Site of infection and mortality in patients with severe sepsis or septic shock. A cohort study of patients admitted to a Danish general intensive care unit

Background: The search for the site of infection has high priority in patients with severe sepsis and septic shock. However, it is questionable whether mortality is associated with the specific site of infection in patients admitted to an intensive care unit (ICU). Therefore, the 30-day and 90-day mortalities in ICU patients admitted with suspected or confirmed community-acquired infection were studied.Methods: A retrospective cohort study was conducted, including all adult patients admitted to a multi-specialty tertiary ICU with severe sepsis or septic shock from November 2008 to October 2010. The site of infection was classified according to criteria set for healthcare associated infections and infections in the acute care setting by Centers for Disease Control and Prevention (CDC). Kaplan-Meier curves and Poisson regression analysis were used to evaluate the association between site of infection and 30- and 90-day all-cause mortality, adjusting for age, sex and comorbidities.Results: Three hundred and eighty-eight patients were included. One or more comorbidities were present in 76% of patients. Across all sites of infection, there were more patients with septic shock than patients with severe sepsis. The most frequent site of infection was pneumonia, followed by gastrointestinal infection. Urinary tract infection was found to be an independent predictor of mortality among septic ICU patients when adjusting for sex, age and comorbidities.Conclusions: The results suggest that identification of correct site of infection is important in the management of severe sepsis and septic shock.