BACKGROUND/OBJECTIVES: Catheter Related Bloodstream Infections (CRBSIs) have been
targeted for reduction at our 500-bed university-affiliated hospital since 2001. Prospective,
house-wide surveillance for CRBSI revealed a cluster of CRBSIs with onset shortly
after insertion (≤14 days) among patients with central catheters inserted in Interventional
Radiology (IR) during the months of June through August 2002.
METHODS: The use of Chloroprep®, a 2% CHG/70% alcohol skin prep (ALC/CHG) was introduced
in June 2003. All IR staff were educated and validated on appropriate aseptic technique
and skin prep using the new product. The occurrence of CRBSI initially decreased,
but we did not observe a sustained reduction in infections. In April 2004, the use
of a CHG sponge dressing (Biopatch®) was introduced. This product was used on all
central catheters inserted in IR and removed at the first scheduled dressing change,
7 days following insertion.
CRBSIs were identified by reviewing daily laboratory reports. A CRBSI was defined
as a positive catheter tip culture with a corresponding positive peripheral blood
culture in a patient with a central catheter in place.
RESULTS: The mean rate of CRBSI in catheters inserted ≤14 days was 0.147 per 1000
patient days for the 7 quarters before the use of the CHG sponge. The CRBSI rate decreased
in the 5 quarters following the intervention to 0.036 per 1000 patient days (p = 0.012).
Following the implementation of the CHG sponge dressing, 7 cases occurred within the
14 day post insertion time period. Of these, all but one were identified after 8 days
CONCLUSIONS: Preparing insertion sites with ALC/CHG, together with use of CHG sponge
dressings decreased our short-onset CRBSI occurrence. We have recently expanded the
use of both products. The ALC/CHG will be used for the insertion of all central catheters.
The CHG sponge is now used on all catheters with the exception of Ports and short-term
lines inserted in the OR. It will also be replaced at each dressing change until the
catheter is discontinued.
One limitation of this study is that we expressed the incidence of infections as the
number of CRBSIs per 1000 patient-days, instead of per 1000 catheter-days, which were
not provided by IR. Although using catheter days would provide a more accurate estimate
of CRBSI rates, we have no reason to believe that the number of catheters or catheter
days used in our hospital decreased in the 4 years of this study.