M E D I C A L M I C R O B I O L O G Y I N V I T E D A R T I C L E

Melvin P. Weinstein and L. Barth Reller, Section Editors

Matrix-Assisted Laser Desorption Ionization
Time of Flight Mass Spectrometry in Clinical
Microbiology
Robin Patel
Division of Clinical Microbiology, Department of Laboratory Medicine and Pathology and Division of Infectious Diseases, Department of Medicine, Mayo
Clinic, Rochester, Minnesota
Despite widespread application of nucleic acid diagnostics, cultures remain integral in modern laboratories.
Because cultures detect a large number of organism types, it is unlikely that they will disappear from clinical
practice in the near future. Their downside is slow turn-around time, impacted by time to growth and identi-
cation of that growth. The latter is expedited using a new proteomic technology, matrix-assisted laser desorp-
tion ionizationtime of ight mass spectrometry (MALDI-TOF MS).
Keywords. mass spectrometry; bacteria; fungi; MALDI-TOF; matrix-assisted laser desorption ionization-time
of ight.
Despite widespread application of nucleic acid diagnos-
tics, cultures remain integral in modern laboratories.
Because cultures detect a large number of organism
types, it is unlikely that they will disappear from clinical
practice in the near future. Their downside is slow
turn-around time, impacted by time to growth and
identication of that growth. The latter is expedited
using a new proteomic technology, matrix-assisted
laser desorption ionizationtime of ight mass spec-
trometry (MALDI-TOF MS).
Historically, identication of cultured bacteria and
fungi has been a complex, algorithmic task. Textbooks,
owcharts, and tables have been used to interpret
colony morphology as well as stains and biochemical
tests performed on colonies. Phenotypic tests may be
contingent on subsequent growth, prolonging time to
identication. Manual biochemical tests (eg, coagulase,
catalase) are rapid, but identify limited organism types
(eg, Staphylococcus aureus). Biochemical test panels
performed manually (eg, API strips, bioMrieux, Dur-
ham, North Carolina) or on automated instruments
(eg, VITEK, bioMrieux; BD Phoenix Automated Mi-
crobiology System, BD, Franklin Lakes, New Jersey; Mi-
croScan, Siemens Healthcare Diagnostics, Tarrytown,
New York) identify more microorganisms, but have
longer turn-around times and costly consumables, and
often require knowledge of the organism type being
tested (eg, gram-positive cocci). Finally, broad-range
PCR followed by sequencing accurately identies bacte-
ria and fungi but has a long turn-around time, and is
expensive and performed in select laboratories. With
MALDI-TOF MS, colonies growing in culture are inex-
pensively and accurately identied in minutes, without
a priori knowledge of microorganism type; users do
not even need to know whether a bacterium or a yeast
is being tested. The product of advances in mass spec-
trometry and bioinformatics, MALDI-TOF MS is revo-
lutionizing bacterial and fungal identication in clinical
microbiology [1].
The concept of using mass spectrometry to identify
bacteria (albeit not based on proteomics) was proposed
in the 1970s [2]; with the Nobel Prizewinning invention
of MALDI-TOF MS in the 1980s [3, 4], whole-organism
Received 29 January 2013; accepted 5 April 2013; electronically published 17
April 2013.
Correspondence: Robin Patel, MD, Mayo Clinic, Division of Clinical Microbiology,
200 First St SW, Rochester, MN 55905 ( patel.robin@mayo.edu).
Clinical Infectious Diseases 2013;57(4):56472
The Author 2013. Published by Oxford University Press on behalf of the Infectious
Diseases Society of America. All rights reserved. For Permissions, please e-mail:
journals.permissions@oup.com.
DOI: 10.1093/cid/cit247
564 CID 2013:57 (15 August) MEDICAL MICROBIOLOGY

b
y

g
u
e
s
t

o
n

A
u
g
u
s
t

2
3
,

2
0
1
4
h
t
t
p
:
/
/
c
i
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

proteomics became possible. It took until the 1990s for develop-
ment of libraries of reference spectra and software for bacterial
identication, and until the past few years for commercialization
of the technology [1, 5]. MALDI-TOF MS, used for a few years
in European clinical microbiology laboratories, and being rap-
idly adopted worldwide, is likely to antiquate most biochemical
microbial identication in the near future.
The simplest MALDI-TOF MS applications test bacterial or
yeast colonies without elaborate preparation and using minimal
consumables, providing rapid and low-cost identication.
Testing begins by picking an unknown colony from a culture
plate to a spot on a MALDI-TOF MS target plate (Figure 1).
The cells may be treated on the target plate (we use formic
acid) and are ultimately overlain with 12 L of matrix. Follow-
ing a short drying period, the plate is placed in the ionization
chamber of the mass spectrometer (Figure 2).
MALDI refers to the matrix, which assists in the desorp-
tion and ionization of microbial analytes through pulses of
energy from an ultraviolet N
2
laser. The matrix (eg, -cyano-4-
hydroxycinnamic acid dissolved in 50% acetonitrile and 2.5%
triuoroacetic acid) isolates microbial molecules from each
other; microbial and matrix molecules on the target plate are
desorbed, with the majority of energy being absorbed by the
matrix, converting it to an ionized state. Through random colli-
sion in the gas phase, charge is transferred from matrix to mi-
crobial molecules; analytes within the mass range specied are
measured, the most abundant being ribosomal proteins. The
cloud of ionized microbial proteins is funneled through a posi-
tively charged electrostatic eld into a time of ight or TOF
mass analyzer, a tube under vacuum. The ions travel toward an
ion detector with small analytes traveling fastest, followed by
progressively larger analytes. A mass spectrum is generated,
representing the number of ions of a given mass impacting the
ion detector over time.
Analytes are not individually characterized; instead, the gen-
erated mass spectrum provides a prole (ngerprint) of the
unknown microorganism. Generated mass spectra are unique
to individual microorganism types, with peaks specic to
genera, species, and even strains. The test isolates mass spec-
trum is automatically compared to a database of reference
Figure 1. Matrix-assisted laser desorption ionizationtime of ight mass spectrometry (MALDI-TOF MS) workow. A bacterial or fungal colony (typically
single) is picked from a culture plate to a spot on a MALDI-TOF MS target plate (a reusable or disposable plate with a number of test spots) using a
wooden or plastic stick, pipette tip, or loop. One or many isolates may be tested at one time. At our institution, cells are treated with formic acid on the target
plate. The spot is then overlain with 12 L of matrix. Following a short drying period, the plate is placed in the ionization chamber of the mass spectrometer
for analysis (see Figure 2). A mass spectrum is generated and automatically compared against a database of mass spectra by the software, resulting in identi-
cation of the organism (Candida parapsilosis in position A4 in the example). More complex preparatory protein extraction, commonly used in older studies, may
be performed prior to target plate inoculation. Although this yields high-quality mass spectra [46], it is technically cumbersome; direct on-plate testing using
formic acid identies a similar number of organisms [37]. Direct on-plate testing must be avoided with organisms hazardous to microbiology technologists (eg,
Brucella species) [28]. Reproduced by permission of Mayo Foundation for Medical Education and Research. All rights reserved.
MEDICAL MICROBIOLOGY CID 2013:57 (15 August) 565

