• critically ill;
  • age;
  • infections


To determine the effect of age on patterns of infection and on outcomes in individuals with infection.


Analysis of data from an international, observational, point-prevalence study: Extended Prevalence of Infection in Intensive Care (EPIC II).


Intensive care units (ICUs; N = 1,265) in 75 countries.


All adults present in a participating ICU on May 8, 2007. Individuals with an infection were included and classified according to age (18–44, 45–64, 65–74, 75–84, ≥85).


Demographic, physiological, bacteriological, and therapeutic data were collected on the day of the study. Outcome data were collected until hospital discharge or for 60 days.


Of the 13,796 adults enrolled in EPIC II, 7,087 (51.4%) had an infection. Of these, 330 (4.7%) were aged 85 and older, 1,405 (19.8%) were 75 to 84, 1,713 (24.2%) were 65 to 74, 2,358 (33.3%) were 45 to 64, and 1,281 (18.1%) were 18 to 44. Severity of illness did not differ between groups. Those aged 85 and older had fewer bloodstream infections than those younger than 75, fewer central nervous system infections than those who were younger than 65, and more abdominal infections than those who were younger than 45. A microbiological diagnosis was established less frequently in participants aged 85 and older than in younger participants. Gram-negative microorganisms were more frequently isolated in those aged 85 and older than in other groups. ICU and hospital mortality were significantly higher in participants aged 85 and older than in those who were younger than 65.


A large proportion of individuals in the ICU with infection are aged 65 and older. Patterns of infection, including site and type of microorganism, vary according to age. Being aged 85 and older was an independent risk factor for mortality in individuals in the ICU with infection.