Bacterial culture of apheresis platelets: a mathematical model of the residual rate of contamination based on unconfirmed positive results
Version of Record online: 2 JUL 2013
© 2013 International Society of Blood Transfusion
Volume 106, Issue 1, pages 23–30, January 2014
How to Cite
Benjamin, R. J., Dy, B., Perez, J., Eder, A. F. and Wagner, S. J. (2014), Bacterial culture of apheresis platelets: a mathematical model of the residual rate of contamination based on unconfirmed positive results. Vox Sanguinis, 106: 23–30. doi: 10.1111/vox.12065
- Issue online: 17 DEC 2013
- Version of Record online: 2 JUL 2013
- Manuscript Revised: 4 JUN 2013
- Manuscript Accepted: 4 JUN 2013
- Manuscript Received: 11 JAN 2013
Platelet septic reactions result from low concentrations of bacteria that escape detection by quality-control BacT/ALERT™ culture testing. We estimate the contamination rate with these bacteria at the time of testing using a mathematical model.
Culture results and reported septic reactions are described for platelets collected between January 2007 and December 2011. Initial positive results with negative confirmatory cultures were reclassified assuming some of the ‘unconfirmed positive results’ represent collections contaminated with low-concentration, dormant bacteria. A mathematical model based on the probability of the detection of bacteria describes the upper limit of the residual rate of contamination.
The rate of confirmed or unconfirmed positive apheresis platelet donations was 188 per million (1:5317) and 110 per million (1:9124), respectively. The rate of post-transfusion sepsis and reported fatalities per distributed component was 1:106 931 and 1:1 015 843, respectively. A linear decrease in unconfirmed positive Bacillus spp. cultures most likely reflected diminishing environmental contamination over time. The remaining unconfirmed positive results identified similar bacteria species as those associated with septic reactions. Assuming that these represent contamination of the collection with low-concentration, dormant bacteria, the model identified a residual contamination of 3524–204 per million (1:284–1:4902) for collections contaminated with 1–20 bacteria, respectively.
Greater than 99·5% of collections contain no viable, aerobic bacteria in solution at the time of early culture testing. For every confirmed positive contaminated collection detected, there are at most 19 collections with low concentrations of dormant bacteria that are not readily detected by early BacT/ALERT™ culture.