b
y

g
u
e
s
t

o
n

A
u
g
u
s
t

2
3
,

2
0
1
4
h
t
t
p
:
/
/
c
i
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

spectra to determine the relatedness of the spectrum to spectra
in the database; a list of most closely related organisms is gener-
ated, each with a numeric ranking (percentage or score) indicat-
ing the level of condence in identication. Depending on how
high the value is, the organism is identied at the family, genus,
or species level. As with any identication system, well-curated
spectral databases representing comprehensive collections of
correctly named and clinically relevant bacteria and fungi are
necessary. In the absence of mass spectral entries in a database,
entries may be manually added (assuming availability of requi-
site isolates). The time from putting the target plate into the in-
strument to nal result is fast; for example, the Bruker system
produces results in 3 minutes per isolate. No mass spectrome-
try expertise is required, and the technology is green with re-
usable target plates available and few disposables.
Commercial MALDI-TOF MS systems are available from
Bruker Daltonics Inc (Billerica, Massachusetts) and bioMrieux
Inc. A system used primarily in France (Andromas, Paris,
France), will not be further discussed. The Bruker Biotyper
system is most well studied and includes a mass spectrometer,
software, and library; over the past few years, improved software
and progressively more comprehensive and better curated mass
spectral libraries have been made available. AnagnosTec
(Zossen, Germany) previously marketed a microbial database
called SARAMIS used with Shimadzus AXIMA Assurance
mass spectrometer (Shimadzu, Columbia, Maryland), which
was acquired by bioMrieux in 2010; with this acquisition, the
systems name was changed to VITEK MS RUO. BioMrieux
developed its own database/operating software/algorithms, re-
ferred to as VITEK MS (including a prerelease version of the
VITEK MS v1.1 database [6], the v1 system/v1.1 database [710],
the v2.0 system/v2.0 database and algorithms, and VITEK MS
Plus, which incorporates the VITEK MS v2.0 and SARAMIS
v4.0 databases). Rapidly implemented progressive improve-
ments in both systems are enhancing the clinical laboratorys
ability to identify a range of organisms but render it challenging
for those reading the literature to compare studies due to vari-
able software, interpretive rules, databases, mass spectrometers,
comparators, specimen preparations, organism sets (ie, en-
riched for unusual organisms, or not), and so forth. Current
systems databases are generally best for common aerobic
gram-negative bacilli and gram-positive cocci, and may lack
some esoteric organisms, such as certain anaerobes, fungi, and
mycobacteria. Given changing nomenclature and ongoing de-
scription of new species and emerging microorganisms, regular
and ongoing updates to databases are imperative to ensure clin-
ical utility. Databases should include entries representing major
phylogenetic lineages within each species. It is possible for
users to add their own mass spectral entries to enhance existing
databases or create their own database by including locally im-
portant strains or strains not well represented in commercial
libraries.
The Bruker and bioMrieux MALDI-TOF MS systems differ
in databases, algorithms used to identify organisms, and instru-
mentation. Numeric rankings are reported on different scales,
and are not directly comparable. Brukers Microex LT mass
spectrometer is a desktop instrument, whereas bioMrieuxs is
a larger, oor model. At this time, neither system is approved
by the US Food and Drug Administration.
Figure 2. Matrix-assisted laser desorption ionizationtime of ight
mass spectrometer. The target plate is placed into the ionization chamber
of the mass spectrometer. Spots to be analyzed are shot by an ultraviolet
N
2
laser desorbing microbial and matrix molecules from the target plate.
The majority of energy is absorbed by the matrix, converting it to an
ionized state. Through random collision in the gas phase, charge is trans-
ferred from matrix to microbial molecules. The cloud of ionized molecules
is funneled through a positively charged electrostatic eld into the time of
ight mass analyzer, a tube under vacuum. The ions travel toward an ion
detector with small analytes traveling fastest, followed by progressively
larger analytes. As ions emerge from the mass analyzer, they collide with
an ion detector generating a mass spectrum representing the number of
ions hitting the detector over time. Although separation is by mass-
to-charge ratio, because the charge is typically single for the described ap-
plication, separation is effectively by molecular weight. Reproduced by
permission of Mayo Foundation for Medical Education and Research. All
rights reserved.
566 CID 2013:57 (15 August) MEDICAL MICROBIOLOGY

