Use of MALDI-TOF for Diagnosis of Microbial Infections

Although mass spectrometry is making its mark on all facets of clinical laboratory medicine, arguably no field is witnessing its impact more than clinical microbiology. The application of MALDI-TOF mass spectrometry (MALDI-TOF MS) to microbial identification is revolutionizing clinical microbiology by providing rapid identification with minimal sample preparation at a potential savings in costs. Across the globe, the degree of implementation of MALDI-TOF MS varies markedly. In Canada, Australia, and much of Europe, MALDI platforms are in routine use in clinical microbiology, whereas the US Food and Drug Administration has yet to provide clinical clearance. In this Q&A, 4 experts from across the globe with first-hand experience implementing MALDI-TOF MS in the microbiology laboratory provide insight into what this technology can and cannot provide, what it takes to bring it in house, and what direction it takes us in the future. The application of MALDI-TOF MS to the diagnosis of microbial infections has been touted as a revolution in clinical microbiology. However, no technology is without its pitfalls. Can you please describe what you feel are the greatest strengths and limitations of MALDI-TOF MS? Gilbert Greub: When it is used to identify bacterial strains and fungi, the main strengths of MALDI-TOF MS are the rapidity of the technique (<10 min), its low cost in terms of reagents and technician processing time [<2 Euros (<2.8 US dollars) per identification] and its overall >95% accuracy at the species level. One of the most important limitations of this technique is its relatively low analytical sensitivity (about 105–106 bacteria/well). Thus, the accuracy of the identification is increased when the identification is done on a colony grown on agar or on a blood culture pellet, i.e., after a culture-based amplification step. Consequently, MALDI-TOF MS is not a tool currently suitable to detect a low amount of bacteria potentially present in physiologically sterile samples such as cerebrospinal fluids. Susan Poutanen: The greatest strengths of MALDI-TOF MS include: (1) the fast turnaround time associated with its use for the identification of bacteria and yeast grown on standard culture media; (2) the cost savings in supplies and work load associated with its use compared to traditional identification methods; (3) the ability to incorporate MALDI-TOF MS into robotic automation in the laboratory; (4) the improvement on the work flow of the laboratory associated with having an earlier organism identification; (5) the improvement in patient care and antimicrobial stewardship associated with having an earlier organism identification; and (6) the potential for future applications, such as identification of filamentous fungi, identification of resistance mechanisms such as the production of carbapenemases, and assisting with epidemiologic typing. The greatest limitations of MALDI-TOF MS include: (1) the up-front cost of purchasing a MALDI-TOF MS instrument; (2) lack of a comparable high-speed susceptibility system, which results in a substantial lag between having a reported organism and reporting its associated susceptibility results; and (3) the potential for technologists to lose the skill set associated with identifying organisms by traditional means, which may result in errors when working in laboratories or sections of laboratories not using MALDI-TOF MS. Jens Jørgen Christensen: In combination with speed, low costs, and ease of use, a major advantage of MALDI-TOF MS is that only highly probable identifications are provided by the scoring algorithms. If the system cannot generate an exact identification, no suggestions are provided, and, instead, low score values and warning comments are given, thereby minimizing false identifications. A limitation is that although the device is robust when examining bacterial strains from many different species, closely related species, such as Streptococcus pneumoniae, S. mitis, and S. oralis, may be difficult to separate. Preparation (extraction, growth) of strains, inclusion of species in the database, and the method used for creation of consensus mass spectra may act as both strengths and limitations; further standardization and optimization will be needed. Markus Kostrzew: The greatest strengths of MALDI-TOF MS are its accuracy and speed, enabling a faster correct treatment of patients. Its main limitations, at least currently, include restricted application areas without prior culture and the discrimination of phylogenetically very closely related microorganisms (e.g., Shigella/Escherichia coli). What do you feel are the major considerations (e.g., technical, economic) to include when implementing MALDI-TOF MS in the clinical microbiology lab? What reasons, if any, are there to hesitate in incorporating MALDI-TOF in the clinical laboratory? Gilbert Greub: When a new tool is made available, it is important to consider the advantages and limitations of the technique and, more specifically: (1) when to use it; (2) how to use it; (3) how to interpret results; (4) how to ensure appropriate traceability of results; and (5) the controls and maintenance that are required. Regarding MALDI-TOF MS applied to clinical microbiology, this technology may be used to identify any strains considered to be clinically significant, when isolated colonies are available on agar, or when a blood culture is positive. Despite its high accuracy, interpretation of results obtained with MALDI-TOF MS is essential. This interpretation is largely influenced by the content and quality of the database, as well as by the identification algorithm. The traceability may be largely improved by using an automated colony-picking system and by an automated transfer of MALDI-TOF MS results to the laboratory information system. Due to the cost of the MALDI-TOF MS instruments [approximately 200 000 Euros (280 000 US dollars)] and of their maintenance [approximately 20 000 Euros (28 000 US dollars) per year], clinical microbiology laboratories identifying <5000 strains per year should opt for other, more cost-effective identification approaches. Another major point to consider when implementing MALDI-TOF MS in a routine diagnostic microbiology laboratory is the use of adequate controls, including negative and positive controls as well as a calibration controls. Internal and external controls should also be considered in the future. Susan Poutanen: The major considerations when implementing a MALDI-TOF MS in the clinical laboratory should include: (1) the breadth and accuracy of the database and the ability for laboratories to be able to verify this; (2) the throughput of the instrument; (3) the ease of use of the instrument; (4) the mechanical reliability of the instrument; (5) the ability for the results of the instrument to be interfaced to a laboratory information system; (6) the best way to implement MALDI-TOF MS to maximize work flow efficiency throughout the laboratory; (7) the cost of the instrument; (8) the number of identifications typically performed in the laboratory and the associated length of time before the laboratory would expect to see a cost return on this investment; and (9) the optimal way to report identifications provided by MALDI-TOF MS to limit confusion that may be caused by reporting new species names—for example, key stakeholders should be consulted regarding their preference to have organisms that may have traditionally been reported with group-level identification (e.g., coagulase-negative staphylococci, viridans group streptococci, S. anginosus group) continue to be reported with group-level identification, despite species-level identification being available through MALDI-TOF MS. There are some laboratories that may want to hold off in incorporating MALDI-TOF MS. Specifically, laboratories with small numbers of specimens need to weigh the benefits of MALDI-TOF MS against the limitations and recognize that it will take longer for them to see the cost return of incorporating MALDI-TOF MS, compared to larger laboratories. It may be prudent for these laboratories to wait until the cost of MALDI-TOF MS instruments is reduced, as is expected in future years. Jens Jørgen Christensen: On the assumption of an adequate validation process, economic considerations and laboratory work flows must be carefully examined and a cost–benefit analysis performed. Communication of MALDI-TOF MS data to the clinicians is also of major importance. It will be important to incorporate MALDI-TOF MS data with other major characteristics used for grouping of bacteria. One must be mindful that reporting to clinicians is a question of useful communication—presenting newly published genus and species names must be balanced with the clinical value of such information. Additionally, laboratories must have a policy in place for how to deal with unfamiliar genera and species. Markus Kostrzew: I don't see a reason to hesitate, but a laboratory must plan very well how the technology can be integrated into the general laboratory work flow. Integration into the existing laboratory information system is essential. What do you feel are the most significant contributions that implementation of MALDI-TOF MS can make in the clinical microbiology laboratory? Gilbert Greub: For the past 3 years, we have routinely used MALDI-TOF MS to identify microbial strains from a positive blood culture pellet. This has a major impact on clinical management and represents one of the most significant contributions of MALDI-TOF MS in clinical microbiology. Indeed, the bacterial identification may partly guide the antimicrobial treatment, improving, for example, by about 30% the adequacy of the empirical antibiotic regimen for gram-negative bacteremia. This may have a major impact on morbidity and mortality. MALDI-TOF MS also allows much faster identification of bacterial colonies present on an agar plate than commercial phenotypic systems, such as the VITEK. Susan Poutanen: Given that many clinical microbiology laboratories are faced