Accepted Manuscript

Direct identification and antimicrobial susceptibility testing of

microorganisms from positive blood cultures following isolation
by lysis-centrifugation

Aaron Campigotto, Lee Goneau, Larissa M. Matukas

PII: S0732-8893(18)30196-2
DOI: doi:10.1016/j.diagmicrobio.2018.06.010
Reference: DMB 14623
To appear in: Diagnostic Microbiology & Infectious Disease
Received date: 16 January 2018
Revised date: 21 May 2018
Accepted date: 13 June 2018

Please cite this article as: Aaron Campigotto, Lee Goneau, Larissa M. Matukas , Direct
identification and antimicrobial susceptibility testing of microorganisms from positive
blood cultures following isolation by lysis-centrifugation. Dmb (2018), doi:10.1016/
j.diagmicrobio.2018.06.010

This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
 ACCEPTED MANUSCRIPT

Direct identification and antimicrobial susceptibility testing of microorganisms from positive blood

cultures following isolation by lysis-centrifugation

Aaron Campigotto*a, Lee Goneau*b, Larissa M. Matukas#,c,d

T
*Both authors contributed equally to this work

IP
CR
a
Division of Microbiology, Department of Pathology and Laboratory Medicine, London Health Sciences

US
Centre, London, ON, Canada; bPublic Health Ontario, Toronto, ON, Canada; cDivision of Infectious

Diseases, Department of Medicine, St. Michael’s Hospital, Toronto, ON, Canada; dDivision of
AN
Microbiology, Department of Laboratory Medicine, St. Michael’s Hospital, Toronto, ON, Canada
M
ED

#Address correspondence: Dr. Larissa M. Matukas, matukasl@smh.ca

PT

Present address: Division of Microbiology, St. Michael’s Hospital, 30 Bond Street, Toronto, ON, Canada,

M5B 1W8
CE
AC

Running title: Direct ID and AST from positive blood cultures

 ACCEPTED MANUSCRIPT

Abstract

We evaluated a direct from positive blood culture pelleting procedure that utilizes a lysis-centrifugation

protocol for the identification of microorganisms by MALDI-TOF MS with subsequent antimicrobial

susceptibility testing (AST) and rapid methicillin- and beta-lactam-resistance screening. The

identification evaluation was performed on 125 cultures and resulted in the correct genus level

T
identification in 91.2% of cultures and a species level concordance of 82.4% compared to routine

IP
subcultured growth. For the AST evaluation, susceptibility results from direct pelleting and subcultured

CR
growth for 187 cultures were compared; an average ±twofold dilution agreement of 98.2% (1650/1681)

US
and 98.6% (1375/1394) for Gram-negatives and -positives respectively was found. Major errors fell

below 5% except for MRSA, which was falsely reported as oxacillin sensitive 17.2% (11/66) of the time.
AN
Lastly, the sensitivity and specificity of rapid MRSA screening was 94.7% (36/38) and 90.0% (9/10)

respectively, while the ESBL screening results were 90.3% (65/72) and 100.0% (13/13) respectively.
M

Keywords: antimicrobial susceptibility, MALDI-TOF, blood culture

ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

1. Introduction

The accurate identification and antimicrobial susceptibility testing (AST) of the pathogen causing a

bloodstream infection is a one of the most important tasks carried out by the clinical microbiology

laboratory. Sepsis is associated with high mortality rates and the time to effective antibiotic therapy is

an important determinant in reducing both morbidity and mortality (1). While clinical presentation,

T
epidemiologic factors, and Gram stain results can guide empiric antimicrobial therapy, definitive

IP
antimicrobial therapy relies on identification and AST. Importantly, obtaining pathogen identification to

CR
the genus or species level, is essential for those organisms that may not be covered by typical empiric

US
agents or intrinsically resistant organisms that cannot be differentiated from susceptible ones by Gram

staining alone (e.g. Pseudomonas species, Enterobacter species). With traditional microorganism
AN
identification, performed by analyzing biochemical reactions and carbohydrate fermentation, growth of

the organism from solid media is usually required; similarly organism growth on solid media is typically
M

required for AST although methods to expedite both traditional identification and susceptibility testing
ED

have been reported (2).

