Effect of a blood culture collection bundle on decreasing the contamination rate.

In this study, we examined the effect of a bundled approach to blood collection for blood culture on decreasing contamination. Commensal organisms were considered contaminants on the basis of the clinical course if they were recovered from only a single blood draw (set) and if a positive result for two sets was not obtained within 72 hours. The main elements of the bundle were blood collection by venipuncture, skin preparation with a chlorhexidine alcohol swab, disinfection of culture bottles, and use of a sterile blood transfer device instead of the two-needle technique for inoculation. In the bundle intervention, chlorhexidine alcohol was first introduced in the hospital, and use of the blood transfer device was increased during the intervention. Both items were used in most patients requiring blood cultures. Blood collection through a line caused contamination in only one case. The contamination rate decreased significantly from 2.0% to 1.0% after introduction of the bundle approach (3-year control period vs. 2-year bundle period, p<0.001), and a significant decrease in the contamination rate was observed for coagulase-negative Staphylococcus (p<0.001). A high contamination rate was found in August and September during the control period. The contamination rate of Bacillus species was significantly higher in those 2 months than in other months. A seasonal increase was not observed during the bundle period. A low contamination rate of 1.0% was achieved using our bundled collection approach for blood culture.