Epidemiology of endemic Pseudomonas aeruginosa: why infection control efforts have failed.

The epidemiology of Pseudomonas aeruginosa was evaluated in an intensive care unit for a period of six months by means of serial surveillance and environmental cultures. One hundred (37%) of 270 patients were noted to be colonized: 63 at the time of their admission and 37 during their stay on the unit. Colonization at the time of admission was associated with length of hospitalization before admission to the intensive care unit, age, gastrointestinal disease, and prior use of antibiotics. The strains acquired on the intensive care unit represented several different serotypes, with little clustering; the source of most strains was not found. In only 12 cases did the acquisition of P. aeruginosa appear to represent cross-infection; the use of barrier isolation could have prevented at most five of these cases. Undetected endogenous gastrointestinal carriage may have been responsible for many other apparent acquisitions. Clinical infection in association with preceding gastrointestinal colonization developed in 20 patients. The data indicate that traditional control measures aimed at the prevention of exogenous acquisition of P. aeruginosa are unlikely to have an impact on the overall incidence of infection and that efforts to prevent infection in patients who are already colonized are necessary.