PT

Matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) has been

introduced into the clinical microbiology lab for faster and more accurate microorganism identification
CE

(3, 4). Similar to traditional identification tests, MALDI-TOF MS identification is valid for organisms
AC

growing from solid media. Various methods for direct identification of organisms from positive blood

cultures generally rely on the lysis of the human blood components in the blood culture broth followed

by the separation and concentration of the microorganism by centrifugation or filtration (5–8); the

microorganism in the resulting concentrate is then identified by MALDI-TOF MS.

The limitations of these procedures include high cost, large volume of blood culture broth required,

lengthy procedure time, lack of consistently reliable gram positive bacterial identification or lack of
 ACCEPTED MANUSCRIPT

viable organism for direct AST of the concentrate (2, 8, 9). The present study evaluated a lysis-

centrifugation method that overcame many of these limitations, in particular the loss of viability for

direct AST (5). Additionally, the use of a centrifugation method allowed for a lower cost per sample and

decreased processing time between samples (10).

We compared the accuracy of MALDI-TOF MS identification from positive blood cultures, using a lysis-

T
centrifugation direct pelleting source to the identification from traditional overnight sub-cultured

IP
growth from solid media. We subsequently demonstrated that this lysis-centrifugation method results

CR
in a pellet that may be used in AST leading to a decreased time to a finalised result, thus providing

US
guidance for directed treatment.
AN
2. Materials and Methods
M

2.1 Blood Culture Specimens for Identification

ED

St. Michael’s Hospital, a predominantly adult 475-bed academic hospital in Toronto, Canada, which has
PT

three adult ICUs and a coronary care unit, provides microbiology services for itself and to a community

hospital system within the greater Toronto area. Blood cultures in the current study were collected
CE

between February and June 2015 (identification portion) and between July and January 2015 (AST
AC

portion), in BacT/Alert Standard Aerobic (SA), Standard Anaerobic (SN) or (Pediatric Fastidious

Antimicrobial Neutralization) PF plus bottles and incubated in the BacT/Alert 3D (bioMérieux, St.

Laurent, Quebec ) automated detection system at the hospital in which the patient was admitted.

Processing of all positive blood cultures, including those from the outside client was performed at the St.

Michael’s Hospital microbiology laboratory during daytime lab hours; when a positive culture was
 ACCEPTED MANUSCRIPT

identified at the outside lab in an offsite automated detection system, the positive blood culture bottle

was also transported to St. Michael’s Hospital for processing during daytime lab hours.

For the identification portion of the study, the first positive blood culture bottle per patient per 24-hour

period was included in the evaluation if Gram-stain demonstrated a single organism. This was designed

to mimic the anticipated use of the pelleting procedure in the laboratory once routinely employed. In

T
addition, this procedure limited replicates of the pelleting procedure on the same microorganism

IP
increasing the diversity of organisms tested for this study.

CR
2.2 Routine Microorganism Identification

US
All blood culture bottles, irrespective of Gram-stain reaction, were subcultured onto 5% sheep blood,
AN
chocolate, MacConkey, and/or colistin-nalidixic acid (CNA) agar (Oxoid, Ontario, Canada). Inoculated

media were incubated at 35⁰C in 5% CO2 with the exception of the MacConkey agar, which was
M

incubated in ambient air. The plates were examined at 24 and 48 hours. If the anaerobic blood culture
ED

bottle was positive, brucella vitamin K agar (BVKA) plate was included and incubated anaerobically for

48 hours. If yeast were seen on Gram-stain, a Sabouraud agar was also inoculated and held aerobically
PT

at 30⁰C.
CE

Following incubation, microorganism growth on solid media was identified by the Bruker Microflex LT

MALDI-TOF MS (Bruker Daltonics). Briefly, isolates were spotted onto a target plate and 1 µl of α-cyano-
AC

4-hydroxycinnamic acid matrix was applied and allowed to dry prior to identification by MALDI-TOF MS.

Identification results were reported on the Biotyper software (3.0). Scores ≥2.000 were accepted as

species level identification, while a Biotyper score between 1.700 and 2.000 was accepted as a genus

level identification consistent with the instrument recommendations and the laboratory’s reporting

structure. If a Biotyper score from the solid media was found to be <1.700, it was discarded from

analysis as we felt this represented either a database or technical limitation of the MALDI-TOF MS. All
 ACCEPTED MANUSCRIPT

Candida species and organisms not achieving a Biotyper score ≥2.000 underwent on plate extraction

with 1 µl of formic acid (70% v/v).