b
y

g
u
e
s
t

o
n

A
u
g
u
s
t

2
3
,

2
0
1
4
h
t
t
p
:
/
/
c
i
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Generally speaking, MALDI-TOF MS performs at least as
well as, if not better than, automated biochemical identication
for commonly encountered bacteria and yeast (Table 1), and
better for unusual organisms. We compared the BD Phoenix
Automated Microbiology System and Biotyper system using
440 common and unusual gram-negative bacilli; the former
correctly identied 83% and 75% of isolates to the genus and
species levels, respectively, whereas MALDI-TOF MS correctly
identied 93% and 82% of isolates to the genus and species
levels, respectively [11]. We also evaluated the Biotyper system
using 217 clinical isolates of staphylococci, streptococci, and
enterococci, 98% and 79% of which were correctly identied to
the genus and species levels, respectively [12]. The performance
of the 2 commercial systems appears to be similar. Martiny et al
reported 93% species-level identications of 1003 routine bac-
terial isolates with both a prerelease version of the VITEK MS
v1.1 database and the Biotyper system [6].
In most cases, organisms are either correctly identied or
yield a low score/percentage, indicating that identication has
not been achieved; the latter typically implies no match in the
database but can occur due to technically poor preparation.
Misidentications are unusual but may occur with some closely
related organisms. Although a comprehensive review of identi-
cation of specic organisms is beyond the scope of this article,
some generalizations can be made. Identication of staphylo-
cocci [13] and -hemolytic streptococci is excellent [14]. The
-hemolytic streptococci are more problematic because Strepto-
coccus mitis and Streptococcus oralis may be poorly differentiat-
ed from one another and from Streptococcus pneumoniae, at
least with the Bruker system [12, 15]; recent data suggest that
VITEK MS may overcome this limitation [8]. Arcanobacterium
haemolyticum, Rhodococcus equi, and all but select closely
related Corynebacterium species are reliably identied [9, 16].
Some closely related organisms, such as Escherichia coli and
Shigella species, are not differentiated. (To avoid potential mis-
identication, we attempt to avoid testing of suspect E. coli and
we perform additional testing if E. coli is suggested.) MALDI-
TOF MS accurately detects Enterobacter cloacae complex,
although it may not discriminate some species within this
complex [9, 17]. Likewise, accurate differentiation of all species
of Acinetobacter may be challenging with current databases, al-
though Acinetobacter baumannii is identiable [18]. Similar
challenges exist with other identication systems. MALDI-TOF
MS is exciting because of the breadth and depth of organism
types that can be identied with one platform, including, but
not limited to, Aeromonas species [19], Campylobacter species,
Arcobacter butzleri, and Yersinia enterocolitica [20]. MALDI-
TOF MS can identify and differentiate Capnocytophaga cani-
morsus from Capnocytophaga cynodegmi [21]. Although typing
within the genus Salmonella is not generally possible, it may be
possible to distinguish Salmonella enterica subspecies enterica
serovar Typhi from other S. enterica serovars [22]. And with
user-supplemented databases, HACEK organisms [23], Legion-
ella species [24], and Nocardia species [25] (the last with specif-
ic extraction procedures) may be identied.
Jamal et al reported species-level identication of 89% and
100% of 274 routinely isolated anaerobic bacteria (enriched in
Bacteroides fragilis) using the Biotyper DB Update v3.3 and the
VITEK MS v1 system/v1.1 database, respectively [10]. Identi-
cation of more esoteric anaerobes, including Prevotella species,
has been successful in 83% of cases with user supplementation
of Brukers Reference Library 3.2.1.0 [26]. We evaluated a
diverse collection of 253 clinical anaerobic isolates using the Bi-
otyper system and a user-supplemented database; 92% and
71% of isolates were correctly identied to the genus and
species levels, respectively [27]. Further commercial database
Table 1. Examples of Recent Evaluations of Matrix-Assisted Laser Desorption IonizationTime of Flight Mass Spectrometry for Routine
Bacterial Identication
Isolates % Identification
System No. Type
Period of
Isolate Collection Genus Level Species Level Country Comparator Reference
Bruker Biotyper 1013 Bacteria 2 mo 99% 97% France Phoenix, API, Biochemical [72]
Bruker Biotyper 468 Bacteria 3 mo 97% 92% Japan MicroScan, API, Phoenix [73]
Bruker Biotyper 2781 Bacteria 1 mo 96% 85% Australia VITEK2, API, Biochemical [74]
Vitek MS
a
767 Bacteria 6 wk 95% 87% France VITEK2 [8]
Bruker Biotyper/
Vitek MSb
986 Bacteria 3 mo 96%/94% 93%/93% Belgium Bruker Biotyper
compared to
Vitek MS
[6]
a
v1 system/v1.1 database.
b
Prerelease version of v1.1 database.
MEDICAL MICROBIOLOGY CID 2013:57 (15 August) 567

