Improving the recovery and detection of bloodstream pathogens from blood culture

Kerry Falconer (Lead / Corresponding author), Robert Hammond, Stephen H. Gillespie

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)
97 Downloads (Pure)


Introduction. Bloodstream infections (BSI) are growing in incidence and present a serious health threat. Most patients wait up to 48 h before microbiological cultures can confirm a diagnosis. Low numbers of circulating bacteria in patients with BSI mean we need to develop new methods and optimize current methods to facilitate efficient recovery of bacteria from the bloodstream. This will allow detection of positive blood cultures in a more clinically useful timeframe. Many bacterial blood recovery methods are available and usually include a combination of techniques such as centrifugation, filtration, serum separation or lysis treatment. Here, we evaluate nine different bacteria recovery methods performed directly from blood culture. Aim. We sought to identify a bacterial recovery method that would allow for a cost-effective and efficient recovery of common BSI pathogens directly from blood culture. Methods. Simulated E. coli ATCC 25922 blood culture was used as a model system to evaluate nine different bacteria recovery methods. Each method was assessed on recovery yield, cost, hands-on time, risk of contamination and ease of use. The highest scoring recovery method was further evaluated using simulated blood cultures spiked with seven of the most frequently occurring bloodstream pathogens. The recovery yield was calculated based on c.f.u. count before and after each recovery method. Independent t-tests were performed to determine if the recovery methods evaluated were significantly different based on c.f.u. ml-1 log recovery. Results. All nine methods evaluated successfully recovered E. coli ATCC 25922 from simulated blood cultures although the bacterial yield differed significantly. The MALDI-TOF intact cell method offered the poorest recovery with a mean loss of 2.94±0.37 log c.f.u. ml-1. In contrast, a method developed by Bio-Rad achieved the greatest bacterial yield with a mean bacteria loss of 0.27±0.013 log c.f.u. ml-1. Overall, a low-speed serum-separation method was demonstrated to be the most efficient method in terms of time, cost and recovery efficiency and successfully recovered seven of the most frequent BSI pathogens with a mean bacteria loss of 0.717±0.18 log c.f.u. ml-1. Conclusion. The efficiency of bacterial recovery can vary significantly between different methods and thereby can have a critical impact on downstream analysis. The low-speed serum-separation method offered a simple and effective means of recovering common BSI pathogens from blood culture and will be further investigated for use in the rapid detection of bacteraemia and susceptibility testing in clinical practice.

Original languageEnglish
Pages (from-to)806-811
Number of pages6
JournalJournal of Medical Microbiology
Issue number6
Publication statusPublished - 3 Jun 2020


  • Blood culture and bacterial recovery
  • Bloodstream infection

ASJC Scopus subject areas

  • Microbiology
  • Microbiology (medical)


Dive into the research topics of 'Improving the recovery and detection of bloodstream pathogens from blood culture'. Together they form a unique fingerprint.

Cite this