2.3 Direct Microorganism Identification

1.5 mL of blood from positive blood culture bottles was added to a 2 ml microcentrifuge tube to which

T
0.5 ml of lysis buffer [0.6% polyoxyethylene 10 oleoyl ether (Brij 97) in 0.4 M 3-cyclohexylamino-1-

IP
propane sulfonic acid (CAPS)] was added. This mixture was vortexed for 3-5 seconds and then incubated

CR
at room temperature for 2 minutes. The resulting solution was centrifuged at 12,000 rpm for 1 minute

30 seconds and the supernatant was discarded by pipette. The pellet was suspended in 1 ml of wash

US
buffer (20 mM sodium phosphate, 0.05% Brij 97, and 0.45% NaCl). Immediately following suspension,

the solution was again centrifuged at 12,000 rpm for 1 minute 30 seconds. The supernatant was again
AN
discarded by pipetting off the liquid, only leaving the pellet.
M

The pellet was spotted onto the MALDI-TOF target plate with a sterile, autoclaved toothpick. Gram
ED

positive bacteria and Candida species underwent on-plate extraction with 1 µl of formic acid (70% v/v).

This was allowed to air dry and 1 µl of α-cyano-4-hydroxycinnamic acid matrix was applied. For Gram
PT

negative bacteria, only the 1 µl of α-cyano-4-hydroxycinnamic acid matrix was applied. If a score was
CE

<2.000, a one-time refire of the MALDI-TOF target was allowed, as is consistent with our identification

procedure from the solid media. Microorganism identification scores reported on the Biotyper software
AC

were recorded.

2.4 Blood Culture Specimens for Bacterial AST testing

187 culture-negative BacT/Alert SA Standard Aerobic blood culture (BC) bottles (bioMérieux) were

spiked with 100 CFU/mL of 24-hour sub-cultured Gram-positive and -negative organisms and incubated

in the BacT/ALERT (bioMérieux) BC system. Organisms were selected based on their resistance profiles,
 ACCEPTED MANUSCRIPT

with ~50% of isolates demonstrating a resistance mechanism including methicillin resistant S. aureus

(MRSA), vancomycin resistant enterococci (VRE), extended-spectrum producing Enterobacteriaceae

(ESBL), carbapenemase-producing organisms (CPO), and multidrug resistant organisms (MDR; for

example, Pseudomonas and Acinetobacter species). Blood extracted from positive bottles was both

pelleted and subcultured to facilitate method comparison. Susceptibility profiles were generated from

T
sub-cultured and pelleted isolates using the Vitek2 (bioMérieux) AST system and AST-GP67 (Gram-

IP
positive) or AST-N220 (Gram-negative) cards. Acceptance criteria included essential and categorical

CR
agreement ≥90%, with combined major and minor errors of <10% (11, 12). Major errors were

considered as a susceptible result determined by one method but resistant by the other method. Minor

US
errors were considered whenever one method resulted in intermediate susceptibility determination
AN
while the other method determined either susceptible or resistant. The pellet was also inoculated to an

oxacillin screen plate to assess for MRSA resistance. Growth on plates was assessed after 18- to 24-
M

hours.
ED

In addition to susceptibility testing, resistance screening was carried out using both standard subculture

and pellet approaches for both S. aureus and Enterobacteriaceae. In total, 48 S. aureus isolates were
PT

assessed for methicillin resistance (MRSA) using the MRSA Latex Test kit (Denka Seiken), while 85
CE

Enterobacteriaceae were assessed for extended spectrum beta-lactamase (ESBL)-production using the β

LACTA Test kit (Bio-Rad). Manufacturer instructions were followed for both assays, with the exception
AC

that pelleted organism was used as the inoculum.

3. Results

3.1 Comparison of MALDI-TOF Identification Using Direct Lysis-Centrifugation and Standard Subculture

Approaches
 ACCEPTED MANUSCRIPT

125 positive blood cultures were included for study, comprising of 64 Gram-positive bacteria, 57 Gram-

negative bacteria, and 4 yeasts (Table 1). MALDI-TOF MS correctly identified organisms to the genus

level in 117 (93.6%) bacterial pellets. Growth of the organisms from the pellet was demonstrated in

representative Gram-positive and Gram-negative organisms and yeast (data not shown). Furthermore,

AST is dependent on viable organisms, our AST would have failed if the organisms had died through the

T
pelleting process. The eight isolates not identified to the genus level, were mainly Gram-positive

IP
organisms (six of eight isolates) of which three were S. pneumoniae. There were no incorrect MALDI-

CR
TOF MS genus identifications when the source was a blood culture pellet.

