Performances and Reliability of Bruker Microflex LT and VITEK MS MALDI-TOF Mass Spectrometry Systems for the Identification of Clinical Microorganisms.

In clinical microbiology laboratories, routine microbial identification is mostly performed using culture based methodologies requiring 24 to 72 hours from culturing to identification. Matrix assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) technology has been established as a cost effective, reliable, and faster alternative identification platform. In this study, we evaluated the reliability of the two available MALDI-TOF MS systems for their routine clinical level identification accuracy and efficiency in a clinical microbiology laboratory setting. A total of 1,341 routine phenotypically identified clinical bacterial and fungal isolates were selected and simultaneously analyzed using VITEK MS (bioMérieux, France) and Microflex LT (Bruker Diagnostics, Germany) MALDI-TOF MS systems. For any isolate that could not be identified with either of the systems and for any discordant result, 16S rDNA gene or ITS1/ITS2 sequencing was used. VITEK MS and Microflex LT correctly identified 1,303 (97.17%) and 1,298 (96.79%) isolates to the species level, respectively. In 114 (8.50%) isolates initial phenotypic identification was inaccurate. Both systems showed a similar identification efficiency and workflow robustness, and they were twice as more accurate compared to routine phenotypic identification in our sample pool. MALDITOF systems with their accuracy and robustness offer a good identification platform for routine clinical microbiology laboratories.

1. Introduction

Rapid and accurate identification of bacteria and yeasts from clinical specimens is crucial for the effective management of infections. In the clinical microbiology laboratories, microbial identification is conventionally done by phenotypic and biochemical analyses mostly using automated systems. These analyses require time ranging from a few hours to several days depending on microbial species in question. Workload and cost requirements for molecular methods, although they provide faster and accurate results, limit their routine use in clinical microbiology laboratories.

MALDI-TOF MS technology makes generation of unique mass spectral fingerprints of microorganisms possible, which are mostly a snapshot of ribosomal proteins ideal for an accurate microbial identification at the species level [1]. MALDI-TOF MS can rapidly and accurately identify a wide range of microorganisms at a reasonable cost using only a portion or the entire colony and a drop of matrix solution. While the MALDI-TOF MS analysis for the identification of intact bacterial cells was demonstrated 17 years ago [2, 3] and was extended to eukaryotic fungal cells 13 years ago [4, 5], not until recently has its potential for routine use been assessed for identification of a wide spectrum of bacteria, yeasts, molds, and mycobacteria that can be isolated in the clinical laboratories [6-13]. The ability of MALDI-TOF MS to directly identify bacteria in positive blood cultures is also important for the effective management of bloodstream infections [14, 15].

By this mean, two different CE marked in vitro diagnostic (IVD) MALDI TOF-MS systems were implemented to our laboratories in Ankara and Istanbul in March, 2012, and since then they have been used as the main identification tool for routine clinical bacterial and yeast isolates. In this study we evaluated performances of Bruker Microflex LT (Bruker Daltonics, Germany) and VITEK MS (bioMérieux, France) MALDI TOF-MS systems for identification of routine clinical microbiology isolates.

2. Materials and Methods

2.1. Clinical Isolates and Collection Strains

A total of 1,341 routine clinical isolates including 1,181 bacterial and 160 fungal specimens were analyzed using VITEK MS and Bruker Microflex LT. The isolates were obtained from clinical specimens including urine, blood, tissue, wounds, bronchoalveolar lavage, tracheal aspirate, sputum and wounds, which had been sent to our laboratory from various intensive care units, health facilities, and laboratories between April 2012 and December 2013. No selection criteria was applied for the isolates in order to obtain a comprehensive diversity.

Both MALDI-TOF MS systems were verified using 18 certified reference microorganisms obtained from reference culture collections.

2.2. Culture Conditions and Identification of Microorganisms

The specimens were routinely inoculated to appropriate media like, Columbia agar with 5% sheep blood, eosin methylene blue (EMB) agar, chromogenic urinary tract infection (CUTI) medium, chocolate agar, oxacillin resistant screening agar (ORSA), Salmonella-shigella agar, Thayer-Martin agar, Candida chromogenic agar, and blood culture bottles at 37°C depending on specimen type.

All of the specimens were incubated overnight at 37°C. The isolates were then Gram stained and identified using the reciprocate cards of VITEK II (bioMérieux, France) automated microbial identification system. The cultures and phenotypic identification results were transported to Ankara Duzen Laboratory Microbiology Department for VITEK MS analyses.

The strains were prepared and identified with VITEK MS system in Ankara and the results were documented. Following these initial identifications, culture plates were transported to our Clinical Laboratory Department located in İstanbul under appropriate conditions for the next day analysis with Bruker Microflex LT system.

2.3. VITEK MS MALDI-TOF System

From the overnight appropriate agar plates a portion or the whole colony was smeared onto the spots of VITEK MS-DS disposable target slides and then the spot was covered with 1 μL of α-cyano-4-hidroxycinnamic acid (CHCA) matrix solution. Spots were completely air dried, and, then, the slides were placed on an adapter and inserted to VITEK MS instrument. Spectra were generated using the MYLA software (bioMérieux, France) and the identification was automatically done with the VITEK MS IVD version 2.4.0-5 software. This software version contains >25,000 spectra, covering 586 species consisting of 508 bacteria and 78 fungi in its database. The software compares the spectrum obtained to the expected spectrum of each organism or organism group. Then the percent probability, a quantitative value, is calculated for each sample. The range of percent probabilities for a correct identification is from 60 to 99 with values closer to 99.9 indicating a closer match. When the obtained percent probability is under 60, then it was considered as no-identification. Confidence level is determined with percent probability and number of choices. During our analyses, although there was a unique identification pattern within the good confidence level, a list of possible organisms was given for certain samples. In such cases if both organisms belonged to the same genus then this result was accepted as a reliable identification to the genus level. In some other cases, when the strain belonged to a complex (e.g., Enterobacter cloacae complex) these results were documented as a correct result according to the cutoff value of the system; however the identification to the complex level was documented in order to compare the performances of the two systems accurately.

2.4. Bruker Microflex LT System

A portion or the whole colony was directly transferred onto the 96 spotted polished steel target plate. The spot was covered with 1 μL of CHCA matrix solution. After completely air drying, the plate was loaded to the Microflex LT instrument. The spectra were generated in linear positive ion mode with delayed extraction in a mass range of 4 to 10 kDa using a 337 nm nitrogen laser with a frequency of 60 Hz. The automated data analysis was processed with MALDI Biotyper Realtime Classification and Biotyper software version 3.0 (Bruker Daltonics). The obtained spectra were analyzed by standard pattern-matching algorithm, which compared the raw spectra with the spectra of the Bruker library by using the standard setting, and the results were listed in a ranking table. The results were expressed as log (score) values, which ranged from 0 to 3 as recommended by the manufacturer. Score values of >1.7 generally indicated relationships at the genus level, and values of >2.0 generally indicated relationships at the species level. The highest score was used for species identification. The Bruker library contains > 80,000 spectra covering 2,048 species and 385 genera. If the result was below 1.7 then the extraction method was performed where the colony was transferred into 1.5-mL screw cap tubes and mixed thoroughly in 0.3 mL of double-distilled ultrapure water. 0.9 mL of pure ethanol was added to tubes and after vortexing they were centrifuged at 13,000 ×g for 2 min. The supernatant was discarded and the pellet was mixed thoroughly with 50 μL of 70% aqueous formic acid. After the addition of 50 μL of acetonitrile, the mixture was centrifuged at 13,000 ×g for 2 min. One microliter of the microorganism extract supernatant was placed onto the polished steel and covered with 1 μL of CHCA matrix solution and loaded to the instrument.

