A prospective study of treatment of carbapenem-resistant Enterobacteriaceae infections and risk factors associated with outcome

BACKGROUND: To describe the clinical and microbiological data of carbapenem-resistant Enterobacteriaceae (CRE) infections, the treatment used, hospital- and infection-related mortality, and risk factors for death. METHODS: A prospective cohort conducted from March 2011 to December 2012. Clinical, demographic, and microbiological data such as in vitro sensitivity, clonality, carbapenemase gene mortality related to infection, and overall mortality were evaluated. Data were analyzed using Epi Info version 7.0 (CDC, Atlanta, GA, USA) and SPSS (Chicago, IL, USA). RESULTS: One hundred and twenty-seven patients were evaluated. Pneumonia, 52 (42 %), and urinary tract infections (UTI), 51 (40.2 %), were the most frequent sites of infection. The isolates were polyclonal; the Bla (KPC) gene was found in 75.6 % of isolates, and 27 % of isolates were resistant to colistin. Mortality related to infection was 34.6 %, and was higher among patients with pneumonia (61.4 %). Combination therapy was used in 98 (77.2 %), and monotherapy in 22.8 %; 96.5 % of them were UTI patients. Shock, age, and dialysis were independent risk factors for death. There was no difference in infection-related death comparing colistin-susceptible and colistin-resistant infections (p = 0.46); neither in survival rate comparing the use of combination therapy with two drugs or more than two drugs (p = 0.32). CONCLUSIONS: CRE infection mortality was higher among patients with pneumonia. Infections caused by colistin-resistant isolates did not increase mortality. The use of more than two drugs on combination therapy did not show a protective effect on outcome. The isolates were polyclonal, and the bla (KPC) gene was the only carbapenemase found. Shock, dialysis, and age over 60 years were independent risk factors for death.