b
y

g
u
e
s
t

o
n

A
u
g
u
s
t

2
3
,

2
0
1
4
h
t
t
p
:
/
/
c
i
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

supplementation will improve anaerobic identication to a
level previously possible only with 16S ribosomal RNA gene se-
quencing or gas liquid chromatography.
MALDI-TOF MS can accurately identify Francisella tularen-
sis and Brucella species; however, the Biotyper library does not
contain and will not identify these organisms. Use of Brukers
Security Relevant database enables Brucella species and F.
tularensis identication [28].
A handful of investigators have reported using MALDI-TOF
MS for mycobacteria. This requires special processing to kill
tested organisms (for safety), disrupt clumped cells, and break
down cell envelopes. Current commercial libraries inadequately
address the clinically relevant Mycobacterium species, but with
appropriate library construction, MALDI-TOF MS should be
able to identify most, with a few caveats expected. Members of
the Mycobacterium tuberculosis complex are identied at the
complex level [29]; some species (eg, Mycobacterium abscessus/
massiliense; Mycobacterium mucogenicum/phocaicum; Myco-
bacterium chimaera/intracellulare) may not be well differentiat-
ed from one another [29], and identication from liquid media
may be more challenging than from colonies on solid media [30].
Beyond bacteria, MALDI-TOF MS can identify yeast, out-
performing some conventional phenotypic systems, and distin-
guishing Candida dubliniensis and albicans; Candida rugosa
and pararugosa; Candida norvegensis, krusei, and inconspicua;
Candida parapsilosis, orthopsilosis, and metapsilosis [31]; and
(with database supplementation) Cryptococcus neoformans and
gattii [3234]. Dhiman et al evaluated the Biotyper system for
identication of 138 common and 103 unusual yeast isolates,
and reported 96% and 85% accurate species-level identication,
respectively [35]. MALDI-TOF MS may outperform current
systems to identify esoteric and sometimes misidentied Candida
species, such as Candida famata [31]. Although older studies
used preparatory extraction for yeasts, we and others use the
same direct testing strategy we use for bacteria (Figure 1) [36, 37].
Few lamentous fungi are included in current commercial
databases. Filamentous fungi exhibit variable phenotypes;
protein spectra may vary with growth conditions or with the
zone of fungal mycelium analyzed. De Carolis et al developed a
library of Aspergillus species, Fusarium species, and Mucorales
using the Biotyper system and identied 97% of 94 isolates to
the species level [38]. Using the VITEK MS v1 system/v1.1 da-
tabase and direct on-plate testing, Iriart et al identied 82% of
44 Aspergillus isolates (including all isolates with species in the
database) [7]. Theel et al developed a dermatophyte library and
used it to identify 93% and 60% of 171 dermatophyte isolates
to the genus and species levels, respectively, using the Biotyper
system [39]. Preliminary studies indicate that Pseudallescheria/
Scedosporium complex species are identiable [40]. Beyond
fungi, MALDI-TOF MS has been used for Prototheca species
identication [41].
Compared to standard methods, MALDI-TOF MS improves
turn-around time for bacterial and fungal identication by an
average of 1.45 days [42], and, because only a small amount of
organism is required, testing can be performed directly from
single colonies on primary culture plates, whereas other
methods may require subculture. Tests for screening for some
enteric pathogens (eg, triple sugar iron agar for Salmonella
species) may be eliminated [43].
MALDI-TOF MS reagent costs are approximately $0.35/test;
implementation reduces reagent and labor costs for identica-
tion, in one model, by 57% within 1 year [42]. An estimated
87% of isolates are identied on the rst day (compared with
9% with standard techniques), with nal identications a day
earlier for most organisms, and several days earlier for bio-
chemically inert, fastidious, or slow-growing organisms [42].
DNA sequencing expenses are avoided for some esoteric organ-
isms, waste disposal is decreased, and quality control and labo-
ratory technologist training/labor for replaced/retired tests are
eliminated [44].
Adoption of MALDI-TOF MS impacts laboratory workow [9].
Time at the bench assessing primary cultures is relatively unaf-
fected, with subcultures needed for susceptibility testing. Gram
staining of colonies may not always be needed, as this informa-
tion is not required for MALDI-TOF MS. For some organisms
(eg, S. aureus), rapid tests performed at the bench will remain.
MALDI-TOF MS has limitations (Table 2). Unlike publically
available sequence databases such as GenBank, MALDI-TOF
MS databases are proprietary. Although low identication per-
centages for some organisms may be improved by user addition
of mass spectral entries of underrepresented species or strains
(to cover intraspecies variability) to the database, or even read-
dition of reference strain spectra [26], doing so may be beyond
the capabilities of some laboratories. Because of low scores/per-
centages, repeat testing is required for approximately 10% of
isolates [42]. Growth on some media (eg, colistin-nalidixic acid
agar) may be associated with low scores/percentages [45], and
tiny or mucoid colonies may fail identication [9, 46]. Rened
criteria are needed to distinguish closely related species and to
differentiate them from the next best taxon match [47]. For
certain species of organisms, genus- or species-specic (includ-
ing lowered) cutoffs may be appropriate [13, 16, 48]. Laboratory
errors may occur, including colony inoculation in erroneous
target plate locations, testing impure colonies, smearing be-
tween spots, failure to clean target plates, and erroneous result
entry into laboratory information systems. There is a learning
curve to applying ideal colony amounts to target plates [9]. In-
strument cost is comparable to that of other common laborato-
ry equipment; laser, software, hardware, or mass spectrometer
failure require an appropriate service plan. Although MALDI-
TOF MS is generally reproducible, sources of variability in-
clude the mass spectrometer, matrix and solvent composition,
568 CID 2013:57 (15 August) MEDICAL MICROBIOLOGY