2.5. Discordant Results

Results obtained from VITEK MS and Microflex LT were compared to initial onsite phenotypic identification results. The decision matrix used for the comparison and classification of results is given in Table 1. Briefly, when MS and the onsite phenotypic identification results agreed at the species level, then this was accepted as a correct result at the species level. When any MS equipment gave a correct result but as a species complex, then this was considered as a correct result and specified in the result table. For any discordant with the onsite phenotypic identification result we performed a second phenotypic identification. If this second phenotypic identification agreed with the MS result then this second result is accepted depending on correctness level, species, or genus. Otherwise a 16S or ITS1/2 sequencing analysis was performed depending on the nature of microorganism (prokaryote or eukaryote).

Table 1

Decision table used for the classification of results during the study.

Type	On-site phenotypic identification	MALDI-TOF MS analysis		Secondary phenotypic identification or molecular identification (16S or ITS1/2 sequencing)	Accepted	Number of m/o's (n = 1341∗)
		VITEK	Bruker
A	Identified as X	Identified as X	Identified as X	N/A	Accepted as X	1203
B	Identified as X or no-identification	Identified as Y	Identified as Y	Identified as Y	Accepted as Y	97
C	Identified as X	Identified as X	Identified as Y	Identified as X (or Y)	Accepted as X (or Y)	16
		Identified as Y	Identified as X
D	Identified as X	Identified as Y	Identified as Z	Identified as Y (or Z) (sequencing only)	Accepted as Y (or Z)	6
E	Identified as X	Identified as Y	Identified as Y	Identified as Z (sequencing only)	Accepted as Z	3
F	Identified as X	No-identification	Identified as X (or Y)	Identified as X (or Y or Z)	Accepted as X (or Y or Z)	12
		Identified as X (or Y)	No-identification
G	Identified as X	No-identification	No-identification	Identified as X (or Y) (sequencing only)	Accepted as X (or Y)	3

Shigella boydii was not included.

2.6. Extraction and Purification of Genomic DNA

Three to four colonies were selected from agar plates, resuspended in 500 μL 1x PBS (pH 7.4), and used for genomic DNA isolation. Genomic DNA isolation from Gram-negative bacteria was performed using DNA4U Bacterial Genomic DNA Isolation Kit (NANObiz, Turkey) as instructed by the manufacturer. DNA isolation from Gram-positive bacteria and other microorganisms were performed using conventional phenol/chloroform extraction method. Briefly, samples in 500 μL 1x PBS were centrifuged at 12,500 g for 5 minutes, resuspended in sterile water, recentrifuged, and resuspended in 500 μL TES buffer (50 mM Tris-HCl, pH 8.0; 1 mM EDTA; 6.7% sucrose). For Gram-positive bacteria 75 μL of 10 mg/mL lysozyme was added and samples were incubated at 37°C for 1 hour. After the incubation 125 μL of 20% SDS was added into each sample and incubated at 37°C for 30 minutes. Later 700 μL of phenol (SIGMA, USA) was added and samples were gently mixed using a vortex. Samples were then centrifuged at 2,000 rpm for 5 minutes and aqueous phases were removed into clean microcentrifuge tubes. Equal amount of 24 : 1 (v/v) chloroform : isoamyl alcohol was added and centrifuged at 2,000 rpm for 5 minutes. Aqueous phase was again removed into a clean microcentrifuge tube, and first 0.1 volume of ice cold 3 M potassium acetate, pH 5.5, and then 2 volumes of ice cold absolute ethanol were added. Samples were then incubated at −80°C for 15 minutes. Ethanol was removed through centrifugation at 10,000 rpm for 15 minutes, and this ethanol wash was repeated once more. Obtained pellets were then incubated at 50°C for approximately 15 minutes. Dry pellets were resuspended in 50 μL of sterile water. Purified genomic DNA was either visualized via agarose gel electrophoresis or directly measured for absorbance at 260 nm.

2.7. PCR Amplification of 16S Regions

The forward primer S-D-Bact-0008-a-S-20 (5′-AGA GTT TGA TCC TGG CTC AG-3′) [16] and the reverse primer rP2 (5′-ACG GCT ACC TTG TTA CGA CTT-3′) [17], which target the domain Bacteria, were used to amplify bacterial 16S rDNA sequences by PCR. Reaction mixture for PCR contained 1x PCR buffer (Fermentas), 2.0 mM MgCl2, 200 μM dNTP (each), 200 ng template genomic DNA, 0.5 μM of each primer, and 1.0 U Taq polymerase (Fermentas). DNA denaturation and amplification cycles were performed as described [16] using GeneAmp PCR System 9700 thermocycler (Applied Biosystems). Amplification products were checked via agarose gel electrophoresis.

2.8. PCR Amplification of ITS Regions

ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) primers were used to amplify internal transcribed spacer regions 1 and 2 of microorganisms [18]. Reaction mixture for PCR contained 1x PCR buffer (Fermentas), 1.5 mM MgCl2, 100 μM dNTP (each), 0.5 μM of each primer, and 1.0 U Taq polymerase (Fermentas) to a final concentration of 15 μL. DNA denaturation and amplification cycles were performed as described [18].

2.9. DNA Sequencing and Sequence Analyses

PCR products were sequenced via Sanger sequencing using 3130 Genetic Analyzer (Applied Biosystems). Obtained sequences were compared to 16S rDNA and/or ITS sequences that are found in public databases: NCBI (http://blast.ncbi.nlm.nih.gov/) and Green Genes 16S rDNA (http://greengenes.lbl.gov/).

3. Results

3.1. Reference Microorganisms

Eighteen reference bacterial strains were tested (Table 2) for verification and comparison of both MALDI-TOF MS systems. Salmonella typhimurium (ATCC 202165), Salmonella enteritidis (DM10), and Salmonella enterica subsp. enterica serovar typhimurium (ATCC 14028) were identified only to the genus level as Salmonella spp. These were all expected as they are within the current MALDI-TOF system limitations [1]. The rest of the reference microorganisms could all be accurately identified to the species level (Table 2).

Table 2

Results of 18 reference bacterial strains that were analyzed by VITEK MS and Bruker Microflex LT for an initial system setup and comparison.

