Increased incidence of central venous catheter–associated bloodstream infections: Possible link to transparent dressing configuration

BACKGROUND/OBJECTIVES: As part of routine surveillance of nosocomial bacteremia, we noted an increase in the hospital-wide rate of catheter-related bloodstream infection (CRBSI) over our baseline of 0.66 per 1000 patient days (ptpd, July 2002–June 2004). We launched an investigation to determine the cause for this increase. Transparent polyurethane dressings are a popular dressing that does not significantly increase the risk of CRBSI. These dressings have been modified to incorporate a woven tape to help adherence to skin. These dressing designs may change infection rates. One of these new dressings was introduced in our dressing kits before this outbreak. METHODS: Examination of case details was used to look for common factors and exposures. After a multidisciplinary review, we implemented stepwise changes in the dressing used for central line care: we changed to a dressing that put the tape on top of, rather than below, the transparent polyurethane. RESULTS: In surveillance of our 280-bed community teaching hospital, we expect no more than six CRBSI a month; we exceeded this threshold in August 2004. The rate peaked with 14 cases in 5718 patient days in September 2004, a rate of 2.45 ptpd (95% confidence interval 1.3-4.1). Cases occurred with several different types of catheters and on different clinical units. Organisms varied, with an increase in the relative contribution of Candida infections. Procedures for insertion and care of central lines were consistent with CDC guidelines. We found no common risk factors for infection, such as total parenteral nutrition, in a majority of cases. Examination of the epidemiology curve showed an increase that corresponded to the introduction of a new central line dressing. Changing to more frequent dressing changes did not stop the outbreak. Replacing the dressing with a new type of dressing resulted in the rates falling toward baseline (1.3 ptpd, CI 0.51-2.6, in December 2004), even with a return to 7-day dressing change intervals. Compliance was incomplete; surveillance continues following the removal of old dressing. CONCLUSION: Shortly after implementing a tape-reinforced transparent dressing, we detected a fourfold increase in the incidence of CRBSI. By changing dressings to a transparent dressing with a different design, we saw a return in our rate of infection to baseline. This suggests that the presence of woven tape under a transparent dressing may increase the rate of CRBSi