Building a continuous multicenter infection surveillance system in the intensive care unit : findings from the initial data set of 9,493 patients from 71 Italian intensive care units

Objective. To describe the epidemiology of infections in intensive care units (ICUs), whether present at admission or acquired during the stay. Methods. Prospective data collection lasting 6 months in 71 Italian adult ICUs. Patients were screened for infections and risk factors at ICU admission and daily during their stay. Main Results. Out of 9,493 consecutive patients admitted to the 71 ICUs, 11.6% had a community-acquired infection, 7.4% a hospital-acquired infection, and 11.4% an ICU-acquired infection. The risk curve of acquiring infection in the ICU was higher in patients who entered without infection than in those already infected (log-rank test, p < .0001; at 15 days, 44.0% vs. 34.6%). Hospital mortality (27.8% overall) was higher in patients admitted with infection than in those who acquired infection in the ICU (45.0% vs. 32.4%, p < .0001). Although the presence of infection per se did not influence mortality, the conditions of severe sepsis and septic shock were strong prognostic factors (odds ratio, 2.3 and 4.8, respectively). Apart from ICU-acquired peritonitis, no other site of infection reached statistical significance as an independent prognostic factor for hospital mortality. Conclusions. Adding specific data on infections and risk factors to a well-established electronic data collection system is a reliable basis for a continuous multicenter infection surveillance program in the ICU. Given the well-established importance of infection prevention programs, our data suggest that the improvement of the treatment of severe sepsis and septic shock is the key to lower infection-related mortality in the ICU. This calls for closer attention to severe infections in surveillance programs.