Microorganism	VITEK	Bruker
S. pyogenes  ATCC 19615	✓	✓
E. coli  ATCC 25922	✓	✓
K. pneumoniae  ATCC 13883	✓	✓
P. aeruginosa  ATCC 27853	✓	✓
S. aureus  ATCC 25923	✓	✓
S. pneumoniae  ATCC 49619	✓	✓
P. vulgaris  ATCC 13315	P. vulgaris/penneri	P. hauseri  (new nomenclature)
S. typhimurium  ATCC 14028	Salmonella group	Salmonella spp.
Y. enterocolitica  ATCC 9610	✓	✓
V. cholera RSKK 913	✓	✓ (V. cholerae biovar albensis)
G. vaginalis  DM 108	✓	✓
S. saprophyticus  DM 17	✓	✓
S. agalactiae  DM 20	✓	✓
MRSA DM 66	S. aureus	S. aureus
S. epidermidis  DM 19	✓	✓
S. enteritidis  DM 10	Salmonella group	Salmonella spp.
A. hydrophila  DM 110	A. hydrophila/caviae	A. caviae
P. shigelloides  DM 111	✓	✓

3.2. Routine Clinical Bacterial Isolates

Our results indicated that both MS systems had high accuracy in the identification of routine clinical bacterial isolates that cover a wide range of species known to cause human infections (Tables 3 and 4). VITEK MS correctly identified 916 (77.56%) clinical bacterial isolates to the species and 231 (19.56%) isolates to the clinically relevant species group, totaling 1,147 (97.12%) correct identification on the species level. From the same pool, Bruker Microflex LT identified 1,142 (96.70%) clinical isolates to the species level.

Table 3

Summary of high-confidence identifications by VITEK MS and Bruker Microflex LT MS systems.

Microorganism groups	Number of isolates	Species content	Correct Identification				Misidentification		No-identification
			Species		Genus only
			VITEK	Bruker	VITEK	Bruker	VITEK	Bruker	VITEK	Bruker
Nonfermentatives	356 (30.1%)	16	352 (98.8%)	350 (98.4%)	—	3 (0.8%)	2 (0.6%)	—	2 (0.6%)	3 (0.8%)
Enterobacteriaceae	309 (26.2%)	23	292 (94.5%)	288 (93.2%)	15 (4.9%)	19 (6.2%)	1 (0.3%)	1 (0.3%)	1 (0.3%)	1 (0.3%)
Staphylococci	273 (23.1%)	10	271 (99.3%)	270 (98.9%)	2 (0.7%)	—	—	2 (0.7%)	—	1 (0.4%)
Streptococci	29 (2.5%)	11	28 (96.6%)	27 (93.2%)	—	1 (3.4%)	—	1 (3.4%)	1 (3.4%)	—
Enterococci and other Gr(+) cocci	160 (13.6%)	5	160 (100.0%)	160 (100.0%)	—	—	—	—	—	—
Coryneforms and other Gr(+) bacilli	30 (2.5%)	14	21 (70.00%)	25 (83.3%)	2 (6.7%)	4 (13.4%)	2 (6.7%)	—	5 (16.6%)	1 (3.3%)
HACEK andother Gr(−) cocci	18 (1.5%)	8	18 (100.0%)	18 (100.0%)	—	—	—	—	—	—
Anaerobes	2 (0.2%)	2	2 (100.0%)	2 (100.0%)	—	—	—	—	—	—
Others	4 (0.3%)	4	3 (75.0%)	2 (50.0%)	—	1 (25.0%)	—	—	1 (25.0%)	1 (25.0%)
TOTAL (bacteria)	1,181	93	1,147 (97.1%)	1,142 (96.7%)	19 (1.6%)	28 (2.4%)	5 (0.4%)	4 (0.3%)	10 (0.9%)	7 (0.6%)
Fungi	160	14	156 (97.5%)	156 (97.5%)	3 (1.9%)	4 (2.5%)	—	—	1 (0.6%)	—
TOTAL (bacteria + fungi)	1,341	107	1,303 (97.2%)	1,298 (96.8%)	22 (1.6%)	32 (2.4%)	5 (0.4%)	4 (0.3%)	11 (0.8%)	7 (0.5%)

Table 4

High-confidence identifications by VITEK MS and Bruker Microflex LT MS systems.

