Assessing risk factors for acquiring extended-spectrum lactamase-producing Gram-negative infections in long-term care facilities: a case-control study
Abstract number: P1118
Tacconelli E., Mantengoli E., Cadeddu C., Cunietti E., Luzzaro F., Cauda R., Rossolini G.M., Tinelli M.
Background: Extended-spectrum b-lactamase-producing Gram-negative (ESBL+Gn) are emerging pathogens among health-care associated infections. However, to our knowledge, risk factors for ESBL+Gn infections in long term care facilities (LTCFs) were rarely investigated.
Methods: A prospective case-case-control study. This approach was chosen because allows to compare the relative contribution of the ESBL production over and above simply having the Gn infection. To homogenize the comparisons, urinary tract infection was chosen as common indicator of infections in LTCFs. ESBL determinants were characterized by hybridization and confirmed by PCR and sequencing. In the first study, cases were defined as patients harbouring the ESBL+Gn while controls were those without Gn infections (random sampling). In the second study, cases were instead defined as those patients harbouring the ESBL-Gn while controls, similar to the first approach, were those patients without Gn infections.
Results: The study involved 279 patients in a LTCF. Cases (1st group) developed UTIs due to the following ESBL+Gn: E. coli (71%), K. pneumoniae (11%), and P. mirabilis (17%). CTX-M-type enzymes (CTX-M-1 and CTX-M-15) were prevalent (84%). TEM-type were produced by 10% P. mirabilis (TEM-92). All K. pneumoniae co-produced CTX-M group 1, SHV and TEM enzymes. Patients with ESBL+Gn UTIs were more likely to have been hospitalised in the previous year (p = 0.04), to suffer from cirrhosis (p = 0.04) and to have permanent urinary catheter (p = 0.02). Compared with controls, these patients had a longer duration of hospital stay before UTI developed (17 vs. 9 days, OR 1.1 per 1-day longer, p = 0.04) and were more likely to have had exposure to antibiotics (OR 9, p < 0.01). The risk was higher in patients with at least 7 days of antibiotic exposure (OR 11). In the casecase comparison using a multinomial logistic regression model, after adjusting for demographic and clinical risk factors, we found a statistically significant risk difference for previous use of quinolones (OR, 11.5, p = 0.01) and 3rd gen. cephalosporins (OR, 3.7, p = 0.02), in patients with ESBL+Gn infections. For ESBL-Gn infections, no specific antibiotic remained a significant risk after adjusted analysis.
Conclusions: Exposure to quinolones and 3rd gen cephalosporins was associated with subsequent UTIs due to ESBL+Gn in a LTCF. Interventions aimed to the reduction of antimicrobial usage in LTCFs should be further developed and implemented.
|Session name:||Abstracts of 21st ECCMID / 27th ICC|
|Location:||Milan, Italy, 7 - 10 May 2011|
|Back to top|