b
y

g
u
e
s
t

o
n

A
u
g
u
s
t

2
3
,

2
0
1
4
h
t
t
p
:
/
/
c
i
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

technologists, culture conditions, and biological variability;
quality control strategies are being dened. Notably, MALDI-
TOF MS may yield different (generally more accurate) identi-
cation than current systems [11, 49].
Antimicrobial susceptibility is not directly determined with
the above-described strategy. By rapid organism identication,
intrinsic antimicrobial resistance characteristics of particular
species (or typical susceptibility of the identied species based
on local antibiograms) may guide therapy. Some investigators
have incubated bacteria with antibiotics and measured antibiot-
ic degradation using MALDI-TOF MS; this requires specic
system conguration and incubation (ie, to allow antibiotic hy-
drolysis) and applies only to resistance mechanisms associated
with antibiotic degradation (eg, -lactamases) [50]. Although
detection of methicillin resistance in S. aureus has been contro-
versial, this does not appear to be feasible using the Biotyper
software in detection ranges and instrument settings preset for
microbial identication [51]. Microbial strain typing is under
development with no current universal protocol; innovations
and advances are expected. With strain typing [52, 53], strains of
S. aureus likely to be methicillin resistant (eg, USA300 [54]), en-
terococci likely to be vancomycin resistant [55], or Bacteroides
fragilis likely to be carbapenem resistant (cA positive) [56]
may be called out (as markers for possible resistance).
Although direct testing of clinical specimens is not generally
feasible because of a high limit of detection, urine may be
tested directly due to the high numbers of bacteria present in
clinically infected urine (although -defensins 1, 2, and 3 may
impair database matching [57]).
Rapid identication of microorganisms growing in blood
culture bottles is promising, but requires preparatory specimen
processing as blood culture bottles contain macromolecules
from blood and growth media. Processing is accomplished with
differential centrifugation and washings, selective lysis of blood
cells (eg, using water, ammonium chloride, saponin, sodium
dodecyl sulfate), separator gels, ltration, and so on; commer-
cial blood culture bottle processing using the Sepsityper
(Bruker Daltonics) is available [5864]. Although results are
valid when obtained, the yield is generally not as good as direct
colony testing with a higher percentage of gram-negative than
gram-positive organisms being identied [58]; polymicrobial
cultures may not be fully characterized (although strategies are
being developed). Most Candida species are identiable [65].
An alternate approach is to subculture positive blood culture
bottles to solid media and test growth after a short period (eg,
24 hours) of incubation [66, 67]. By testing of positive blood
culture bottles using MALDI-TOF MS, the time to organism
identication may be reduced by a day or more, which may
serve as a guide for empirical treatment [68], but antimicrobial
susceptibility is not provided. Recently, MALDI-TOF MS has
been tied with direct antimicrobial susceptibility testing from
positive blood culture bottles [69, 70], improving time to
optimal therapy and decreasing hospital length of stay and
costs [71].
In summary, MALDI-TOF MS is a quantum leap forward
for rapid and accurate identication of cultured bacteria and
fungi in clinical microbiology. This is incontrovertibly bene-
cial to laboratories, and decreased expense is helpful to labora-
tories and patients alike. MALDI-TOF MS implementation
may impact patient outcomes and/or cost of care, depending
on its application. Further, this new technology may provide
better denition of the epidemiology, pathogenesis, and antimi-
crobial susceptibility of hitherto unrecognized (or misidenti-
ed) microorganisms.
Notes
Acknowledgments. I thank Dr Nancy L. Wengenack, Scott A. Cun-
ningham, Brenda L. Dylla, and Sherry M. Ihde for their thoughtful reviews
of this manuscript.
Potential conicts of interest. Author certies no potential conicts of
interest.
The author has submitted the ICMJE Form for Disclosure of Potential
Conicts of Interest. Conicts that the editors consider relevant to the
content of the manuscript have been disclosed.
Table 2. Advantages and Limitations of Matrix-Assisted Laser
Desorption IonizationTime of Flight Mass Spectrometry in Clini-
cal Microbiology
Advantages Limitations
Rapid turn-around time Does not provide antimicrobial
susceptibility
Single colony requirement Requires isolated colony
Automated, high
throughput
Repeat testing occasionally
required
Broad applicability to
bacteria and fungi,
including esoteric
organisms
Not cleared by the US Food and
Drug Administration
Databases may be
expanded by users
Identification limited by entries in
database; databases vary, not
publicly available
Minimal consumables Laser, hardware, software, mass
spectrometer failure potential
testing downtime
Cost-effective Instrument cost
Generalizes to bacteria and
yeast
Misidentifications: failure to test
pure colonies or to clean target
plates
Some closely related organisms
not differentiated (eg,
Escherichia coli and Shigella
species)
Growth on some media
associated with no identification
(without extraction)
Tiny or mucoid colonies may not
be identified
MEDICAL MICROBIOLOGY CID 2013:57 (15 August) 569