Microorganism groups	# of isolates	Correct Identification					Misidentification		No-identification
		Species			Genus only
		VITEK		Bruker	VITEK	Bruker	VITEK	Bruker	VITEK	Bruker
		Species	Species Complex
Nonfermentatives (16 species)
Achromobacter xylosoxidans	6	4	2	6	—	—	—	—	—	—
Acinetobacter baumannii	194	—	192	192	—	—	—	—	2 (F × 2)	2 (F × 2)
Acinetobacter guillouiae	1	—	1	1	—	—	—	—	—	—
Acinetobacter johnsonii	1	1	—	1	—	—	—	—	—	—
Acinetobacter lwoffii	1	1	—	1	—	—	—	—	—	—
Acinetobacter radioresistens	1	1	—	1	—	—	—	—	—	—
Aeromonas hydrophila	3	—	3	2	—	1 (C)	—	—	—	—
Aeromonas sobria	1	—	1	—	—	—	—	—	—	1 (F)
Alcaligenes faecalis	2	2	—	2	—	—	—	—	—	—
Burkholderia cepacia	2	—	2	—	—	2 (C × 2)	—	—	—	—
Comamonas aquatica	1	—	—	1	—	—	1 (D)	—	—	—
Delftia acidovorans	3	3	—	3	—	—	—	—	—	—
Elizabethkingia meningoseptica	1	1	—	1	—	—	—	—	—	—
Pseudomonas aeruginosa	125	124	—	125	—	—	1 (C)	—	—	—
Pseudomonas fluorescens	1	1	—	1	—	—	—	—	—	—
Stenotrophomonas maltophilia	13	13	—	13	—	—	—	—	—	—
Total	356	151 (42.4%)	201 (56.4%)	350 (98.4%)	—	3 (0.8%)	2 (0.6%)	—	2  (0.6%)	3  (0.8%)
Enterobacteriaceae (23 species)
Chryseobacterium indologenes	1	1	—	1	—	—	—	—	—	—
Citrobacter braakii	3	—	3	3	—	—	—	—	—	—
Citrobacter freundii	5	3	2	5	—	—	—	—	—	—
Citrobacter sedlakii	1	1	—	1	—	—	—	—	—	—
Citrobacter youngae	2	—	2	1	—	1 (C)	—	—	—	—
Enterobacter aerogenes	4	4	—	4	—	—	—	—	—	—
Enterobacter asburiae	2	—	2	1	—	1 (C)	—	—	—	—
Enterobacter cancerogenous	1	1	—	1	—	—	—	—	—	—
Enterobacter cloacae	11	—	11	9	—	2 (C × 2)	—	—	—	—
Escherichia coli	123	123	—	123	—	—	—	—	—	—
Klebsiella oxytoca	5	5	—	5	—	—	—	—	—	—
Klebsiella pneumonia	64	64	—	64	—	—	—	—	—	—
Kluyvera cryocrescens	1	—	—	—	—	—	—	—	1 (G)	1 (G)
Morganella morganii	12	12	—	12	—	—	—	—	—	—
Proteus mirabilis	28	28	—	28	—	—	—	—	—	—
Providencia rettgeri	3	3	—	3	—	—	—	—	—	—
Providencia stuartii	1	1	—	1	—	—	—	—	—	—
Raoultella ornithinolytica	3	2	1	3	—	—	—	—	—	—
Raoultella planticola	1	—	1	1	—	—	—	—	—	—
Salmonella spp.	15	—	—	—	15	15	—	—	—	—
Serratia liquefaciens	1	1	—	1	—	—	—	—	—	—
Serratia marcescens	21	21	—	21	—	—	—	—	—	—
Shigella boydii	1	—	—	—	—	—	1 (N/A)	1 (N/A)	—	—
Total	309	270 (87.4%)	22 (7.1%)	288 (93.3%)	15 (4.9%)	19 (6.1%)	1 (0.3%)	1  (0.3%)	1  (0.3%)	1  (0.3%)
Staphylococci (10 species)
Staphylococcus aureus	52	52	—	52	—	—	—	—	—	—
Staphylococcus capitis	8	8	—	8	—	—	—	—	—	—
Staphylococcus caprae	1	1	—	1	—	—	—	—	—	—
Staphylococcus epidermidis	68	68	—	65	—	—	—	2 (C × 2)	—	1 (F)
Staphylococcus haemolyticus	54	54	—	54	—	—	—	—	—	—
Staphylococcus hominis	82	82	—	82	—	—	—	—	—	—
Staphylococcus lugdunensis	1	1	—	1	—	—	—	—	—	—
Staphylococcus pettenkoferi	1	—	—	1	1 (C)	—	—	—	—	—
Staphylococcus sciuri	2	2	—	2	—	—	—	—	—	—
Staphylococcus simulans	4	3	—	4	1 (D)	—	—	—	—	—
Total	273	271 (99.3%)	—	270 (98.9%)	2 (0.7%)	—	—	2  (0.7%)	—	1  (0.4%)
Streptococci (11 species)
Leuconostoc mesenteroides	1	—	—	—	—	—	—	1 (F)	1 (F)	—
Streptococcus agalactiae	9	9	—	9	—	—	—	—	—	—
Streptococcus anginosus	1	1	—	1	—	—	—	—	—	—
Streptococcus dysgalactiae	3	3	—	3	—	—	—	—	—	—
Streptococcus gallolyticus	1	1	—	1	—	—	—	—	—	—
Streptococcus mitis	1	—	1	—	—	1 (C)	—	—	—	—
Streptococcus oralis	1	—	1	1	—	—	—	—	—	—
Streptococcus parasanguinis	1	1	—	1	—	—	—	—	—	—
Streptococcus pneumonia	4	4	—	4	—	—	—	—	—	—
Streptococcus pyogenes	6	6	—	6	—	—	—	—	—	—
Streptococcus salivarius	1	1	—	1	—	—	—	—	—	—
Total	29	26 (89.7%)	2 (6.9%)	27 (93.2%)	—	1 (3.4%)	—	1  (3.4%)	1  (3.4%)	—
Enterococci and other Gr(+) cocci (5 species)
Aerococcus viridans	4	4	—	4	—	—	—	—	—	—
Enterococcus avium	4	4	—	4	—	—	—	—	—	—
Enterococcus faecalis	54	54	—	54	—	—	—	—	—	—
Enterococcus faecium	84	84	—	84	—	—	—	—	—	—
Enterococcus gallinarum	14	14	—	14	—	—	—	—	—	—
Total	160	160 (100%)	—	160 (100%)	—	—	—	—	—	—
Coryneforms and other Gr(+) bacilli (14 species)
Actinomyces naeslundii	1	—	—	—	—	1 (F)	—	—	1 (F)	—
Bacillus cereus	1	—	1	1	—	—	—	—	—	—
Bacillus simplex	1	1	—	1	—	—	—	—	—	—
Corynebacterium amycolatum	4	—	4	4	—	—	—	—	—	—
Corynebacterium appendicis	1	—	—	—	1 (E)	1 (E)	—	—	—	—
Corynebacterium coyleae	1	—	—	1	—	—	1 (D)	—	—	—
Corynebacterium glaucum	1	—	—	—	—	—	—	—	1 (G)	1 (G)
Corynebacterium jeikeium	2	2	—	2	—	—	—	—	—	—
Corynebacterium mucifaciens	3	—	—	3	—	—	1 (D)	—	2 (F)	—
Corynebacterium striatum	10	10	—	10	—	—	—	—	—	—
Corynebacterium urealyticum	1	1	—	1	—	—	—	—	—	—
Corynebacterium ureicelerivorans	2	—	—	—	1 (E)	2 (E, F)	—	—	1 (F)	—
Lactobacillus paracasei	1	—	1	1	—	—	—	—	—	—
Listeria monocytogenes	1	1	—	1	—	—	—	—	—	—
Total	30	15 (50.0%)	6 (20.0%)	25 (83.4%)	2 (6.7%)	4 (13.3%)	2 (6.7%)	—	5  (16.6%)	1  (3.3%)
HACEK and other Gr(−) cocci (8 species)
Haemophilus haemolyticus	1	1	—	1	—	—	—	—	—	—
Haemophilus influenza	3	3	—	3	—	—	—	—	—	—
Haemophilus parahaemolyticus	2	2	—	2	—	—	—	—	—	—
Haemophilus parainfluenzae	1	1	—	1	—	—	—	—	—	—
Legionella pneumophila	7	7	—	7	—	—	—	—	—	—
Moraxella catarrhalis	1	1	—	1	—	—	—	—	—	—
Neisseria meningitides	1	1	—	1	—	—	—	—	—	—
Pasteurella multocida	2	2	—	2	—	—	—	—	—	—
Total	18	18 (100%)	—	18 (100%)	—	—	—	—	—	—
Anaerobes (2 species)
Clostridium perfringens	1	1	—	1	—	—	—	—	—	—
Clostridium sporogenes	1	1	—	1	—	—	—	—	—	—
Total	2	2 (100%)	—	2 (100%)	—	—	—	—	—	—
Others (4 species)
Bergeyella zoohelcum	1	1	—	1	—	—	—	—	—	—
Myroides odoratimimus	1	1	—	—	—	1 (C)	—	—	—	—
Rhodococcus equi	1	1	—	1	—	—	—	—	—	—
Wautersiella falsenii	1	—	—	—	—	—	—	—	1 (G)	1 (G)
Total	4	3 (75.0%)	—	2 (50.0%)	—	1 (C) (25.0%)	—	—	1 (G)  (25.0%)	1 (G) (25.0%)
Total (bacteria)	1,181	916 (77.6%)	231 (19.6%)	1,142 (96.7%)	19 (1.6%)	28 (2.4%)	5 (0.4%)	4  (0.3%)	10  (0.8%)	7  (0.6%)
Fungi (14 species)
Aspergillus brasiliensis	1	—	—	1	—	—	—	—	1 (F)	—
Candida albicans	43	43	—	41	—	2 (C × 2)	—	—	—	—
Candida dubliniensis	1	1	—	1	—	—	—	—	—	—
Candida glabrata	44	44	—	44	—	—	—	—	—	—
Candida kefyr	11	11	—	11	—	—	—	—	—	—
Candida krusei	4	4	—	4	—	—	—	—	—	—
Candida lusitaniae	2	2	—	2	—	—	—	—	—	—
Candida norvegensis	1	1	—	1	—	—	—	—	—	—
Candida orthopsilosis	2	—	—	2	2 (D × 2)	—	—	—	—	—
Candida parapsilosis	14	14	—	13	—	1 (C)	—	—	—	—
Candida tropicalis	32	31	1	32
Cryptococcus neoformans	1	1	—	1	—	—	—	—	—	—
Meyerozyma caribbica	1	—	—	—	1 (E)	1 (E)	—	—	—	—
Trichosporon asahii	3	3	—	3	—	—	—	—	—	—
Total (fungi)	160	155 (96.9%)	1 (0.6%)	156 (97.5%)	3 (1.9%)	4 (2.5%)	—	—	1  (0.6%)	—
TOTAL (bacteria + fungi)	1,341	1,071 (79.9%)	232 (17.3%)	1,298 (96.8%)	22 (1.6%)	32 (2.4%)	5 (0.4%)	4  (0.3%)	11 (0.8%)	7  (0.5%)
		1,303 (97.2%)

