Limitations of current practices in detection of bacterially contaminated blood products associated with suspected septic transfusion reactions.

ABSTRACT

BACKGROUND: In the setting of suspected septic transfusion reactions, bacterial culture of both the transfused patient and the residual blood component is recommended. Primary bacterial contamination can occur at the time of component collection. Clinically insignificant "secondary contamination" can occur during post-transfusion component discard, retrieval for culture, or manipulation of the bag at the time of culture sampling. STUDY DESIGN AND METHODS: This retrospective, multi-center study analyzes positive residual component culture results and companion patient blood cultures from 15 hospitals, 1 blood center, and all cultured transfusion reactions within the province of Quebec, Canada, over a 5-year period. Imputability was assigned as "definite" (concordant growth), "possible" (discordant growth or lack of growth in patient culture), or "unable to assess" (patient not cultured). RESULTS: There were 373 positive component cultures from 360 unique transfusion reactions, with 276 (76.7%) companion patient blood cultures performed, of which 10 (2.8%) yielded the pathogen detected in the positive component. Of these 10 definite pathogens, 7 (2 Staphylococcus aureus, 3 other staphylococci, and 1 Streptococcus pyogenes and 1 Bacillus sp.) were associated with platelet and 3 (Aeromonas veronii, Staphylococcus epidermidis, and Enterococcus faecalis) with RBC transfusions. RBC and plasma components comprised 70% of positive component cultures. DISCUSSION: The process of performing residual component culture is vulnerable to secondary contamination. The significance of microorganisms recovered from component culture cannot be interpreted in isolation. In the context of low prevalence of primary contamination of blood components, the positive predictive value of a positive component culture result is very low.