b
y

g
u
e
s
t

o
n

A
u
g
u
s
t

2
3
,

2
0
1
4
h
t
t
p
:
/
/
c
i
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

References
1. Seng P, Drancourt M, Gouriet F, et al. Ongoing revolution in bacteriol-
ogy: routine identication of bacteria by matrix-assisted laser desorp-
tion ionization time-of-ight mass spectrometry. Clin Infect Dis 2009;
49:54351.
2. Anhalt JP, Fenselau C. Identication of bacteria using mass spectrome-
try. Anal Chem 1975; 47:21925.
3. Karas M, Hillenkamp F. Laser desorption ionization of proteins with
molecular masses exceeding 10,000 daltons. Anal Chem 1988;
60:2299301.
4. Tanaka K, Hiroaki W, Ido Y, Akita S, Yoshida Y, Yoshida T. Protein
and polymer analyses up to m/z 100 000 by laser ionization time-
of-ight mass spectrometry. Rapid Commun Mass Spectrom 1988;
2:1513.
5. Mellmann A, Cloud J, Maier T, et al. Evaluation of matrix-assisted laser
desorption ionization-time-of-ight mass spectrometry in comparison
to 16S rRNA gene sequencing for species identication of nonferment-
ing bacteria. J Clin Microbiol 2008; 46:194654.
6. Martiny D, Busson L, Wybo I, El Haj RA, Dediste A, Vandenberg O.
Comparison of the Microex LT and Vitek MS systems for routine
identication of bacteria by matrix-assisted laser desorption ionization-
time of ight mass spectrometry. J Clin Microbiol 2012; 50:131325.
7. Iriart X, Lavergne RA, Fillaux J, et al. Routine identication of medical
fungi by the new Vitek MS matrix-assisted laser desorption ionization-
time of ight system with a new time-effective strategy. J Clin Microbiol
2012; 50:210710.
8. Dubois D, Grare M, Prere MF, Segonds C, Marty N, Oswald E. Perfor-
mances of the Vitek MS matrix-assisted laser desorption ionization-time
of ight mass spectrometry system for rapid identication of bacteria in
routine clinical microbiology. J Clin Microbiol 2012; 50:256876.
9. Harris P, Winney I, Ashhurst-Smith C, OBrien M, Graves S. Compari-
son of Vitek MS (MALDI-TOF) to standard routine identication
methods: an advance but no panacea. Pathology 2012; 44:5835.
10. Jamal WY, Shahin M, Rotimi VO. Comparison of two matrix-assisted
laser desorption ionization-time of ight (MALDI-TOF) mass spec-
trometry methods and API 20AN for identication of clinically rele-
vant anaerobic bacteria. J Med Microbiol 2013; 62:5404.
11. Saffert RT, Cunningham SA, Ihde SM, Jobe KE, Mandrekar J, Patel R.
Comparison of Bruker Biotyper matrix-assisted laser desorption ioniza-
tion-time of ight mass spectrometer to BD Phoenix automated micro-
biology system for identication of gram-negative bacilli. J Clin
Microbiol 2011; 49:88792.
12. Alatoom AA, Cunningham SA, Ihde SM, Mandrekar J, Patel R. Com-
parison of direct colony method versus extraction method for identi-
cation of gram-positive cocci by use of Bruker Biotyper matrix-assisted
laser desorption ionization-time of ight mass spectrometry. J Clin Mi-
crobiol 2011; 49:286873.
13. Richter C, Hollstein S, Woloszyn J, Kaase M, Gatermann SG, Szabados
F. Evaluation of species-specic score cut-off values for various Staphy-
lococcus species using a MALDI Biotyper-based identication. J Med
Microbiol 2012; 61:140916.
14. Cherkaoui A, Emonet S, Fernandez J, Schorderet D, Schrenzel J. Evalu-
ation of matrix-assisted laser desorption ionization-time of ight mass
spectrometry for rapid identication of beta-hemolytic streptococci.
J Clin Microbiol 2011; 49:30045.
15. Davies AP, Reid M, Hadeld SJ, et al. Identication of clinical isolates
of alpha-hemolytic streptococci by 16S rRNA gene sequencing, matrix-
assisted laser desorption ionization-time of ight mass spectrometry
using MALDI Biotyper, and conventional phenotypic methods: a com-
parison. J Clin Microbiol 2012; 50:408790.
16. Alatoom AA, Cazanave CJ, Cunningham SA, Ihde SM, Patel R. Identi-
cation of non-diphtheriae Corynebacterium by use of matrix-assisted
laser desorption ionization-time of ight mass spectrometry. J Clin Mi-
crobiol 2012; 50:1603.
17. Pavlovic M, Konrad R, Iwobi AN, Sing A, Busch U, Huber I. A dual ap-
proach employing MALDI-TOF MS and real-time PCR for fast species
identication within the Enterobacter cloacae complex. FEMS Micro-
biol Lett 2012; 328:4653.
18. Alvarez-Buylla A, Culebras E, Picazo JJ. Identication of Acinetobacter
species: is Bruker Biotyper MALDI-TOF mass spectrometry a good al-
ternative to molecular techniques? Infect Genet Evol 2012; 12:3459.
19. Lamy B, Kodjo A, Laurent F. Identication of Aeromonas isolates by
matrix-assisted laser desorption ionization time-of-ight mass spec-
trometry. Diagn Microbiol Infect Dis 2011; 71:15.
20. He Y, Li H, Lu X, Stratton CW, Tang YW. Mass spectrometry biotyper
system identies enteric bacterial pathogens directly from colonies
grown on selective stool culture media. J Clin Microbiol 2010; 48:
388892.
21. Zangenah S, Ozenci V, Borang S, Bergman P. Identication of blood
and wound isolates of C. canimorsus and C. cynodegmi using VITEK2
and MALDI-TOF. Eur J Clin Microbiol Infect Dis 2012; 31:26317.
22. Kuhns M, Zautner AE, Rabsch W, et al. Rapid discrimination of Salmo-
nella enterica serovar Typhi from other serovars by MALDI-TOF mass
spectrometry. PLoS One 2012; 7:e40004.
23. Couturier MR, Mehinovic E, Croft AC, Fisher MA. Identication of
HACEK clinical isolates by matrix-assisted laser desorption ionization-
time of ight mass spectrometry. J Clin Microbiol 2011; 49:11046.
24. Gaia V, Casati S, Tonolla M. Rapid identication of Legionella spp. by
MALDI-TOF MS based protein mass ngerprinting. Syst Appl Micro-
biol 2011; 34:404.
25. Verroken A, Janssens M, Berhin C, et al. Evaluation of matrix-assisted
laser desorption ionization-time of ight mass spectrometry for identi-
cation of Nocardia species. J Clin Microbiol 2010; 48:401521.
26. Wybo I, Soetens O, De Bel A, et al. Species identication of clinical Pre-
votella isolates by matrix-assisted laser desorption ionization-time of
ight mass spectrometry. J Clin Microbiol 2012; 50:14158.
27. Schmitt BH, Cunningham SA, Dailey AL, Gustafson DR, Patel R. Iden-
tication of anaerobic bacteria by Bruker Biotyper matrix assisted laser
desorption ionization-time of ight mass spectrometry with on-plate
formic acid preparation. J Clin Microbiol; 2013; 51:7826.
28. Cunningham SA, Patel R. Importance of using Brukers Security Rele-
vant Library for Biotyper identication of Burkholderia pseudomallei,
Brucella species and Francisella tularensis. J Clin Microbiol 2013;
51:163940.
29. Saleeb PG, Drake SK, Murray PR, Zelazny AM. Identication of myco-
bacteria in solid-culture media by matrix-assisted laser desorption ioni-
zation-time of ight mass spectrometry. J Clin Microbiol 2011;
49:17904.
30. Bille E, Dauphin B, Leto J, et al. MALDI-TOF MS Andromas strategy
for the routine identication of bacteria, mycobacteria, yeasts, Aspergil-
lus spp. and positive blood cultures. Clin Microbiol Infect 2012;
18:111725.
31. Castanheira M, Woosley LN, Diekema DJ, Jones RN, Pfaller MA.
Candida guilliermondii and other species of Candida misidentied as
Candida famata: assessment by Vitek 2, DNA sequencing analysis, and
matrix-assisted laser desorption ionization-time of ight mass spec-
trometry in two global antifungal surveillance programs. J Clin Micro-
biol 2013; 51:11724.
32. Firacative C, Trilles L, Meyer W. MALDI-TOF MS enables the rapid
identication of the major molecular types within the Cryptococcus
neoformans/C. gattii species complex. PLoS One 2012; 7:e37566.
33. Sendid B, Ducoroy P, Francois N, et al. Evaluation of MALDI-TOF
mass spectrometry for the identication of medically-important yeasts
in the clinical laboratories of Dijon and Lille hospitals. Med Mycol
2013; 51:2532.
34. McTaggart LR, Lei E, Richardson SE, Hoang L, Fothergill A, Zhang SX.
Rapid identication of Cryptococcus neoformans and Cryptococcus
gattii by matrix-assisted laser desorption ionization-time of ight mass
spectrometry. J Clin Microbiol 2011; 49:30503.
570 CID 2013:57 (15 August) MEDICAL MICROBIOLOGY