Identification on genus only level was also similar for both systems. VITEK MS could identify 19 (1.61%) and Bruker Microflex LT could identify 28 (2.37%) isolates only to the genus level. Genus level, but not to species level identification of Salmonella spp. (n = 15) accounts for the main decrease in the percentage for correct species level identification for both systems, which is a known system limitation.

Misidentification numbers were similar for VITEK MS and Bruker Microflex LT with 5 (0.42%) and 4 (0.33%) isolates, respectively. Shigella boydii was incorrectly identified as E. coli by both systems, but this is also a known system limitation.

Except one Leuconostoc mesenteroides isolate, the remaining misidentified isolates were identified by at least one system. L. mesenteroides was misidentified as Staphylococcus pettenkoferi by Bruker Microflex LT, whereas it could not be identified by VITEK MS.

A total of three isolates, Wautersiella falsenii, Kluyvera cryocrescens, and Corynebacterium glaucum, could not be identified by both systems (Table 4).

3.3. Fungal Samples

We have observed a very similar identification accuracy in both systems. Out of 160 fungal samples VITEK MS correctly identified 155 (96.88%) isolates to the species, 1 (0.62%) isolate to the species complex, and 3 (1.88%) isolates to the genus level. Similarly, Bruker Microflex LT identified 156 (97.50%) fungal isolates to the species level and 4 (2.50%) isolates to the genus level.

There was no misidentification in both systems. VITEK MS could not identify an Aspergillus brasiliensis isolate as this microorganism is not present in the system's library, whereas Bruker Microflex LT identified this isolate to the species level.

4. Discussion

Our results indicated that, using any of the two available MALDI-TOF MS systems, a species level correct identification accuracy of over 96% could be achieved without any prior information about the tested samples. We have reached this high accuracy level using the regular sample extraction methods without any strain specific pretreatment in order to determine the efficiency of MS systems in the routine clinical microbiology settings.

In our laboratory, conventional biochemistry based phenotypic microbial identification methods costs about $15 and typically takes from 6 to 18 hours depending on the tested microorganisms. MALDI-TOF MS system, on the other hand, requires only 5 to 30 minutes for the identification and it costs less than $1 per sample.

For this study we collected various clinical specimens which were sent to our laboratory with an initial on-site identification that had been carried out using a VITEK II Compact phenotypic identification platform. During our analyses; for 97 (7.23%) isolates both MS systems gave a different result than this initial on-site phenotypic identification. Our further molecular (16S or ITS1/2 sequencing) or phenotypic analysis resulted in favor of MS systems for all of these 97 isolates. Additional 17 (0.01%) isolates were also confirmed to be different than the on-site phenotypic identification and these isolates were misidentified at least by one of the MS systems. As a result, a total of 114 (8.50%) isolates were misidentified by phenotypic methods, whereas an average of 37 (2.76%) isolates were misidentified or could not be identified by MS systems used in this study, making them at least twice as more accurate (Table 6). Using the wrong ID card and in some few instances wrong labeling might account for these incorrect on-site identifications. MS systems, as they do not need prior strain information, proved themselves more accurate and efficient for this instance.

Table 6

Detailed results of microorganisms that were either misidentified or could not be identified by at least one MALDI-TOF system or phenotypic identification.