US
One hundred and fourteen isolates from subculture growth met the more stringent species

identification acceptance criteria compared to 103 (82.4%) from the direct blood culture pellet source.
AN
This resulted in a concordance rate to the species level, using a Biotyper score of ≥2.000 by both

methods, of 90.4% (Table 2). Organisms that failed to identify to species level were predominantly
M

Gram-positive bacteria, regardless of which source was used. Of the 64 Gram-positive isolates identified
ED

to the species level from the subcultured solid media, 52 (81.3%) were also identified from the direct

pellet source. Fewer coagulase negative staphylococci species (CNSS) were identified to the species
PT

level from either subcultured growth (20/23, 87%) or the direct pellet source (16/23, 70%), whereas, all
CE

S. aureus (n=13) were identified correctly from either source method (Table 1).
AC

Gram-negative bacteria were identified to the species level in 100% and 93% from subcultured growth

and the direct pelleting source respectively. The four fungemias due to C. albicans, C. parapsilosis, and C.

tropicalis, were all adequately named to the species level from either the subcultured growth or from

the direct pelleting source.

3.2 Evaluation of Antimicrobial Susceptibility Testing Results Using Direct Lysis-Centrifugation and

Standard Subculture Approaches

 ACCEPTED MANUSCRIPT

A comparison between using standard subcultured growth versus blood culture pellets as the source for

AST testing was performed on 187 isolates (95 were Gram-negative, 92 were Gram-positive); these

isolates were unique from those used for the bacterial identification study.

A total of 3075 antibiotic comparisons were evaluated, with an average ±twofold dilution agreement of

98.2% (1650/1681) and 98.6% (1375/1394), and categorical agreement of 98.6% (1628/1651) and 98.5%

T
(1373/1394) for Gram-negative and -positive cards respectively (Table 3ab). Major error rates fell below

IP
5% for all antibiotics tested except for MRSA, which was falsely reported as oxacillin sensitive 17.2%

CR
(11/66) of the time resulting in a very major error. Minor error rates were high (12.6%) for the

US
antimicrobial agents amoxicillin-clavulanic acid and nitrofurantoin in Gram-negatives tested using the

GN220 AST card. However, this could be attributed to intermediate MIC values, which fall only one two-
AN
fold dilution away from resistance interpretation. Thus, these errors could be attributed to normal test-

to-test variation using the Vitek2 instrument. High essential agreement values (97.9%) for these agents
M

support this hypothesis. Data logged in the laboratory information system facilitated comparison of AST
ED

resulting time for both methods. Pelleting reduced the average time from Gram-stain result to Vitek2

AST report from 41.6 to 11.6 hours compared to the sub-culture method (Table 4). The presence of
PT

methicillin resistance, detected through the use of plating to oxacillin screen plates, was correctly
CE

identified in 100% (30/30) of pelleted S. aureus isolates .

AC

3.3 Rapid Antimicrobial Resistance Screening Using Lysis-Centrifugated Isolates

Forty-eight Staphylococcus aureus isolates were assessed for methicillin resistance (MRSA), while 85

Enterobacteriaceae were assessed for extended spectrum beta-lactamase (ESBL)-production. The

sensitivity and specificity of the MRSA Latex Test was 94.7% (36/38) and 90.0% (9/10), while the ESBL β

LACTA Test was 90.3% (65/72) and 100.0% (13/13) respectively (Table 5). The ESBL β LACTA test was

reliably interpreted without difficulty. However, MRSA Latex Test results were much less strongly
 ACCEPTED MANUSCRIPT

reactive when conducted with pelleted S. aureus compared to those done using 24-hour sub-cultured

growth, leading to potential falsely-negative interpretation for inexperienced users.

4. Discussion

T
With the ability of the MALDI-TOF MS to rapidly and reliably identify microorganisms, its use in the

IP
clinical microbiology laboratory has become routine and innovative methods to utilize this technology

CR
are evolving. Direct MALDI-TOF MS identification and AST from sterile sites, like blood cultures studied

here, has the potential to decrease organism identification by hours if not days. In this study, we were

US
able to compare the use of a direct pelleting procedure, which is a modification of published methods,
AN
to routine MALDI-TOF MS identification from subcultured growth on solid media; compared to the

previously published methods, the modifications, specifically the use of a pelleting procedure as
M

opposed to a filtration method, means that there is both less equipment to purchase and less
ED

equipment to clean and sterilize between use.

