Role of Matrix-Assisted Laser Desorption Ionization–Time-of-Flight Mass Spectrometry for Species Identification of Acinetobacter Strains

Introduction

According to most recent scientific literature, Acinetobacter species are the second most common nonfermenting gram-negative pathogens isolated from clinical samples after Pseudomonas aeruginosa.^[1] The genus Acinetobacter comprises 30 named species and 9 genomic species. Consequently, more relevant species have been grouped together as Acinetobacter calcoaceticus–Acinetobacter baumannii complex (ABC) (A. baumannii, A. calcoaceticus, Acinetobacter genospecies 3 and A. genospecies 13TU).^[2,3] However, there are other species that may also have clinical significance. They cause wide range of infections such as bacteremia, pneumonia, skin and soft tissue infections, and urinary tract infections (UTIs). Non-baumannii Acinetobacter species are emerging pathogens and have also been isolated from patients with bacteremia, endocarditis, and meningitis.^[4–7]

The currently available phenotypic identification systems cannot precisely distinguish Acinetobacter species. The conventional biochemical methods (CBMs) are tedious and time consuming and still only vaguely divide into groups and are not able to speciate the organisms. Automated phenotypic systems have been increasingly used in many clinical laboratories for identification of the bacteria and susceptibility testing. These systems decrease the labor and time required when compared with that for conventional phenotypic methods. Some of the commonly used automated systems are MicroScan WalkAway 96 Plus, BD Phoenix, and VITEK 2. These systems accurately identify few species but have failed to discriminate within the A. calcoaceticus–ABC.^[8]

Matrix-assisted laser desorption ionization–time-of-flight (MALDI-TOF) mass spectrometry (MS) measures highly abundant (ribosomal) proteins found in all microorganisms, and the characteristic patterns obtained from these proteins are matched to a library to identify an organism reliably and accurately. This method is simple, fast, and cost-effective, requiring only small amounts of samples. However, nonfermenting gram-negative bacilli, including Acinetobacter species, may be misidentified because of an incomplete database.^[9]

Several genotypic methods have been developed to differentiate Acinetobacter species. The 16S ribosomal RNA and RNA polymerase β-subunit (rpoB) gene sequencing approaches have been widely used. There is an abundance of rpoB polymorphisms in these Acinetobacter species; hence, it was suggested to be fairly accurate method for the identification of Acinetobacter species.^[10,11] However, these molecular techniques are unsuitable for routine identification of Acinetobacter species, as they are laborious, expensive, and time consuming.^[12]

Very few Indian studies are available regarding the evaluation of MALDI-TOF MS to identify Acinetobacter species. Therefore, the purpose of the present study was to assess the ability of MALDI-TOF MS to identify Acinetobacter species in a tertiary care center in India.

Materials and Methods

A cross-sectional study was conducted in the department of microbiology at a tertiary care teaching hospital located in Rishikesh, Uttarakhand, after the approval of the Institutional Ethics Committee (letter no. AIIMS/IEC/20/258).

The objectives of this study were as follows:

1. To identify Acinetobacter up to species level using MALDI-TOF MS.

2. To evaluate MALDI-TOF MS in comparison with automated phenotypic methods.

3. To compare the antibiotic susceptibility test results Kirby–Bauer disk diffusion method and MicroScan WalkAway 96 Plus automated identification and antimicrobial susceptibility testing (ID/AST) system.

Results

Discussion

In recent few years, Acinetobacter species has emerged as an important pathogen along with an increasing trend toward drug resistance.

In our study, Acinetobacter infection was found to be more common in males, with male to female ratio of 1.70:1, which is similar to other studies and mainly belonged to 31 to 60 years (56%). A similar trend was noted in study conducted by Cucunawangsih et al where maximum isolates (73.8%) were obtained from age group of 40 to 60 years^[13] and in contrast, Dimple and Nupur reported maximum isolation from the age group of <10 years (22.6%) followed by age group of 41 to 50 years (20.8%).^[14] The trend seen in our study may be because patients in the age group >40 years have more chances of having comorbid conditions making them more prone to bacterial infections such as MDR Acinetobacter infections.

In the present study, most of the cases were from medical wards and specialized/critical care units. Similar findings were seen in study by Dimple and Nupur, Tripathi et al, and Rajmane et al reported higher isolation from intensive care units.^[14–16]

Isolation of majority of Acinetobacter species was from LRTIs (59%), and majority of isolates were obtained from respiratory samples (51%) especially ET aspirates (34%). Several other studies show that Acinetobacter species were mainly obtained from respiratory samples.^[17,18] Current findings strengthens the evidence that Acinetobacter infections are one of the most common causes of hospital acquired respiratory tract infections, due to their propensity to survive and persist in moist environment such as suction tubes, ventilators for a long period of time and due to MDR to commonly used antibiotics.