Correct ID	Number of isolates	On-site phenotypic identification	Secondary phenotypic identification or molecular identification∗  (16S or ITS1/2 sequencing)	VITEK	Bruker
Nonfermentatives
Achromobacter xylosoxidans	1	Achromobacter denitrificans	∗ Achromobacter xylosoxidans	Achromobacter denitrificans/xylosoxidans	Achromobacter xylosoxidans
Acinetobacter baumannii	1	Cedecea lapegei	Acinetobacter baumannii	Acinetobacter baumannii complex	Acinetobacter baumannii
Acinetobacter baumannii	1	Escherichia coli	Acinetobacter baumannii complex	Acinetobacter baumannii complex	Acinetobacter baumannii
Acinetobacter baumannii	1	Klebsiella pneumoniae	Acinetobacter baumannii	Acinetobacter baumannii complex	Acinetobacter baumannii
Acinetobacter baumannii	1	Pseudomonas aeruginosa	Acinetobacter baumannii	Acinetobacter baumannii complex	Acinetobacter baumannii
Acinetobacter baumannii	1	Staphylococcus hominis	Acinetobacter baumannii complex	Acinetobacter baumannii complex	Acinetobacter baumannii
Acinetobacter baumannii	2	Acinetobacter baumannii	Acinetobacter baumannii	No identification	Acinetobacter baumannii
Acinetobacter baumannii	2	Acinetobacter baumannii	Acinetobacter baumannii	Acinetobacter baumannii complex	No identification
Acinetobacter guillouiae	1	Acinetobacter lwoffii	∗ Acinetobacter guillouiae	Acinetobacter baumannii complex	Acinetobacter guillouiae
Acinetobacter johnsonii	1	Acinetobacter lwoffii	∗ Acinetobacter johnsonii	Acinetobacter johnsonii	Acinetobacter johnsonii
Acinetobacter radioresistens	1	Acinetobacter lwoffii	∗ Acinetobacter radioresistens	Acinetobacter radioresistens	Acinetobacter radioresistens
Aeromonas hydrophila	1	Aeromonas sobria	∗ Aeromonas hydrophila	Aeromonas hydrophila/caviae	Aeromonas hydrophila
Aeromonas hydrophila	1	Aeromonas hydrophila/caviae	Aeromonas hydrophila	Aeromonas hydrophila/caviae	Aeromonas caviae
Aeromonas sobria	1	Aeromonas sobria	∗ Aeromonas sobria	Aeromonas hydrophila/caviae, sobria	No identification
Alcaligenes faecalis	1	Chromobacterium violaceum	∗ Alcaligenes faecalis	Alcaligenes faecalis	Alcaligenes faecalis
Burkholderia cepacia	2	Burkholderia cepacia	∗ Burkholderia cepacia	Burkholderia cepacia/vietnamiensis	Burkholderia cenocepacia
Comamonas aquatica	1	Comamonas testosteroni	∗ Comamonas aquatic	Delftia acidovorans	Comamonas aquatica
Delftia acidovorans	1	No identification	Delftia acidovorans	D. acidovorans	D. acidovorans
Delftia acidovorans	1	No identification	Delftia acidovorans	Delftia acidovorans	Delftia acidovorans
Pseudomonas aeruginosa	3	Acinetobacter baumannii	Pseudomonas aeruginosa	Pseudomonas aeruginosa	Pseudomonas aeruginosa
Pseudomonas aeruginosa	1	Enterobacter aerogenes	Pseudomonas aeruginosa	Pseudomonas aeruginosa	Pseudomonas aeruginosa
Pseudomonas aeruginosa	1	Pseudomonas aeruginosa	Pseudomonas aeruginosa	Porphyromonas gingivalis	Pseudomonas aeruginosa
Stenotrophomonas maltophilia	1	Pseudomonas aeruginosa	Stenotrophomonas maltophilia	Stenotrophomonas maltophilia	Stenotrophomonas maltophilia
Enterobacteriaceae
Chryseobacterium indologenes	1	No identification	Chryseobacterium indologenes	Chryseobacterium indologenes	Chryseobacterium indologenes
Citrobacter braakii	1	Citrobacter freundii	∗ Citrobacter braakii	Citrobacter braakii/farmeri/freundii	Citrobacter braakii
Citrobacter freundii	1	No identification	Citrobacter freundii	Citrobacter freundii	Citrobacter freundii
Citrobacter sedlakii	1	Citrobacter amalonaticus	∗ Citrobacter sedlakii	Citrobacter sedlakii	Citrobacter sedlakii
Citrobacter youngae	1	No identification	Citrobacter youngae	Citrobacter freundii/youngae	Citrobacter youngae
Citrobacter youngae	1	Citrobacter youngae	∗ Citrobacter youngae	Citrobacter freundii/youngae	Citrobacter freundii
Enterobacter asburiae	1	Escherichia coli	Enterobacter cloacae complex	Enterobacter cloacae/asburiae	Enterobacter asburiae
Enterobacter asburiae	1	Enterobacter cloacae complex	∗ Enterobacter asburiae	Enterobacter cloacae/asburiae	Enterobacter cloacae
Enterobacter cancerogenous	1	Enterobacter aerogenes	Enterobacter cancerogenous	Enterobacter cancerogenous	Enterobacter cancerogenous
Enterobacter cloacae	1	Enterobacter cancerogenus	Enterobacter cloacae complex	Enterobacter cloacae/asburiae	Enterobacter cloacae
Enterobacter cloacae	1	No identification	Enterobacter cloacae complex	Enterobacter cloacae/asburiae	Enterobacter cloacae
Enterobacter cloacae	1	No identification	Enterobacter cloacae	Enterobacter cloacae/asburiae	Enterobacter cloacae
Enterobacter cloacae	1	Enterobacter cloacae	∗ Enterobacter cloacae	Enterobacter cloacae/asburiae	Enterobacter asburiae
Enterobacter cloacae	1	Enterobacter cloacae	Enterobacter cloacae	Enterobacter cloacae/asburiae	Enterobacter asburiae
Enterobacter cloacae	1	No identification	Enterobacter cloacae	Enterobacter cloacae/asburiae	Enterobacter cloacae
Klebsiella oxytoca	1	Escherichia coli	Klebsiella oxytoca	Klebsiella oxytoca	Klebsiella oxytoca
Klebsiella pneumoniae	1	Acinetobacter baumannii	Klebsiella pneumonia	Klebsiella pneumoniae	Klebsiella pneumoniae
Klebsiella pneumoniae	1	Enterobacter aerogenes	Klebsiella pneumonia	Klebsiella pneumoniae	Klebsiella pneumoniae
Klebsiella pneumoniae	1	Enterobacter aerogenes	∗ Klebsiella pneumoniae	Klebsiella pneumoniae	Klebsiella pneumoniae
Klebsiella pneumoniae	1	Escherichia coli	Klebsiella pneumonia	Klebsiella pneumoniae	Klebsiella pneumoniae
Kluyvera cryocrescens	1	Raoultella planticola	∗ Kluyvera cryocrescens	No identification	No identification
Proteus mirabilis	1	Stenotrophomonas maltophilia	Proteus mirabilis	Proteus mirabilis	Proteus mirabilis
Raoultella ornithinolytica	1	No identification	∗ Raoultella ornithinolytica	Raoultella ornithinolytica/planticola	Raoultella ornithinolytica
Shigella boydii	1	Shigella boydii	Shigella boydii	Escherichia coli	Escherichia coli
Staphylococci
Staphylococcus aureus	1	Acinetobacter baumannii	Staphylococcus aureus	Staphylococcus aureus	Staphylococcus aureus
Staphylococcus hominis	1	Acinetobacter baumannii	Staphylococcus hominis	Staphylococcus hominis	Staphylococcus hominis
Staphylococcus epidermidis	1	Enterococcus faecium	Staphylococcus epidermidis	Staphylococcus epidermidis	Staphylococcus epidermidis
Staphylococcus simulans	1	Granulicatella elegans	∗ Staphylococcus simulans	Staphylococcus haemolyticus	Staphylococcus simulans
Staphylococcus haemolyticus	1	Staphylococcus aureus	Staphylococcus haemolyticus	Staphylococcus haemolyticus	Staphylococcus haemolyticus
Staphylococcus pettenkoferi	1	Staphylococcus auricularis	∗ Staphylococcus pettenkoferi	Staphylococcus auricularis/capitis	Staphylococcus pettenkoferi
Staphylococcus hominis	1	Staphylococcus epidermidis	Staphylococcus hominis	Staphylococcus hominis	Staphylococcus hominis
Staphylococcus aureus	1	Staphylococcus haemolyticus	Staphylococcus aureus	Staphylococcus aureus	Staphylococcus aureus
Staphylococcus capitis	1	Staphylococcus haemolyticus	Staphylococcus capitis	Staphylococcus capitis	Staphylococcus capitis
Staphylococcus aureus	1	Staphylococcus hominis	Staphylococcus aureus	Staphylococcus aureus	Staphylococcus aureus
Staphylococcus epidermidis	1	Staphylococcus hominis	Staphylococcus epidermidis	Staphylococcus epidermidis	Staphylococcus epidermidis