b
y

g
u
e
s
t

o
n

A
u
g
u
s
t

2
3
,

2
0
1
4
h
t
t
p
:
/
/
c
i
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

35. Dhiman N, Hall L, Wohlel SL, Buckwalter SP, Wengenack NL. Perfor-
mance and cost analysis of matrix-assisted laser desorption ionization-
time of ight mass spectrometry for routine identication of yeast. J
Clin Microbiol 2011; 49:16146.
36. Van Herendael BH, Bruynseels P, Bensaid M, et al. Validation of a
modied algorithm for the identication of yeast isolates using matrix-
assisted laser desorption/ionisation time-of-ight mass spectrometry
(MALDI-TOF MS). Eur J Clin Microbiol Infect Dis 2012; 31:8418.
37. Theel ES, Schmitt BH, Hall L, et al. Formic acid-based direct, on-plate
testing of yeast and Corynebacterium species by Bruker Biotyper
matrix-assisted laser desorption ionization-time of ight mass spec-
trometry. J Clin Microbiol 2012; 50:30935.
38. De Carolis E, Posteraro B, Lass-Florl C, et al. Species identication of
Aspergillus, Fusarium and Mucorales with direct surface analysis by
matrix-assisted laser desorption ionization time-of-ight mass spec-
trometry. Clin Microbiol Infect 2012; 18:47584.
39. Theel ES, Hall L, Mandrekar J, Wengenack NL. Dermatophyte identi-
cation using matrix-assisted laser desorption ionization-time of ight
mass spectrometry. J Clin Microbiol 2011; 49:406771.
40. Coulibaly O, Marinach-Patrice C, Cassagne C, Piarroux R, Mazier D,
Ranque S. Pseudallescheria/Scedosporium complex species identica-
tion by matrix-assisted laser desorption ionization time-of-ight mass
spectrometry. Med Mycol 2011; 49:6216.
41. Murugaiyan J, Ahrholdt J, Kowbel V, Roesler U. Establishment of a
matrix-assisted laser desorption ionization time-of-ight mass spec-
trometry database for rapid identication of infectious achlorophyllous
green micro-algae of the genus Prototheca. Clin Microbiol Infect 2012;
18:4617.
42. Tan KE, Ellis BC, Lee R, Stamper PD, Zhang SX, Carroll KC. Prospec-
tive evaluation of a matrix-assisted laser desorption ionization-time of
ight mass spectrometry system in a hospital clinical microbiology lab-
oratory for identication of bacteria and yeasts: a bench-by-bench
study for assessing the impact on time to identication and cost-effec-
tiveness. J Clin Microbiol 2012; 50:33018.
43. Sparbier K, Weller U, Boogen C, Kostrzewa M. Rapid detection of Sal-
monella sp. by means of a combination of selective enrichment broth
and MALDI-TOF MS. Eur J Clin Microbiol Infect Dis 2012;
31:76773.
44. Gaillot O, Blondiaux N, Loiez C, et al. Cost-effectiveness of switch to
matrix-assisted laser desorption ionization-time of ight mass spec-
trometry for routine bacterial identication. J Clin Microbiol 2011;
49:4412.
45. Anderson NW, Buchan BW, Riebe KM, Parsons LN, Gnacinski S, Lede-
boer NA. Effects of solid-medium type on routine identication of bac-
terial isolates by use of matrix-assisted laser desorption ionization-time
of ight mass spectrometry. J Clin Microbiol 2012; 50:100813.
46. Xiao D, Zhao F, Lv M, et al. Rapid identication of microorganisms iso-
lated from throat swab specimens of community-acquired pneumonia
patients by two MALDI-TOF MS systems. Diagn Microbiol Infect Dis
2012; 73:3017.
47. De Bel A, Wybo I, Vandoorslaer K, Rosseel P, Lauwers S, Pierard D.
Acceptance criteria for identication results of gram-negative rods by
mass spectrometry. J Med Microbiol 2011; 60:6846.
48. Szabados F, Tix H, Anders A, Kaase M, Gatermann SG, Geis G. Evalua-
tion of species-specic score cutoff values of routinely isolated clinically
relevant bacteria using a direct smear preparation for matrix-assisted
laser desorption/ionization time-of-ight mass spectrometry-based bac-
terial identication. Eur J Clin Microbiol Infect Dis 2012; 31:110919.
49. Dupont C, Sivadon-Tardy V, Bille E, et al. Identication of clinical co-
agulase-negative staphylococci, isolated in microbiology laboratories,
by matrix-assisted laser desorption/ionization-time of ight mass spec-
trometry and two automated systems. Clin Microbiol Infect 2010;
16:9981004.
50. Sparbier K, Schubert S, Weller U, Boogen C, Kostrzewa M. Matrix-
assisted laser desorption ionization-time of ight mass spectrometry-
based functional assay for rapid detection of resistance against beta-
lactam antibiotics. J Clin Microbiol 2012; 50:92737.
51. Szabados F, Kaase M, Anders A, Gatermann SG. Identical MALDI
TOF MS-derived peak proles in a pair of isogenic SCCmec-harboring
and SCCmec-lacking strains of Staphylococcus aureus. J Infect 2012;
65:4005.
52. Wolters M, Rohde H, Maier T, et al. MALDI-TOF MS ngerprinting
allows for discrimination of major methicillin-resistant Staphylococcus
aureus lineages. Int J Med Microbiol 2011; 301:648.
53. Mencacci A, Monari C, Leli C, et al. Identication of nosocomial out-
breaks of Acinetobacter baumannii using MALDI-TOF mass spectrom-
etry. J Clin Microbiol 2013; 51:6036.
54. Boggs SR, Cazares LH, Drake R. Characterization of a Staphylococcus
aureus USA300 protein signature using matrix-assisted laser desorp-
tion/ionization time-of-ight mass spectrometry. J Med Microbiol
2012; 61:6404.
55. Grifn PM, Price GR, Schooneveldt JM, et al. Use of matrix-assisted
laser desorption ionization-time of ight mass spectrometry to identify
vancomycin-resistant enterococci and investigate the epidemiology of
an outbreak. J Clin Microbiol 2012; 50:291831.
56. Wybo I, De Bel A, Soetens O, et al. Differentiation of cA-negative and
cA-positive Bacteroides fragilis isolates by matrix-assisted laser desorp-
tion ionization-time of ight mass spectrometry. J Clin Microbiol 2011;
49:19614.
57. Kohling HL, Bittner A, Muller KD, et al. Direct identication of bacte-
ria in urine samples by matrix-assisted laser desorption/ionization
time-of-ight mass spectrometry and relevance of defensins as interfer-
ing factors. J Med Microbiol 2012; 61:33944.
58. Buchan BW, Riebe KM, Ledeboer NA. Comparison of the MALDI Bio-
typer system using Sepsityper specimen processing to routine microbi-
ological methods for identication of bacteria from positive blood
culture bottles. J Clin Microbiol 2012; 50:34652.
59. Meex C, Neuville F, Descy J, et al. Direct identication of bacteria from
BacT/ALERT anaerobic positive blood cultures by MALDI-TOF MS:
MALDI Sepsityper kit versus an in-house saponin method for bacterial
extraction. J Med Microbiol 2012; 61:15116.
60. Martiny D, Dediste A, Vandenberg O. Comparison of an in-house
method and the commercial Sepsityper kit for bacterial identication
directly from positive blood culture broths by matrix-assisted laser de-
sorption-ionisation time-of-ight mass spectrometry. Eur J Clin Micro-
biol Infect Dis 2012; 31:226981.
61. Loonen AJ, Jansz AR, Stalpers J, Wolffs PF, van den Brule AJ. An evalu-
ation of three processing methods and the effect of reduced culture
times for faster direct identication of pathogens from BacT/ALERT
blood cultures by MALDI-TOF MS. Eur J Clin Microbiol Infect Dis
2012; 31:157583.
62. Saffert RT, Cunningham SA, Mandrekar J, Patel R. Comparison of
three preparatory methods for detection of bacteremia by MALDI-TOF
mass spectrometry. Diagn Microbiol Infect Dis 2012; 73:216.
63. Fothergill A, Kasinathan V, Hyman J, Walsh J, Drake T, Wang YF.
Rapid identication of bacteria and yeasts from positive BacT/ALERT
blood culture bottles by using a lysis-ltration method and MALDI-
TOF mass spectrum analysis with SARAMIS database. J Clin Microbiol
2013: 51:8059.
64. Drancourt M. Detection of microorganisms in blood specimens using
matrix-assisted laser desorption ionization time-of-ight mass spec-
trometry: a review. Clin Microbiol Infect 2010; 16:16205.
65. Spanu T, Posteraro B, Fiori B, et al. Direct MALDI-TOF mass spec-
trometry assay of blood culture broths for rapid identication of
Candida species causing bloodstream infections: an observational study
in two large microbiology laboratories. J Clin Microbiol 2012; 50:
1769.
66. Ngan GJ, Lin RT, Teo JW. Utility of the Bruker Biotyper matrix-assisted
laser desorption ionisation time-of-ight mass spectrometer in a clini-
cal microbiology laboratory. Pathology 2012; 44:4936.
MEDICAL MICROBIOLOGY CID 2013:57 (15 August) 571

