Microbiological and epidemiological analysis, and clinical outcome of patients with bloodstream infection from an Esquel hospital in the period 2007-200

In order to evaluate the bacteremic episodes, the blood cultures performed in 867 patients during a 29-month period were reviewed. The incidence of significant bloodstream infections was 10/1000 admissions/year. About 47% of bacteremias were community-acquired (CAB), 22% nosocomial (NB), and 31% health care-associated (HCAB). The five most common pathogens were: Streptococcus pneumoniae, Staphylococcus aureus, Escherichia coli, Enterococcus spp. and coagulase-negative staphylococci, whereas the main sources were: respiratory, intraabdominal and unknown. The major organism in CAB was S. pneumoniae, and S. aureus in NB and HCAB. The ineffective empirical-antimicrobial-therapy [IEAT (total 26.5%)] was much higher in NB and HCAB vs. CAB (p < 0.05). There was no significant difference in mortality and IEAT between known vs. unknown sources. Old age (> 70 yrs), previous antibiotic therapy, hospitalization in clinical medical services, Enterococcus spp. and methicillin-resistant Staphylococcus isolates were more associated with IEAT (p < 0.05). Previous antibiotic therapy and longer hospital stays were more common in NB vs. CAB and HCAB (p < 0.05). Overall and septicemia-associated mortality was 28.9% and 21.7%, respectively. Univariate associations with increased mortality (p < 0.05) included: hypothermia, WBC < 1499/mm3, hypotension, presentation with septic-shock, hospital-stay 10 days, and polymicrobial bacteremia. The early administration of effective empirical antimicrobial therapy according to our epidemiological characteristics and an aggressive hemodynamic treatment in presence of septic shock could reduce bacteremia-associated morbidity and mortality, costs and length of hospital stays.