceae (CRE) and non-CP-CRE.

Design: A comparative historical study.

Setting: A 1,000-bed tertiary-care university hospital.

Patients: Adults with CP-CRE positive rectal swab cultures, non-CP-CRE positive rectal swab cultures, and negative rectal swab cultures (non-CRE).

Methods: CP-CRE and non-CP-CRE colonized adult patients versus patients not colonized with CRE hospitalized during 24 months were included. We identified patients retrospectively through the microbiology laboratory, and we reviewed their files for demographics, underlying diseases, Charlson Index, treatment, and outcome.

Results: This study included 447 patients for whom a rectal swab for CRE was obtained: 147 positive for CP-CRE, 147 positive for non-CP-CRE, and 147 negative for both. Patients with CP-CRE and non-CP-CRE versus no CRE more frequently resided in nursing homes (P<0.001), received antibiotics 3 months prior to admission (P < .001), and received glucocorticosteroids 3 months prior to admission (P = .047 and P < .001, respectively). Risk factors unique for non-CP-CRE versus CP-CRE colonization included mechanical ventilation and patient movement between hospital departments. Non-CP-CRE was a predictor for mechanical ventilation 2.5 that of CP-CRE colonization. In-hospital mortality was highest among non-CP-CRE-colonized patients. On COX multivariate regression for mortality prediction age, Charlson index and steroid treatment 3 months before admission influenced mortality (P = .027, P = .023, and P = .013, respectively).

Conclusions: Overlapping and unique risk factors are associated with CP-CRE and non-CP-CRE colonization. Non-CP-CRE colonized patients had a higher in-hospital mortality rate.