Conventionally, isolates which were nonlactose fermenting, oxidase negative, nonmotile, gram-negative coccobacilli, and citrate utilizer are identified as Acinetobacter species. However, the emergence of resistant strains of A. baumannii raised the need for further speciation of the bacteria. The currently available CBMs can only broadly classify them into A. calcoaceticus–ABC and A. lwoffii. The phenotypic techniques which are currently available are insufficient in accurately identifying and differentiating the closely related and clinically important Acinetobacter species. Moreover, it is tedious and time consuming.

Although various molecular methods are available, they are not regularly used in diagnostic laboratories. Automated systems such as MicroScan WalkAway 96 Plus ID/AST system have been introduced to overcome these difficulties, but it has also failed to distinguish Acinetobacter species further. MALDI-TOF MS is a proteomic-based method, which provides rapid species identification of Acinetobacter strains. MALDI-TOF is increasingly used in diagnostic microbiology for the routine identification of bacteria till the species level.^[8]

In our study, we performed a comparative evaluation of CBMs, MALDI-TOF MS and MicroScan WalkAway 96 Plus automated systems for the correct identification of Acinetobacter species. We although found that there was a substantial agreement between CBMs with MALDI-TOF MS (kappa 0.653, p < 0.05) and MicroScan WalkAway 96 Plus automated systems (kappa 0.74, p < 0.05). Majority of A. baumannii identified with biochemical methods were concordant with MALDI-TOF and MicroScan WalkAway systems currently, while other Acinetobacter species had discordant results.

We found a moderate agreement between MALDI-TOF MS and MicroScan WalkAway 96 Plus ID/AST system (kappa 0.519, p < 0.05) for identification of Acinetobacter species. This drop in agreement is due to discordant results between the two systems in identification of not only other Acinetobacter species but also A. baumannii few of which were identified as S. maltophilia in Microscan Walkaway system. Study by Lee et al suggested that the MALDI-TOF is more advantageous than MicroScan WalkAway 96 Plus ID/AST system and VITEK 2 for routine diagnosis in clinical microbiology laboratories. Furthermore, an isolate not primarily identified by MALDI-TOF can be retested by using other automated methods. In addition, the low rate of misidentification obtained by MALDI-TOF is an advantage along with offering more rapid and reliable ID of other bacterial isolates.^[19]

In our study, MALDI-TOF was able to speciate the organism as A. baumannii, A. nosocomialis, A. lwoffii, A. junii, A. johnsonii, and A. tandoii; 81 (81%) of the isolates were identified with high confidence level, 17 (17%) of the isolates were identified with low confidence level, and 2 (2%) were not identified. For these two isolates, the best match among the choices provided was taken as the ID, that is, A. baumannii and A. tandoii, which was similar to the studies by Hsueh et al, Espinal et al, and Šedo et al.^[20–22] Hence, we too infer that MALDI-TOF has an inherent limitation in identifying genus Acinetobacter up to species.

MicroScan WalkAway 96 Plus ID/AST system can identify Acinetobacter species as ABC and A. lwoffii group but cannot speciate it further. In our study, MicroScan WalkAway 96 Plus ID/AST system identified 95% isolates as ABC, 3% isolates as A. lwoffii group, and 2 (2%) isolates as S. maltophilia with low probability ID. Low probability indicates that the chances of the isolate being S. maltophilia is low and it may be an erroneous result. There is a paucity of published studies on performance of MicroScan WalkAway 96 Plus ID/AST system. One study by Hernandez-Duran et al showed that there was 89% concordance of identification by MicroScan WalkAway system with CBMs and VITEK 2 system, as compared with VITEK 2, MicroScan system presented a longer delay in obtaining results and greater difficulty in the correct identification of gram-negative bacteria including Acinetobacter species.^[23]

Common isolates in the present study were A. baumannii (92%) and A. nosocomialis (2%) belonging to ABC (94%). Other species isolated were A. lwoffii (3%), A. junii (1%), A. johnsonii (1%), and A. tandoii (1%). Three A. lwoffii and one each of A. junii, A. johnsonii were obtained from ET aspirate and pus, respectively. Few studies have reported these agents where confirmation has been obtained by gene sequencing.^[6,24] A single A. tandoii was isolated from a urine sample in our study with low confidence, hence further molecular studies are necessary for the identification of species. Karah et al isolated A. tandoii from blood sample by rpoB sequencing, but it was later found to be Acinetobacter baylyi after 16S rDNA gene sequencing suggesting that A. tandoii may not be a pathogen.^[25]

