Background: United Blood Services (UBS) began bacterial testing of platelets (PLTs) using one-bottle cultures in September 2003. Collection of 7-day PLTs using two-bottle cultures began in April 2006. This study compares our experience using both systems.
Study design and methods: PLTs from 13 UBS regional centers cultured from September 1, 2003, to September 1, 2007, were included in the analysis. Positive-signal bottles from a commercially available microbial detection system (BacT/ALERT, bioMérieux) were sent, with corresponding PLTs if available, for confirmatory testing using plate culture media. AABB definitions were used with modifications.
Results: A total of 51,265 7-day PLT collections and 191,521 5-day PLT collections were tested with bacterial cultures. The overall true-positive (TP) rate for the two-bottle system (1:8544) was comparable to the TP rate with the previous one-bottle system (1:6344). In three of six yield cases, only the anaerobic bottle was positive (two cases of Group D Streptococci, one case of Corynebacterium spp.). The false-positive (FP) and indeterminate (IND) rates in the anaerobic bottle (1:1767 and 1:1830, respectively) were significantly higher than those in the aerobic bottle (1:6408 and 1:17,088, respectively; p < 0.001). One confirmed transfusion-related septic reaction, classified as a late TP after investigation, was reported out of 242,786 tested PLT donations.
Conclusion: The rate of TP cases by the two-bottle system was not increased over the one-bottle system, although anaerobic-bottle-only positive cases were detected. FP and IND rates were increased in the two-bottle system, attributable to the anaerobic bottle. Observation of only one documented transfusion-related septic reaction in 4 years of bacterial screening at UBS is reassuring, although limitations in passive surveillance and higher rates of reactions reported by others indicate the need for continued vigilance.