Staphylococcus epidermidis	1	Staphylococcus lentus	Staphylococcus epidermidis	Staphylococcus epidermidis	Staphylococcus epidermidis
Staphylococcus hominis	1	Staphylococcus saprophyticus	Staphylococcus hominis	Staphylococcus hominis	Staphylococcus hominis
Staphylococcus hominis	1	Staphylococcus warneri	∗ Staphylococcus hominis	Staphylococcus hominis	Staphylococcus hominis
Staphylococcus hominis	2	Staphylococcus warneri	∗ Staphylococcus hominis	Staphylococcus hominis	Staphylococcus hominis
Staphylococcus haemolyticus	1	Staphylococcus warneri	Staphylococcus haemolyticus	Staphylococcus haemolyticus	Staphylococcus haemolyticus
Staphylococcus epidermidis	1	Staphylococcus epidermidis	Staphylococcus epidermidis	Staphylococcus epidermidis	Bacillus pumilus
Staphylococcus epidermidis	1	Staphylococcus epidermidis	Staphylococcus epidermidis	Staphylococcus epidermidis	Dermobacter hominis
Staphylococcus aureus	1	No identification	Staphylococcus aureus	Staphylococcus aureus	Staphylococcus aureus
Staphylococcus lugdunensis	1	No identification	Staphylococcus lugdunensis	Staphylococcus lugdunensis	Staphylococcus lugdunensis
Staphylococcus epidermidis	1	Staphylococcus epidermidis	Staphylococcus epidermidis	Staphylococcus epidermidis	No identification
Streptococci
Leuconostoc mesenteroides	1	Staphylococcus capitis	∗ Leuconostoc mesenteroides	No identification	Staphylococcus pettenkoferi
Streptococcus agalactiae	3	No identification	Streptococcus agalactiae	Streptococcus agalactiae	Streptococcus agalactiae
Streptococcus dysgalactiae	1	No identification	∗ Streptococcus dysgalactiae	Streptococcus dysgalactiae	Streptococcus dysgalactiae
Streptococcus mitis	1	Streptococcus mitis	Streptococcus mitis	Streptococcus parasanguinis/mitis/oralis	Streptococcus pneumoniae
Streptococcus salivarius	1	Streptococcus lentus	Streptococcus salivarius	Streptococcus salivarius	Streptococcus salivarius
Enterococci & other Gr(+) cocci
Enterococcus faecalis	1	Escherichia coli	Enterococcus faecalis	Enterococcus faecalis	Enterococcus faecalis
Enterococcus faecium	1	Enterococcus faecalis	Enterococcus faecium	Enterococcus faecium	Enterococcus faecium
Enterococcus faecalis	2	Enterococcus faecium	Enterococcus faecalis	Enterococcus faecalis	Enterococcus faecalis
Enterococcus gallinarum	1	Enterococcus faecium	∗ Enterococcus gallinarum	Enterococcus gallinarum	Enterococcus gallinarum
Enterococcus faecium	2	Enterococcus gallinarum	Enterococcus faecium	Enterococcus faecium	Enterococcus faecium
Enterococcus faecalis	1	Serratia fonticola	Enterococcus faecalis	Enterococcus faecalis	Enterococcus faecalis
Enterococcus faecalis	1	Staphylococcus hominis	Enterococcus faecalis	Enterococcus faecalis	Enterococcus faecalis
Coryneforms & other Gr(+) bacilli
Corynebacterium striatum	1	Actinomyces spp.	Corynebacterium striatum	Corynebacterium striatum	Corynebacterium striatum
Corynebacterium urealyticum	1	Brucella spp.	Corynebacterium urealyticum	Corynebacterium urealyticum	Corynebacterium urealyticum
Corynebacterium appendicis	1	Corynebacterium jeikeium	∗ Corynebacterium appendicis	Corynebacterium pseudodiphtericum	Corynebacterium pseudodiphtericum
Corynebacterium ureicelerivorans	1	Corynebacterium jeikeium	∗ Corynebacterium ureicelerivorans	Corynebacterium afermentans	Corynebacterium afermentans
Corynebacterium mucifaciens	2	Corynebacterium jeikeium	∗ Corynebacterium mucifaciens	No identification	Corynebacterium mucifaciens
Corynebacterium ureicelerivorans	1	Corynebacterium jeikeium	∗ Corynebacterium ureicelerivorans	No identification	Corynebacterium coyleae
Corynebacterium glaucum	1	Corynebacterium jeikeium	∗ Corynebacterium glaucum	No identification	No identification
Corynebacterium mucifaciens	1	Clostridium spp.	∗ Corynebacterium mucifaciens	Rhizobium radiobacter	Corynebacterium mucifaciens
Corynebacterium coyleae	1	Dermococcus/Kytococcus/  Micrococcus	∗ Corynebacterium coyleae	Arthrobacter cumminsii	Corynebacterium coyleae
Lactobacillus paracasei	1	Pediococcus pentosaceus	∗ Lactobacillus paracasei	Lactobacillus casei/paracasei	Lactobacillus paracasei
Corynebacterium striatum	1	Propionibacterium acnes	Corynebacterium striatum	Corynebacterium striatum	Corynebacterium striatum
Bacillus simplex	1	Sphingomonas paucimobilis	Bacillus simplex	Bacillus simplex	Bacillus simplex
Corynebacterium amycolatum	1	Bacillus spp.	Corynebacterium amycolatum	Corynebacterium amycolatum/xerosis	Corynebacterium amycolatum
Corynebacterium jeikeium	2	Bacillus spp.	Corynebacterium jeikeium	Corynebacterium jeikeium	Corynebacterium jeikeium
Corynebacterium striatum	2	Bacillus spp.	Corynebacterium striatum	Corynebacterium striatum	Corynebacterium striatum
Actinomyces naeslundii	1	Actinomyces naeslundii	∗ Actinomyces naeslundii	No identification	Actinomyces spp.
HACEK & other Gr(−) cocci
Moraxella catarrhalis	1	Arcanobacterium hemoliticum	Moraxella catarrhalis	Moraxella catarrhalis	Moraxella catarrhalis
Haemophilus influenzae	1	Bacillus spp.	Haemophilus influenzae	Haemophilus influenzae	Haemophilus influenzae
Pasteurella multocida	1	Bacillus spp.	∗ Pasteurella multocida	Pasteurella multocida	Pasteurella multocida
Others
Bergeyella zoohelcum	1	No identification	∗ Bergeyella zoohelcum	Bergeyella zoohelcum	Bergeyella zoohelcum
Wautersiella falsenii	1	Weeksella virosa	∗ Wautersiella falsenii	No identification	No identification
Myroides odoratimimus	1	Myroides spp.	∗ Myroides odoratimimus	Myroides odoratimimus	Myroides spp.
Fungi
Candida glabrata	2	Candida albicans	Candida glabrata	Candida glabrata	Candida glabrata
Candida albicans	1	Candida famata	Candida albicans	Candida albicans	Candida albicans
Candida tropicalis	1	Candida famata	Candida tropicalis	Candida tropicalis/glabrata	Candida tropicalis
Candida tropicalis	1	Candida famata	∗ Candida tropicalis	Candida tropicalis	Candida tropicalis
Candida albicans	2	Candida glabrata	Candida albicans	Candida albicans	Candida albicans
Candida glabrata	1	Candida lusitaniae	Candida glabrata	Candida glabrata	Candida glabrata
Candida tropicalis	2	Candida parapsilosis	Candida tropicalis	Candida tropicalis	Candida tropicalis
Candida glabrata	1	Candida rugose	Candida glabrata	Candida glabrata	Candida glabrata
Candida kefyr	1	Candida sphaerica	Candida kefyr	Candida kefyr	Candida kefyr
Candida orthopsilosis	1	Candida tropicalis	∗ Candida orthopsilosis	Candida parapsilosis	Candida orthopsilosis
Candida parapsilosis	2	Candida spp.	∗ Candida parapsilosis	Candida parapsilosis	Candida parapsilosis
Candida orthopsilosis	1	Candida spp.	∗ Candida orthopsilosis	Candida parapsilosis	Candida orthopsilosis
Meyerozyma caribbica	1	Candida spp.	∗ Meyerozyma caribbica	Candida guilliermondii	Candida guilliermondii
Candida albicans	1	Stephanoascus ciferrii	Candida albicans	Candida albicans	Candida albicans
Candida krusei	1	No identification	Candida krusei	Candida krusei	Candida krusei
Candida albicans	1	Candida albicans	Candida albicans	Candida albicans	Candida tropicalis
Candida albicans	1	Candida albicans	Candida albicans	Candida albicans	Candida glabrata
Candida parapsilosis	1	Candida parapsilosis	∗ Candida parapsilosis	Candida parapsilosis	Candida orthopsilosis
Aspergillus brasiliensis	1	Aspergillus brasiliensis	Aspergillus brasiliensis	No identification	Aspergillus brasiliensis

