OBJECTIVES: Investigate and control an increase in bloodstream infections (BSIs) in an outpatient hemodialysis center. PATIENTS AND DESIGN: A retrospective cohort study was conducted for patients receiving dialysis at the center from February 2000 to April 2001. A case-control study compared microbiological data for all BSIs that occurred during the study period with those for BSIs that occurred during a baseline period January 1999 to January 2000). BSI rates before and after a 1-month intervention (May 2001) were assessed. A case was defined as a new BSI during the study period. RESULTS: The outbreak was polymicrobial, with approximately 30 species. The baseline BSI rate was 0.7 per 100 patient-months. From February 2000 to April 2001, the BSI rate increased to 4.2 per 100 patient-months. Overall, 75% of the BSIs were associated with central venous catheters (CVCs), but CVC use did not fully explain the increase in BSIs. In January 2000, when the center changed ownership, prepackaged CVC dressing kits and biweekly infection control monitoring were discontinued. Beginning in May 2001, staff were educated on CVC care, chlorhexidine replaced povidone-iodine for cutaneous antisepsis, gauze replaced transparent dressings, antimicrobial ointments containing polyethylene glycol at CVC exit sites were discontinued, and patients with CVCs were educated on cutaneous hygiene. After the intervention period, by October 2001, rates decreased to less than 1 BSI per 100 patient-months. CONCLUSIONS: Proper cutaneous antisepsis and access site care is crucial in preventing BSIs in patients receiving hemodialysis. Infection control programs, staff and patient education, and use of optimal antisepsis agents or prepackaged kits are useful toward this end.
OBJECTIVES: Investigate and control an increase in bloodstream infections (BSIs) in an outpatient hemodialysis center. PATIENTS AND DESIGN: A retrospective cohort study was conducted for patients receiving dialysis at the center from February 2000 to April 2001. A case-control study compared microbiological data for all BSIs that occurred during the study period with those for BSIs that occurred during a baseline period January 1999 to January 2000). BSI rates before and after a 1-month intervention (May 2001) were assessed. A case was defined as a new BSI during the study period. RESULTS: The outbreak was polymicrobial, with approximately 30 species. The baseline BSI rate was 0.7 per 100 patient-months. From February 2000 to April 2001, the BSI rate increased to 4.2 per 100 patient-months. Overall, 75% of the BSIs were associated with central venous catheters (CVCs), but CVC use did not fully explain the increase in BSIs. In January 2000, when the center changed ownership, prepackaged CVC dressing kits and biweekly infection control monitoring were discontinued. Beginning in May 2001, staff were educated on CVC care, chlorhexidine replaced povidone-iodine for cutaneous antisepsis, gauze replaced transparent dressings, antimicrobial ointments containing polyethylene glycol at CVC exit sites were discontinued, and patients with CVCs were educated on cutaneous hygiene. After the intervention period, by October 2001, rates decreased to less than 1 BSI per 100 patient-months. CONCLUSIONS: Proper cutaneous antisepsis and access site care is crucial in preventing BSIs in patients receiving hemodialysis. Infection control programs, staff and patient education, and use of optimal antisepsis agents or prepackaged kits are useful toward this end.