BACKGROUND: Extended-spectrum β-lactamase-producing Escherichia coli (ESBLEC) is an increasing cause of hospital-acquired infection. Risk factors for ESBLEC colonization and infection have been reported, but information is lacking about the risk factors for acquiring ESBLEC infection in patients with prior colonization. AIM: To identify risk factors for development of infection in patients colonized with ESBLEC. METHODS: A retrospective study was performed at Hôpital Necker-Enfants Malades, Paris from 2007 to 2010. A multi-variable model was created to compare a group of patients with nosocomial ESBLEC infection following documented ESBLEC colonization with a control group of patients colonized with ESBLEC (case-control design). FINDINGS: In total, 118 patients were included: 40 (26 adults, 14 children) with colonization and infection and 78 (51 adults, 27 children) with colonization alone. The median time from colonization to infection was 12.5 days [25-75% confidence interval (CI) 5-40]. ESBLEC infections included urinary tract infection (85%), bacteraemia (7.5%) and lower respiratory tract infection (7.5%). On multi-variate analysis, use of β-lactam/β-lactamase inhibitor prior to infection [odds ratio (OR) 3.2, 95% CI 1.073-9.864); P = 0.037] and urinary catheterization were reported as risk factors for ESBLEC infection in colonized patients (OR 5.2, 95% CI 1.984-13.569; P = 0.0008). CONCLUSION: Identification of these risk factors will be helpful to identify patients colonized with ESBLEC who will require antibiotics for ESBLEC in the case of nosocomial infection. Limiting the use of specific antibiotics and controlling the duration of urinary catheterization will be helpful for prevention of ESBLEC infection.
BACKGROUND: Extended-spectrum β-lactamase-producing Escherichia coli (ESBLEC) is an increasing cause of hospital-acquired infection. Risk factors for ESBLEC colonization and infection have been reported, but information is lacking about the risk factors for acquiring ESBLEC infection in patients with prior colonization. AIM: To identify risk factors for development of infection in patients colonized with ESBLEC. METHODS: A retrospective study was performed at Hôpital Necker-Enfants Malades, Paris from 2007 to 2010. A multi-variable model was created to compare a group of patients with nosocomial ESBLEC infection following documented ESBLEC colonization with a control group of patients colonized with ESBLEC (case-control design). FINDINGS: In total, 118 patients were included: 40 (26 adults, 14 children) with colonization and infection and 78 (51 adults, 27 children) with colonization alone. The median time from colonization to infection was 12.5 days [25-75% confidence interval (CI) 5-40]. ESBLEC infections included urinary tract infection (85%), bacteraemia (7.5%) and lower respiratory tract infection (7.5%). On multi-variate analysis, use of β-lactam/β-lactamase inhibitor prior to infection [odds ratio (OR) 3.2, 95% CI 1.073-9.864); P = 0.037] and urinary catheterization were reported as risk factors for ESBLEC infection in colonized patients (OR 5.2, 95% CI 1.984-13.569; P = 0.0008). CONCLUSION: Identification of these risk factors will be helpful to identify patients colonized with ESBLEC who will require antibiotics for ESBLEC in the case of nosocomial infection. Limiting the use of specific antibiotics and controlling the duration of urinary catheterization will be helpful for prevention of ESBLEC infection.