Epidemiology of infections caused by extended-spectrum β-lactamase-producing Escherichia coli and Klebsiella spp. a nested case-control study from a tertiary hospital in London

Information on risk factors for acquisition of extended-spectrum ß-lactamase (ESBL)-producing organisms and their outcomes in patients with invasive infections is scant. The objectives of this study were to evaluate risk factors and all-cause mortality associated with infection due to ESBL-producing organisms using a nested case-control design, and to document transmission within a hospital employing molecular and conventional epidemiological methods. From December 2003 to April 2005, 50 patients with bloodstream infections (BSIs) due to ESBL-producing E. coli and Klebsiella spp. were recruited. Controls (N = 50) were chosen, within the same period, from patients with non-ESBL-producing BSIs by simple random sampling; account was taken of potential confounding factors. Cases and controls were followed-up until November 2005, and outcomes were recorded as discharged or deceased. Molecular methods, supported by conventional epidemiology, were used to study the transmission of organisms. Logistic regression analyses showed prior ß-lactam antibiotics [odds ratio (OR) 11.57; 95% confidence intervals (CI) 2.31-51.15; P = 0.003], hospital stay >15 days (OR 2.63; 95% CI 1.01-6.89; P = 0.04) and prior admission to the intensive care unit (OR 13.98; 95% CI 1.88-19.15; P = 0.006) to be independent risk factors for the acquisition of ESBL-producing organisms. In the first 15 days of follow-up, a significant proportion of patients with ESBL-producing organisms died; however, there was no difference in mortality between cases and controls at the end of the follow-up period. Molecular epidemiology identified five clusters amongst the ESBL-producing isolates. Conventional epidemiological analyses supported the evidence of transmission in three of these clusters.