b
y

g
u
e
s
t

o
n

A
u
g
u
s
t

2
3
,

2
0
1
4
h
t
t
p
:
/
/
c
i
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

67. Kroumova V, Gobbato E, Basso E, Mucedola L, Giani T, Fortina G.
Direct identication of bacteria in blood culture by matrix-assisted
laser desorption/ionization time-of-ight mass spectrometry: a new
methodological approach. Rapid Commun Mass Spectrom 2011;
25:22479.
68. Clerc O, Prodhom G, Vogne C, Bizzini A, Calandra T, Greub G.
Impact of matrix-assisted laser desorption ionization time-of-ight
mass spectrometry on the clinical management of patients with gram-
negative bacteremia: a prospective observational study. Clin Infect Dis
2013; 56:11017.
69. Romero-Gomez MP, Gomez-Gil R, Pano-Pardo JR, Mingorance J.
Identication and susceptibility testing of microorganism by direct
inoculation from positive blood culture bottles by combining MALDI-
TOF and Vitek-2 Compact is rapid and effective. J Infect 2012;
65:51320.
70. Wimmer JL, Long SW, Cernoch P, et al. Strategy for rapid identication
and antibiotic susceptibility testing of gram-negative bacteria directly
recovered from positive blood cultures using the Bruker MALDI Bio-
typer and the BD Phoenix system. J Clin Microbiol 2012; 50:24524.
71. Perez K, Olsen R, Musick W, et al. Integrating rapid pathogen identi-
cation and antimicrobial stewardship signicantly decreases hospital
costs [Epub ahead of print]. Arch Pathol Lab Med. 2013; doi:10.5858/
arpa.2012-0651-0A.
72. Bessde E, Angla-Gre M, Delagarde Y, Sep Hieng S, Mnard A,
Mgraud F. Matrix-assisted laser-desorption/ionization biotyper: expe-
rience in the routine of a university hospital. Clin Microbiol Infect
2011; 17:5338.
73. Sogawa K, Watanabe M, Sato K, et al. Use of the MALDI BioTyper
system with MALDI-TOF mass spectrometry for rapid identication of
microorganisms. Anal Bioanal Chem 2011; 400:190511.
74. Neville SA, Lecordier A, Ziochos H, et al. Utility of matrix-assisted
laser desorption ionization-time of ight mass spectrometry following
introduction for routine laboratory bacterial identication. J Clin Mi-
crobiol 2011; 49:29804.
572 CID 2013:57 (15 August) MEDICAL MICROBIOLOGY

b
y

g
u
e
s
t

o
n

A
u
g
u
s
t

2
3
,

2
0
1
4
h
t
t
p
:
/
/
c
i
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m