Bloodstream infections as a marker of community-acquired sepsis severity. Results from the Portuguese community-acquired sepsis study (SACiUCI study).

The impact of bloodstream infection (BSI) on admission to hospital on the outcome of patients with community-acquired sepsis (CAS) admitted to intensive-care units (ICU) is largely unknown. We selected 803 adult patients consecutively admitted with CAS to one of 17 Portuguese ICU, in whom blood cultures were collected before initiation of antibiotic therapy during a 12-month period. A BSI was identified on hospital admission in 160 (19.9%) patients. Those with and without BSI had similar mean Simplified Acute Physiology Score (SAPS) II and age. The presence of BSI was independently associated with mortality in ICU (adjusted odds ratio 1.86; 95% confidence interval 1.20-2.89; p 0.005). On the 4th day in ICU, patients with BSI were found to be significantly more dependent on vasopressor support (p 0.002) but not on ventilatory support. Cumulative ICU mortality was significantly higher in BSI patients from the 9th day onwards. A seasonal variation of BSI isolates was noted: gram-negative BSI were more common in the summer, whereas in the winter, gram-positive infections were more frequent (p 0.024), without mortality differences.