Experience with a hospital wide outbreak of vancomycin-resistant enterococci

Background: Vancomycin-resistant enterococci (VRE) were first detected in our institution in 1991. An outbreak was recognized in late 1992 when there was a sudden rise in the number of patients per month with VRE. Little information exists concerning the natural history of infection with these pathogens, and the effect of antimicrobial therapy is unclear. Recent guidelines emphasize prudent use of vancomycin and prompt institution of barrier precautions to limit the spread of vancomycin resistance. Methods: Data were obtained by review of microbiologic and clinical records, Patients were categorized according to site of infection, and outcome of therapy was assessed. Hospital antibiotic usage was analyzed to determine any correlation with the outbreak Infection control measures instituted in 1993 included patient isolation, environmental cleaning, and a reemphasis of barrier precautions. Surveillance cultures were performed to assess the extent of the outbreak in January 1995. Results: VRE were detected in clinical cultures from 159 patients from 1991 through 1994. Mortality rate was 48%, but in most cases death could not be attributed to enterococcal infection. Patients with wound infections healed without specific therapy. Many patients with bacteremia had resolution with ampicillin or without specific therapy. Patients were widely scattered throughout the hospital from the beginning of the outbreak. Hospital usage of cefotaxime correlated with the number of cases. Infection control measures were not successful. Surveillance culture results in January 1995 revealed that 53% of all medical and surgical inpatients had fecal colonization with VRE. Genetic analysis of selected isolates revealed that one strain predominated, but at least seven distinct strains were identified. Conclusions: Our data suggest that many infections with VRE resolve without specific therapy. The infection control measures we used were ineffective, possibly because of the multiple strains present in our hospital. Isolation of all patients with VRE is impractical when there is widespread fecal carriage.