PT

Multiple methods of direct identification from positive blood cultures have been proposed with correct

identification to the species level ranging from 67% to 93%. The performance of the direct pellet
CE

method described here was within the upper ranges reported (5, 7). Commonly found with these

published procedures is the ability to correctly identify Gram-negative bacteria more frequently than
AC

Gram-positive bacteria, a finding we also encountered. To partially overcome this limitation, formic acid

was applied to all Gram-positive bacteria. Many (7/12, 58%) of the Gram-positive bacteria that were

unable to be identified to the species level were CNSS; identification to the genus level, however, was

successful. This identification only to the genus level is unlikely to be significant in many clinical

situations since the isolation of a CNSS may represent contamination. It is important to note that S.

aureus was consistently distinguished from the CNSS.

 ACCEPTED MANUSCRIPT

The overall concordance of the pelleting technique to identify isolates to the genus level was higher

than most other described methods at 93.6%. For instance, the use of a short incubation plate may only

reach a correct genus level identification greater than 80% after 4 hour (13) or even up to 8 hours of

incubation (14). No isolate was identified to the genus level through the direct pelleting procedure

incorrectly.

T
The major limitations for Gram-positive bacterial identification to the species level occurred most often

IP
with CNSS, where the species differed between the identification from solid media and the direct

CR
pelleting method, and S. pneumoniae, where no identification was able to be obtained from the direct

US
method while identification from subcultured media was correct. In each of the three cases of S.

pneumoniae bacteremia, the blood culture bottle remained in the automated culture system for an
AN
additional eight to twelve hours after flagging positive, as a result of not having a 24-hours laboratory

for processing cultures. In each of these instances, only a few colonies were recovered on the subculture
M

plate indicating a very low viable bacterial load in the bottle. The minimum bacterial density required to
ED

produce a pellet for MALDI-TOF MS identification has been shown to be 106 CFU/ml, and with this

density a minimum of 6 ml of blood was required to produce a pellet; in the above instance, this
PT

threshold may not have been achieved due to the autolysis of S. pneumoniae prior to its processing (15).
CE

Automated AST conducted using this lysis-centrifugation approach was a reliable method for rapidly
AC

determining resistance profiles of bacteria. Removing the subculture step has the potential to allow

same-day susceptibility results, while antimicrobial resistance screening from pellets permitted

resistance determination in approximately one hour post blood culture positivity. However, we did

observe low discrimination of MRSA identification from blood culture pellets tested using the Vitek2

automated system, with only 66.6% of MRSA correctly identified (20/30) compared to the standard sub-

culture approach. Importantly, this was not of significant concern, as subculturing pellets of MALDI-TOF
 ACCEPTED MANUSCRIPT

MS identified S. aureus directly to oxacillin screening plates correctly identified the presence of

methicillin resistance. Thus, oxacillin screening plates can be used in place of the Vitek2 for the

detection of MRSA while still providing results in less than 24-hours. Furthermore, as latex agglutination

was able to detect 9/10 MRSA isolates, we believe that additional studies examining this technique are

warranted to determine if this method may supplement or replace oxacillin screen plates.

T
5. Conclusion

IP
CR
The ability of this procedure to generate a reliable identification in almost 90% of positive blood

cultures, and permit rapid AST resulting and resistance screening, when compared to identification from

US
solid media, demonstrates that this method may have an impact on changing microbiology practices.

Further, these approaches have the potential to greatly impact patient management through rapid and
AN
judicious use of appropriate antibiotics, while improving upon infection prevention and control
M

initiatives through timely isolation of patients infected with antimicrobial resistant organisms. We are

currently in the process of implementation within our laboratory with further investigations required to
ED

determine the impact of this intervention on antimicrobial prescribing.

PT
CE

Funding
AC

This research did not receive any specific grant from funding agencies in the public, commercial, or not-

for-profit sectors.
 ACCEPTED MANUSCRIPT

T
IP
CR
References

1. Kumar A, Roberts D, Wood KE, Light B, Parrillo JE, Sharma S, Suppes R, Feinstein D, Zanotti S,

US
Taiberg L, Gurka D, Cheang M. 2006. Duration of hypotension before initiation of effective
AN
antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care

Med2006/04/21. 34:1589–1596.
M

2. Lupetti A, Barnini S, Castagna B, Capria A-LL, Nibbering PH. 2010. Rapid identification and
ED

antimicrobial susceptibility profiling of Gram-positive cocci in blood cultures with the Vitek 2
PT

system. Eur J Clin Microbiol Infect Dis 29:89–95.

