Targeted bacteremia surveillance throughout a year--comparison of community-acquired and hospital-acquired infection

To keep an eye on severe nosocomial infection and to evaluate the clinical difference of blood-stream infection between community-acquired and hospital-acquired infection, a survey of blood culture was performed in National Tokyo Medical Center from the period between November 2000 and October 2001. There were 252 episodes detected in 219 patients (80 community-acquired episodes in 80 patients and 172 hospital-acquired episodes in 139 patients). The three most common foci of infection/pathogens were as follows: in the community-acquired cases; urinary tract, pneumonia, infective endocarditis/Escherichia coli, viridant group of streptococci, Streptococcus pneumoniae, and in the hospital-acquired cases; intra-venous catheter, urinary tract, neutropenia-related bacteremia/Staphylococcus aureus, coagulase negative Staphylococcus, Enterococcus. Fifteen patients with community-acquired bacteremia and 37 patients with hospital-acquired bacteremia had been died within a month of the episode; the mortality was not significantly different between the both. The average of peak serum concentrations of C-reactive protein during the episodes of community-acquired bacteremia was higher than that of hospital-acquired bacteremia. These findings probably show that life threatening bloodstream infections seemed to be more common in the community. The rate of nosocomial bacteremia was approximately 1%, and no outbreak was observed during the period. Targeted bacteremia surveillance is maybe useful and efficient method to detect severe hospital-acquired infections.