[Epidemiology of infection and sepsis in intensive care unit patients].
Since 1992, epidemiological and clinical studies have classified severe infections into three categories: sepsis, severe sepsis and septic shock. Microbiological documentation is not always provided. We used a different approach, focusing on the infection itself, whether or not it is microbiologically documented or associated with sepsis. In an international prospective cohort study, all patients admitted to the participating units from May 1997 to May 1998 were followed until hospital discharge. Twenty-eight intensive care units (ICU) in eight countries enrolled 14,364 patients. Of these, 6011 stayed in the ICU for less than 24 hours and 8353 for more than 24 hours. Overall, 3034 infectious episodes were recorded at ICU admission (crude incidence rate 21.1%). Among patients hospitalized for more than 24 hours, 1581 infectious episodes occurred in the ICU (crude incidence rate 18.9%), including 713 cases (45%) in patients who were already infected at ICU admission. These rates varied among the ICUs. Respiratory, gastrointestinal, urinary tract and primary bloodstream infections represented about 80% of all infections. Hospital-acquired and Intensive Care Unit-acquired infections were more frequently microbiologically documented than community-acquired infections (71% and 86%, respectively, vs 55%). About 28% of all infections were associated with sepsis, 24% with severe sepsis and 30% with septic shock (18% were not classified). Crude in-hospital mortality rates ranged from 16.9% in uninfected patients to 53.6% in patients who were both infected at the time of ICU admission and subsequently acquired an infection during the ICU stay. The in-hospital mortality rate increased with severity, from 20% for sepsis to 40% for severe sepsis and 60% for septic shock, but also depended on the origin of infection (community vs hospital/ICU). Crude incidence rates of ICU infection were high, varying among ICUs and patient subsets. Thus, vital outcome depends not only on the severity of sepsis but also on the characteristics of the infection.