Molecular identification (16S or ITS1/2 sequencing).

Identification of microorganisms using a MALDI-TOF system is based on prerecorded protein spectra that are present in the system library [19]. These spectra are mainly based on ribosomal proteins and therefore MALDI-TOF systems have intrinsic limitations to differentiate closely related species or strains such as Salmonella spp. In our study the number of identification on the genus only level was 19 (1.61%) for VITEK MS and 28 (2.37%) for Bruker Microflex LT, mainly due to this systems' limitations. Genus level, but not to species level, identification of Salmonella spp. (n = 15) accounts for the main decrease in the percentage for species level correct identification for both systems.

When we compared the identification accuracies of the two systems we have seen similar results. However we have observed a relatively higher number of correct genera but wrong species count for the Bruker Microflex LT especially with closely related genera or complex members. VITEK MS prefer to give multiple results as species complexes leading to fewer errors while dealing with close species like Streptococcus parasanguinis/mitis/oralis, Enterobacter cloacae/asburiae, Citrobacter freundii/youngae, Aeromonas hydrophila/caviae, and Burkholderia cepacia/vietnamiensis. However, Bruker Microflex LT tends to report single species and that was the reason of correct genus but wrong species results in 8 cases. This grouping style favors VITEK MS during the analysis of clinical samples because distinguishing these closely related genera is clinically irrelevant; however, it can cause certain limitations for research studies and for the analysis of environmental, food, and industrial samples.

Two Corynebacterium spp. (C. ureicelerivorans and C. appendicis) isolates confirmed with 16S rDNA sequencing could only be correctly identified to the genus level by both systems. Both systems identified this isolate as C. pseudodiphtericum. In general, these are rare clinical isolates where molecular identification is needed mainly due to the MALDI-TOF MS library limitations [20]. One C. ureicelerivorans and one Actinomyces naeslundii isolates, which could not be identified by VITEK MS, were identified to the genus only level by Bruker Microflex LT. Two Staphylococcus species S. simulans and S. pettenkoferi were identified to the genus level by VITEK MS; however, Bruker Microflex LT could identify both of them to the species level. On the other hand, one Myroides odoratimimus isolate could only be identified to genus level by Bruker Microflex LT, whereas it was identified to species level by VITEK MS.

VITEK MS had a relatively higher no-identification rate compared to Bruker Microflex LT in the Coryneforms and other Gram-positive bacilli group (Table 5). It has already been shown that probably due to the thick peptidoglycan layer, as it might interfere with the laser ionization, better results had been obtained with a pretreatment prior to MS run for Gram-positive bacteria [21].

Table 5

Results summary of MALDI-TOF MS systems for the microorganisms that either were misidentified or could not be identified.

Microorganism groups	Number of isolates	Species content	Misidentification		No-identification
			VITEK	Bruker	VITEK	Bruker
Nonfermentatives
Acinetobacter baumannii	194		—	—	2 (F × 2)	2 (F × 2)
Aeromonas sobria	1		—	—	—	1 (F)
Total	356	16	2	—	2	3
Enterobacteriaceae
Kluyvera cryocrescens	1		—	—	1 (G)	1 (G)
Shigella boydii	1		1 (N/A)	1 (N/A)	—	—
Total	309	23	1	1	1	1
Staphylococci
Staphylococcus epidermidis	68		—	2 (C × 2)	—	1 (F)
Total	273	10	—	2	—	1
Streptococci
Leuconostoc mesenteroides	1		—	1 (F)	1 (F)	—
Total	29	11	—	1	1	—
Enterococci and other Gr(+) cocci
Total	160	5	—	—	—	—
Coryneforms and other Gr(+) bacilli
Actinomyces naeslundii	1		—	—	1 (F)	—
Corynebacterium coyleae	1		1 (D)	—	—	—
Corynebacterium glaucum	1		—	—	1 (G)	1 (G)
Corynebacterium mucifaciens	3		1 (D)	—	2 (F)	—
Corynebacterium ureicelerivorans	2		—	—	1 (F)	—
Total	30	14	2	—	5	1
HACEK and other Gr(−) cocci
Total	18	8	—	—	—	—
Anaerobes
Total	2	2	—	—	—	—
Others
Wautersiella falsenii	1		—	—	1 (G)	1 (G)
Total	4	4	—	—	1 (G)	1 (G)
Total (bacteria)	1,181	93	5	4	10	7
Fungi
Aspergillus brasiliensis	1		—	—	1 (F)	—
Total (fungi)	160	14	—	—	1	—
Total (bacteria + fungi)	1,341	107	5	4	11	7

Among the 3 species that could not be identified by both systems (Table 4) W. falsenii was not included in the VITEK MS library both in genus and in species level. C. glaucum was represented on the genus level but not on the species level in the VITEK MS library. K. cryocrescens, on the other hand, was already in the library. As we had only one isolate for the MALDI-TOF analysis, it could have been a strain specific issue. Bruker had spectra for all these three species in its library. W. falsenii and C. glaucum are represented with 2 spectra, whereas K. cryocrescens is represented with only 1 spectrum in Bruker BioTyper library. It is likely that this low number of spectral representation in the library was the reason for no identification. In the upcoming updates once their spectral representation is increased or included in the library, we believe both systems will perform better in the identification.

Evaluating all results together, we have observed that most of the species with incorrect and/or no identifications were not common microorganisms encountered in clinical settings and also they are either not represented in systems' libraries or represented with low number of spectra. We believe that these issues will be solved once systems' libraries are updated in the near future.

VITEK MS' species complex approach (reporting the closely related species together), which is a guaranteed attitude, increases the rate of correct identification and it is an advantage for the identifications from the routine clinical isolates. Bruker Microflex LT, however, confidently acts and reports single species, which can provide an advantage in research studies as well as in analyses of environmental, food, and industrial samples. However, consequently, both systems correctly identified 97% of routine clinical isolates making them reliable tools for clinical microbiology laboratories.

Comparing final confirmed identifications, MALDI-TOF MS systems showed a better identification accuracy compared to phenotypic identification. In addition to a better accuracy, a faster workflow and considerable lower cost provide clinical laboratories with a better identification tool [22, 23].