3. Sauer S, Kliem M. 2010. Mass spectrometry tools for the classification and identification of
CE

bacteria. Nat Rev Microbiol 8:74–82.

AC

4. Tan KE, Ellis BC, Lee R, Stamper PD, Zhang SX, Carroll KC. 2012. Prospective evaluation of a

matrix-assisted laser desorption ionization-time of flight mass spectrometry system in a hospital

clinical microbiology laboratory for identification of bacteria and yeasts: a bench-by-bench study

for assessing the impact on ti. J Clin Microbiol 50:3301–8.

5. Fothergill A, Kasinathan V, Hyman J, Walsh J, Drake T, Wang YF. 2013. Rapid identification of
 ACCEPTED MANUSCRIPT

bacteria and yeasts from positive-blood-culture bottles by using a lysis-filtration method and

matrix-assisted laser desorption ionization-time of flight mass spectrum analysis with the

SARAMIS database. J Clin Microbiol 51:805–809.

6. Foster AGW. 2013. Rapid identification of microbes in positive blood cultures by use of the vitek

ms matrix-assisted laser desorption ionization-time of flight mass spectrometry system. J Clin

T
Microbiol 51:3717–3719.

IP
CR
7. Ferroni A, Suarez S, Beretti JL, Dauphin B, Bille E, Meyer J, Bougnoux ME, Alanio A, Berche P,

Nassif X. 2010. Real-time identification of bacteria and Candida species in positive blood culture

US
broths by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin

Microbiol 48:1542–1548.
AN
8. Meex C, Neuville F, Descy J, Huynen P, Hayette MP, De Mol P, Melin P. 2012. Direct identification
M

of bacteria from BacT/ALERT anaerobic positive blood cultures by MALDI-TOF MS: MALDI
ED

Sepsityper kit versus an in-house saponin method for bacterial extraction. J Med Microbiol

61:1511–1516.
PT

9. Klein S, Zimmermann S, Köhler C, Mischnik A, Alle W, Bode KA. 2012. Integration of matrix-
CE

assisted laser desorption/ionization time-of-flight mass spectrometry in blood culture

diagnostics: A fast and effective approach. J Med Microbiol 61:323–331.

AC

10. Machen A, Drake T, Wang YF (Wayne), Anderson A, Melendez A. 2014. Same Day Identification

and Full Panel Antimicrobial Susceptibility Testing of Bacteria from Positive Blood Culture Bottles

Made Possible by a Combined Lysis-Filtration Method with MALDI-TOF VITEK Mass Spectrometry

and the VITEK2 System. PLoS One 9:e87870.

11. Clark R, Lewinski M, Loeffelholz M, Tibbetts R. 1997. Cumitech 31A, Verification and Validation of
 ACCEPTED MANUSCRIPT

Procedures in the Clinical Microbiology Laboratory. ASM Press, Washington, DC.

12. CLSI. 2015. Verification of Commercial Microbial Identification and Antimicrobial Susceptibility

Testing Systems,1st ed. CLSI guideline M52. Wayne, PA: Clinical and Laboratory Standard

Institute.

T
13. Chen Y, Porter V, Mubareka S, Kotowich L, Simor AE. 2015. Rapid Identification of Bacteria

IP
Directly from Positive Blood Cultures by Use of a Serum Separator Tube, Smudge Plate

CR
Preparation, and Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry. J

Clin Microbiol 53:3349–52.

US
14. Idelevich EA, Schüle I, Grünastel B, Wüllenweber J, Peters G, Becker K. 2014. Rapid identification
AN
of microorganisms from positive blood cultures by MALDI-TOF mass spectrometry subsequent to

very short-term incubation on solid medium. Clin Microbiol Infect 20:1001–1006.

M

15. Stevenson LG, Drake SK, Murray PR. 2010. Rapid identification of bacteria in positive blood
ED

culture broths by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J

PT

Clin Microbiol 48:444–7.

CE
AC
 ACCEPTED MANUSCRIPT

TABLE 1 MALDI-TOF MS identification by routine and direct pelleting method.