In our study, we compared the antibiotic sensitivity results of Kirby–Bauer disk diffusion method with MicroScan WalkAway 96 Plus ID/AST system. When there was discrepancy in two methods, Kirby–Bauer disk diffusion method's results were taken as reference standard. We found that there was a 100% categorical agreement between both systems with respect to susceptibility of aminoglycosides and cephalosporins. Total of 16 errors were observed. These results were similar to other studies where a substantial number of very major and major errors were also reported.^[18,26,27] Habib Babay et al suggested the due to different carbapenem resistance mechanisms among Acinetobacter species may lead to disparity in detection of resistance by Kirby–Bauer and MicroScan WalkAway 96 Plus system. Hence, a second, independent antimicrobial susceptibility testing method to validate susceptibility results may be used where possible, especially in a setting of critically ill patients with bacterial infection when only automated systems are used.^[5]

MDR Acinetobacter species are defined as resistance to more than two of the following five drug classes: antipseudomonal cephalosporins (ceftazidime or cefepime), antipseudomonal carbapenems (imipenem or meropenem), ampicillin–sulbactam, fluoroquinolones (ciprofloxacin or levofloxacin), and aminoglycosides (gentamicin, tobramycin, or amikacin). XDR Acinetobacter species are defined as MDR Acinetobacter species which show additional resistance to carbapenems. Pan drug-resistant Acinetobacter species are defined as XDR Acinetobacter species that are also resistant to polymyxins and tigecycline.^[28] In the present study, 94% of Acinetobacter species were found to be XDR, 5% were found to be sensitive to all the drugs tested, and one was found to be MDR. This is similar to other studies; however, occasional reports have shown majority of Acinetobacter species to be sensitive to meropenem and amikacin. This difference may be attributed to the variations in geographical distribution of various Acinetobacter species, their sensitivity patterns, and laboratory quality control practices.^{[15,17,18,29]}

Increasing use of colistin for MDR gram-negative bacterial infections has led to the emergence of colistin resistance in several countries worldwide which may vary between regions, with majority studies showing resistance rates of < 10%.^[30] In our study, MicroScan WalkAway 96 Plus system based on broth microdilution (BMD) method was used and the results were interpreted as per CLSI 2019 guidelines.^[31] Out of 100 isolates,99% were found sensitive to colistin and only 1% was found intermediate. Since the joint CLSI and EUCAST Polymyxin Breakpoints Working Group recommend only reference BMD method as the reference test for determining susceptibility to colistin the Kirby–Bauer disk diffusion method was not done.^[32]

Jayol et al stated that there is a high rate of major errors (26.9%) observed with the MicroScan WalkAway 96 Plus ID/AST system when compared with manual BMD, due to an overestimation of the minimum inhibitory concentrations (MICs) for the nonfermenting gram-negative bacilli including Acinetobacter species.^[33] In a study conducted in our institute by Singh et al, categorical agreement between MicroScan WalkAway 96 Plus ID/AST system and Mikrolatest MIC AST kit was found to be 71.4% for A. baumannii and 100% for A. junii and A. johnsonii. Total two major errors for A. baumannii were observed.^[34] This shows that even though automated systems are less tedious, it has its own inherent errors. The automated systems depend on the library of information for their results, and if the organism provided matches another organism in the system, the results will be inconsistent, which has to be confirmed with further testing. Hence, even after advent of various automated systems, the role of microbiologists will not decrease as they are required to interpret the results correctly and as applicable. Furthermore, routine infection control practices have a role in control of MDR and XDR infections as the hospital equipment could serve as a reservoir for Acinetobacter species.

The limitations of the current study were as follows: The gold standard of identification of Acinetobacter species is gene sequencing and the reference test for determining susceptibility in the form of MIC to colistin is BMD. However, these could not be performed due to logistical issues.

While MALDI-TOF identification costs few rupees, other automated systems cost in hundreds and molecular methods cost thousands, which in a developing country like India amount to a huge financial burden to the patients.

Conclusion

In conclusion for the species identification of Acinetobacter strains, MALDI-TOF is a rapid, reliable method and can thus be routinely used in clinical laboratories.

Phenotypic methods are currently used to determine the antibiotic susceptibility of bacterial isolates. Various commercial automated systems are widely used routinely in clinical laboratories. Although these systems are rapid, convenient, and decrease manpower requirement, they have inherent limitations. Hence, standardized techniques and quality control practices have to be followed in these settings.

While it is not prudent to completely automate, we can use these methods along with conventional phenotypic methods of identification and AST. The feedback from the conventional phenotypic methods can be used to further upgrade the automated systems.