The Impact of a Hospital-Wide Intervention to Reduce Catheter-Associated Bloodstream Infections: Results from a Nine-Year Program outside the Intensive Care Units in a Middle-Income Country

Background: Central line-associated bloodstream infections remain a frequent healthcare-related infection with substantial morbidity and mortality associated rates.

Objective: Implement a bundle to reduce catheter-related infections outside intensive care units (ICUs).

Methods: Central line-associated bloodstream infection rates were monitored in a 749-beds teaching hospital in Porto Alegre, Brazil, before and during implementation of a hospital-wide prevention program. The program, with special emphasis on a set of interventions consisted in the creation of a catheter insertion protocol in august 2006 (first intervention), when all catheters were only inserted in the ward for ambulatory surgical procedures, with emphasis on hand hygiene, full-barrier precaution, and use of chlorhexidine for skin cleaning. In July 2007 (second intervention) education and a training program for catheter insertion and follow-up were implemented. In March 2008 (third intervention) goals for catheter infection rates were established and feedback was given to hospital wards. We monitored central line-associated bloodstream infection rates in adult open wards on a monthly basis from January 2002 to October 2009. Secondary outcomes measures were number of catheter-days and central line use ratio. Results: In the study period, we observed 1,484 infections in 1,872,937 catheter-days for the entire hospital.  Outside ICUs the mean central line-associated bloodstream infection rates decreased in 64% from 2002 to 2009, respectively, 9.3 to 3.3 infections per 1.000 days of catheter use. In the same period, the rate of catheter use increased from 10.0 in 2002 to 16.1 in 2009, an increase in 61%. By segmented regression analysis each intervention isolated did not result in significant reduction in the rate of catheter infection. Considering the three interventions taken together, comparing the period before the first intervention and after the third intervention we observed a significant reduction in level of infection (P = 0.03) by segmented regression analysis. Conclusions: The three segmented interventions promoted a sustained reduction in central line-associated bloodstream infection in the eight-year studied period. The implementation of a hospital-wide program demonstrated to be an achievable target that resulted in 102 prevented infections in 41 months after the first intervention outside intensive care units.