Routine
Direct identification
identification
Organism Total Biotyper score
Biotyper score
≥2.0 1.7 - 2.0 ≥2.0 1.7 - 2.0 <1.7
Gram-postive bacteria
Staphylococci
S. aureus 13 13 0 13 0 0
Coagulase-negative
23 20 3 16 7 0

T
staphylococci
S. auricularis 2 1 1 0 2 0

IP
S. epidermidis 13 11 2 9 4 0
S. hominis 6 6 0 6 0 0

CR
S. lugdunesis 1 1 0 1 0 0
S. warneri 1 1 0 0 1 0
Streptococci

US
S. agalactiae 1 1 0 1 0 0
S. pneumoniae 3 3 0 0 0 3
S. gordonii 1 0 1 0 1 0
AN
S. milleri group1 2 2 0 1 0 1
2
AHS 2 1 1 1 0 1
Other
M

E. faecalis 4 4 0 4 0 0
E. faecium 1 1 0 1 0 0
ED

M. luteus 4 3 1 4 0 0
3
ASB 4 3 1 2 1 1
L. monocytogenes 1 1 0 1 0 0
PT

4
Corynebacterium spp. 2 1 1 0 2 0
B. longum 1 0 1 0 1 0
P. acnes 1 1 0 1 0 0
CE

A. odontolyticus 1 0 1 1 0 0
Gram positive total 64 54 10 46 12 6
AC

Gram-negative bacteria
Enterobacteriaceae
E. coli 36 36 0 34 2 0
K. pneumoniae 8 8 0 8 0 0
Enterobacter spp5 3 3 0 3 0 0
S. marcescens 1 1 0 1 0 0
C. koseri 1 1 0 1 0 0
P. stuartii 1 1 0 1 0 0
Salmonella spp 1 1 0 1 0 0
Other
P. aeruginosa 2 2 0 1 0 1
 ACCEPTED MANUSCRIPT

P. multocida 1 1 0 1 0 0
M. catarrhalis 1 1 0 1 0 0
B. fragilis 1 1 0 1 0 0
P. nigrescens 1 0 1 0 0 1
Gram negative total 57 56 1 53 2 2

Yeast 4 4 0 4 0 0
Total 125 114 11 103 14 8
1
S. intermedius, S. anginosus; 2S. sanguinis, S. mitis; 3B. cereusx2, B. pumilus, P.

T
odorifir; 4C. afermentens, C. Striatum; 5E. cloacae complex, E. aerogenes x2

IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

TABLE 2 Identification results from positive blood cultures by MALDI-TOF MS

with Biotyper score ≥2.0001.
Routine method (n, %) Direct method (n, %)
Total (n=125) 114 (91.2) 103 (82.4)
Gram-positive (n=64) 54 (84.4) 46 (71.9)
Gram-negative (n=57) 56 (98.2) 53 (93.0)
Candida (n=4) 4 (100) 4 (100)

Adjusted identification to

T
species level2

IP
Total (n=114) - 103 (90.4)
Gram-positive (n=54) - 46 (85.2)

CR
Gram-negative (n=56) - 53 (94.6)
Candida (n=4) - 4 (100)
1
Species level identification; 2Only isolates with a Biotyper score ≥2.000 on

US
routine identification included
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

TABLE 3a Automated antibiotic susceptibility test results for Gram-negative organisms

compared using both pellet and routine methods.

Antibiotic Comparisons EA%1 CA%2 (EA+CA)% ME%3 MinE%4 (ME+MinE)%

Ampicillin 95 100.0 100.0 100.0 0.0 0.0 0.0

Amox/Clav 95 97.9 87.4 92.6 0.0 12.6 12.6
Pip/Tazo 89 97.8 96.6 97.2 0.0 3.2 3.2
Cefalotin 72 97.2 97.2 97.2 1.0 1.1 2.1
Cefazolin 95 98.9 97.9 98.4 0.7 1.1 1.8

T
Cefoxitin 95 100.0 96.8 98.4 0.0 3.2 3.2

IP
Cefixime 95 100.0 96.8 98.4 0.0 3.2 3.2
Cefpodoxime 95 98.9 100.0 99.5 0.0 0.0 0.0
Cefotaxime 95 96.8 95.8 96.3 0.0 4.2 4.2

