Rapid identification of bacteria from bioMerieux BacT/ALERT blood culture bottles by MALDI-TOF MS

Abstract

Several studies have reported poor results when trying to identify microorganisms directly from the bioMérieux BacT/ALERT blood culture system using matrix-assisted laser desorption/ionisation-time of flight (MALDI-TOF) mass spectrometry. The aim of this study is to evaluate two new methods, Sepsityper and an enrichment method for direct identification of microorganisms from this system. For both methods the samples were processed using the Bruker Microflex LT mass spectrometer (Biotyper) using the Microflex Control software to obtain spectra. The results from direct analysis were compared with those obtained by subculture and subsequent identification. A total of 350 positive blood cultures were processed simultaneously by the two methods. Fifty-three cultures were polymocrobial or failed to grow any organism on subculture, and these results were not included as there was either no subculture result, or for polymicrobial cultures it was known that the Biotyper would not be able to distinguish the constituent organisms correctly. Overall, the results showed that, contrary to previous reports, it is possible to identify bacteria directly from bioMérieux blood culture bottles, as 219/297 (74%) correct identifications were obtained using the Bruker Sepsityper method and 228/297 (77%) were obtained for the enrichment method when there is only one organism was present. Although the enrichment method was simpler, the reagent costs for the Sepsityper method were approximately pound 4.00 per sample compared to pound 0.50. An even simpler and cheaper method, which was less labour-intensive and did not require further reagents, was investigated. Seventy-seven specimens from positive signalled blood cultures were analysed by inoculating prewarmed blood agar plates and analysing any growth after 1-, 2- and 4-h periods of incubation at 37 degrees C, by either direct transfer or alcohol extraction. This method gave the highest number of correct identifications, 66/77 (86%), and was cheaper and less labour-intensive than either of the two above methods.

MALDI-TOF mass spectrometry following short incubation on a solid medium is a valuable tool for rapid pathogen identification from positive blood cultures

“…In comparison to direct MALDI-TOF MS, similarly low identification rates (0-15%) were encountered for Gram-positive rods (La Moussaoui et al, 2010;Saffert et al, 2012;Stevenson et al, 2010), but results were contradictory for yeasts with most researchers describing poor performance (Buchan et al, 2012;Ferreira et al, 2011;Haigh et al, 2013) and others reporting high fungal species identification rates of >90% (Spanu et al, 2012;Yan et al, 2011). Thus, even though rapid identification of yeasts would be desirable as mortality of fungal bloodstream infection is high especially in the event of delayed antifungal therapy (Garey et al, 2006), our protocol does unfortunately not provide this opportunity.…”

“…The published studies on direct MALDI-TOF MS are based on various methods for specimen processing including differential centrifugation (Christner et al, 2010;Ferreira et al, 2011;Haigh et al, 2013;La Scola and Raoult, 2009;Loonen et al, 2012;Moussaoui et al, 2010;Saffert et al, 2012;Vlek et al, 2012) or detergent lysis either in the form of in-house protocols (Chen et al, 2013;Clerc et al, 2013;Leli et al, 2013;Martiny et al, 2012b;Meex et al, 2012;Prod'hom et al, 2010;Saffert et al, 2012;Spanu et al, 2012) or the commercially available Sepsityper kit (Buchan et al, 2012;Chen et al, 2013;Haigh et al, 2013;Idelevich et al, 2014;Klein et al, 2012;Kok et al, 2011;Lagace-Wiens et al, 2012;Loonen et al, 2012;Martinez et al, 2014;Martiny et al, 2012b;Meex et al, 2012;Saffert et al, 2012;Schubert et al, 2011;Yan et al, 2011). Further, those studies differ in instrumentation with respect to MALDI-TOF MS and automated blood culture system, and protocol for MALDI-TOF MS analysis itself including extraction method, replicate testing and interpretation rules for analysis of results.…”

“…One explanation might be that low confidence scores reflect suboptimal sample conditions (low analyte concentration, prominent background signal) rather than a low degree of relatedness between the sample isolate and the best database match. Consequently, others have already applied modified interpretive criteria, but there is inconsistence regarding the "right" threshold for valid identification to the species level: it was used ≥1.9 (La Stevenson et al, 2010), ≥1.8 (Martinez et al, 2014), ≥1.7 (Haigh et al, 2013;Moussaoui et al, 2010;Saffert et al, 2012;Vlek et al, 2012), ≥1.6 (Martiny et al, 2012b), ≥1.5 (Schubert et al, 2011), and even lower cut-offs (La Meex et al, 2012;Moussaoui et al, 2010), sometimes provided that the first 2-4 successive proposals from the identification matches ranking list referred to a single species. With such modified thresholds, improved identification rates were reported with overall correct species identification in about 75% in average (about 90% and 55% for Gram-negative and Gram-positive bacteria, respectively).…”

“…without preceding subculture, from positive blood culture bottles (Buchan et al, 2012;Chen et al, 2013;Christner et al, 2010;Ferreira et al, 2011;Haigh et al, 2013;Klein et al, 2012;Kok et al, 2011;La Scola and Raoult, 2009;Lagace-Wiens et al, 2012;Leli et al, 2013;Loonen et al, 2012;Martinez et al, 2014;Martiny et al, 2012b;Meex et al, 2012;Moussaoui et al, 2010;Prod'hom et al, 2010;Saffert et al, 2012;Schubert et al, 2011;Spanu et al, 2012;Stevenson et al, 2010;Yan et al, 2011). As the blood culture broth contains macromolecules from blood and growth media which have to be removed prior to MALDI-TOF MS analysis to reveal species-specific protein spectra of bacteria and yeasts, all procedures for direct MALDI-TOF MS are associated with additional expenditure of consumables, costs and hand-on processing time.…”

Fast and accurate bacterial species identification in biological samples using LC-MS/MS mass spectrometry and machine learning

“…This proof of concept could pave the way to the development of new peptide signatures for the analysis of other types of clinical specimens (bronchoalveolar lavage (BAL), stool, hemoculture…) (47)(48)(49), but also for the detection of foodborne or waterborne pathogens (50,51), to reduce the turnaround time required to obtain a genus-and/or speciesspecific identification of microorganisms by classical or molecular microbiology methods or MALDI-TOF mass spectrometry. For some of these applications, without any culture, the sensitivity of the method might be too low to detect the bacteria but, it is expected that a short-term culture in a liquid medium might be enough to reach a detectable level (<1 x 10 3 CFU/mL for certain peptides) without the isolation of colonies on a culture plate.…”