Improving the performance of culture-based bacterial screening by increasing the sample volume from 4 mL to 8 mL in aerobic culture bottles
Article first published online: 29 DEC 2011
© 2011 American Association of Blood Banks
Volume 52, Issue 7, pages 1576–1582, July 2012
How to Cite
Souza, S., Bravo, M., Poulin, T., Vanderpool, S., Kamel, H., Tomasulo, P. and Su, L. L. (2012), Improving the performance of culture-based bacterial screening by increasing the sample volume from 4 mL to 8 mL in aerobic culture bottles. Transfusion, 52: 1576–1582. doi: 10.1111/j.1537-2995.2011.03489.x
- Issue published online: 10 JUL 2012
- Article first published online: 29 DEC 2011
- Received for publication August 30, 2011; revision received October 31, 2011, and accepted November 8, 2011.
BACKGROUND: In the setting of bacterial detection of apheresis platelets (PLTs), the manufacturer recommended PLT inoculation volume for BacT/ALERT culture bottles (bioMérieux) ranges from 4 to 10 mL. This study compares the rate of capture of true-positive (TP) contaminations between aerobic culture bottles inoculated with either 4 or 8 mL of sample and assesses if a larger sample volume reduces time to detection.
STUDY DESIGN AND METHODS: Detection of TP samples and mean time to detection were compared for 4- and 8-mL samples collected between September 1, 2003, and May 2, 2011.
RESULTS: A total of 180,263 and 283,114 PLT collections were tested with an 8- and 4-mL sample, respectively. Analysis of TP rates by volume sampled show an increase in the rate of detection of TP with the 8-mL sample relative to the 4-mL sample (139 vs. 106 per million events; odds ratio, 1.31; 95% confidence interval, 0.77-2.23). Comparison of mean time to detection for TP shows a decrease in mean time to detection using 8 mL compared with 4 mL (12.36 ± 3.7 hr to 15.97 ± 6.3 hr, p = 0.012).
CONCLUSION: Doubling the sample volume to 8 mL showed a trend in improvement for the rate of detection of TP and shortened the mean time of detection for TP by 23% when compared to a sample volume of 4 mL. The decrease in mean time to detection using a larger sample volume suggests that a shorter release time after inoculation could be achieved without significantly increasing patient risk.