CR
Ceftazidime 95 95.8 98.9 97.4 0.0 1.1 1.1
Imipenem 95 93.7 95.8 94.7 5.0 4.2 9.2
Amikacin 95 97.9 100.0 98.9 0.0 0.0 0.0

US
Gentamicin 95 98.9 97.9 98.4 0.0 2.1 2.1
Tobramycin 95 100.0 98.9 99.5 0.0 1.1 1.1
Ciprofloxacin 95 98.9 98.9 98.9 1.3 1.1 2.3
AN
Tetracycline 95 97.9 97.9 97.9 1.8 0.0 1.8
Nitrofurantoin 95 97.9 87.4 92.6 0.0 12.6 12.6
Cotrimoxazole 95 97.9 98.9 98.4 0.8 0.0 0.8
M

1
Essential Agreement; 2Categorical Agreement; 3Major Error; 4Minor Error
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

TABLE 3b Automated antibiotic susceptibility test results for Gram-positive organisms compared using both pellet and
routine methods.
Antibiotic Comparisons EA%1 CA%2 (EA+CA)% VME%3 ME%4 MinE%5 (VME+ME+MinE)%
Benzylpenicillin 92 100.0 100.0 100.0 0.0 0.0 0.0 0.0
Ampicillin 26 100.0 100.0 100.0 0.0 0.0 0.0 0.0
Oxacillin 66 83.3 84.8 84.1 17.2 0.0 0.0 17.2
Gentamicin 66 98.5 98.5 98.5 0.0 0.0 1.1 1.1
Ciprofloxacin 92 100.0 98.9 99.5 0.0 0.0 1.1 1.1

T
Levofloxacin 92 98.9 100.0 99.5 0.0 0.0 0.0 0.0
Moxifloxacin 92 100.0 98.9 99.5 0.0 0.0 1.1 1.1

IP
Erythromycin 92 94.6 93.5 94.0 0.0 0.0 6.5 6.5
Clindamycin 92 100.0 100.0 100.0 0.0 0.0 0.0 0.0

CR
Synercid 92 100.0 100.0 100.0 0.0 0.0 0.0 0.0
Linezolid 92 100.0 100.0 100.0 0.0 0.0 0.0 0.0
Vancomycin 92 100.0 100.0 100.0 0.0 0.0 0.0 0.0

US
Tetracycline 92 100.0 100.0 100.0 0.0 0.0 0.0 0.0
Tigecycline 92 100.0 100.0 100.0 0.0 0.0 0.0 0.0
Nitrofurantoin 92 100.0 97.8 98.9 0.0 0.0 2.2 2.2
AN
Rifampicin 66 100.0 100.0 100.0 0.0 0.0 0.0 0.0
Cotrimoxazole 66 98.5 100.0 99.2 0.0 0.0 0.0 0.0
1
Essential Agreement; 2Categorical Agreement; 3Very Major Error; 4Major Error; 5Minor Error
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

TABLE 4 Time from blood culture bottle positivity to MALDI identification and Vitek 2 susceptibility
test result.
Sub-culture Pellet Sub-culture Pellet
Organism identification identification susceptibility susceptibility
(mean, SD) (mean, SD) (mean, SD) (mean, SD)
Gram Negative 26.1 hrs (1.61) 1.0 hrs (0.24) 38.6 hrs (3.32) 12.0 hrs (1.21)
Gram Positive 29.9 hrs (1.75) 0.9 hrs (0.23) 44.5 hrs (3.77) 11.2 hrs (1.23)

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

TABLE 5 Performance characteristics of rapid ESBL (β LACTA) and MRSA (Denka) screening from
bacterial pellet.
True True False False Sensitivity Specificity
Positive Negative Positive Negative (%) (%)
MRSA
(Denka) 36 9 1 0 94.7 90
ESBL (β
LACTA) 65 13 0 7 90.3 100

T
IP
CR
US
AN
M
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Highlights

 Positive blood cultures underwent a lysis-centrifugation protocol to pellet

microorganisms for subsequent MALDI-TOF identification and antimicrobial
susceptibility testing
 Compared to MALDI-TOF identification from subcultured growth, a correct genus level
identification occurred in 91.2 % of isolates, while the special level concordance was
82.4%
 There was a >98% agreement for Vitek2 susceptibility results between subcultured and

T
pelleted organisms

IP
 Rapid MRSA and ESBL screening from the pellet demonstrated a >90% sensitivity and
specificity

CR
US
AN
M
ED
PT
CE
AC