Health-care-associated bloodstream infections at admission to the ICU.

BACKGROUND Infections occurring among outpatients having recent contact with the health-care system have been recently classified as health-care-associated infections to distinguish them from hospital- and community-acquired infections. Patients with bloodstream infections (BSIs) were studied to assess health-care-associated infections at admission in the ICU. METHODS This work was a multicenter, prospective, observational study of all adult patients with BSI at ICU admission at 27 Spanish hospitals and one Argentine hospital. Cases of BSI were classified as community-acquired BSI (CAB), health-care-associated BSI (HCAB), or hospital-acquired BSI (HAB), and their characteristics were compared. RESULTS Of 726 BSIs, 343 (47.2%) were CABs, 252 (34.7%) were HABs, and 131 (18.0%) were HCABs. Potentially antibiotic-resistant pathogens were more frequently isolated in HABs (34.8%) and HCABs (27.6%) than in CABs (10.3%) (P < .001). Logistic regression analysis revealed that HABs (OR, 4.6; 95% CI, 2.9-7.3), HCABs (OR, 3.1; 95% CI, 1.8-5.4), and BSIs of unknown origin (OR, 1.7; 95% CI, 1.0-2.8) were independently associated with the isolation of potentially antibiotic-resistant pathogens. The incidence of inappropriate treatment was significantly higher in HABs (OR, 3.4; 95% CI, 2.1-5.3) and in HCABs (OR, 1.8; 95% CI, 1.0-3.2) than in CABs. CONCLUSIONS One in five BSIs diagnosed at ICU admission is health-care-associated. The incidence of potentially drug-resistant pathogens in HCABs is more similar to that of HABs, and they should be treated as such until culture data are available.