with an increasing work load and yet decreasing numbers of staff and healthcare dollars, the cost efficiencies of MALDI-TOF MS and potential for its incorporation into automation are the most significant contributions associated with the implementation of this technology into a clinical laboratory. Jens Jørgen Christensen: Implementation of MALDI-TOF MS affords 3 significant contributions. First, there is the rapid identification of cultured bacterial strains and direct identification from positive blood cultures, which provide clinical guidance and permits optimal antibiotic treatment at least 24 h before what was possible with previous techniques. Direct identification of pathogens can be of great significance for the initial treatment of serious invasive infections. Second, with respect to identification of fastidious bacteria, MALDI-TOF MS requires only minute amounts of material, typically a fraction of a colony, thereby eliminating the need to inoculate multiple growth plates. Third, antibiotic-susceptibility testing and resistance-determinant testing are also within reach, which will provide information on, for example, methicillin susceptibility or carbapenemase production. This application, however, requires further investigation and standardization. Markus Kostrzew: The implementation of MALDI-TOF MS in a clinical microbiology laboratory can reduce the work load of the staff by substituting many partially elaborate tests for the majority of isolates. The fast time to result enables the early reporting of microbial identification, which is appreciated by many physicians. When implementing MALDI-TOF MS in the clinical microbiology laboratory, laboratorians have the choice of several instruments, each with their own testing algorithm. What do you feel is the likelihood of establishing a harmonized microbial database and algorithm for identification? Do you believe this would be advantageous? Gilbert Greub: It might be advantageous to have a single harmonized microbial database. However, the most important issue is that users may also add spectrum from well-characterized bacterial strains and species in a common open-source, web-based database. Such addition of new strains to the database should be controlled and validated to avoid the addition of poor-quality spectra that may lead to misidentification. Harmonization of algorithm is not mandatory; on the contrary, it might be of value to implement several algorithms chosen by the end user, since some algorithms may be better suited to applications such as Staphylococcus aureus typing or for the identification of closely related bacteria such as streptococci, whereas others might be ideal for the identification at the species level, of corynebacteria for example. Susan Poutanen: Given the current competitive nature of the manufacturers involved with selling MALDI-TOF MS instruments, I do not see harmonization of instruments or microbial databases a likely possibility, at least in the near future. While harmonization would be advantageous from the point of view of knowing that the same organism will be identified the same way by any MALDI-TOF MS instrument and so a patient's results can readily be compared from one laboratory using MALDI-TOF MS to another, there are also disadvantages. Having a diversity of instruments acts as a buffer reducing the number of laboratories reporting a potential systematic error associated with using a single MALDI-TOF MS. For example, if a systematic misidentification occurs with a specific organism in only one MALDI-TOF MS instrument and if all laboratories used that system, all would make this error. However, the number of laboratories reporting this error would be minimized by having a variety of MALDI-TOF MS instruments/databases used. In addition, if only one MALDI-TOF MS instrument or database were used, it may take longer for systematic misidentifications to be detected by laboratories, since they will potentially see verification of their results from other laboratories reporting the same error and not recognize the error until a later time. Jens Jørgen Christensen: Database and algorithm developments have largely been Having algorithms may be to identify On the other this will also is and the of all of information on microorganisms are a need for incorporating MALDI-TOF MS data into that algorithms must be as as possible to the future of bacterial The future might well be a combination of current algorithms and the of new these developments will be or from the only the future will A combination is Markus Kostrzew: Harmonization across different is not instruments have not only different algorithms but also different spectra that are The spectra are the of any further and the of information from these spectra (e.g., is one of the most in mass spectrometry what do you believe MALDI-TOF MS will be used in the microbiology laboratory? Do you believe it to be competitive or a for current Gilbert Greub: MALDI-TOF MS has been used in laboratory since to routinely identify all isolated strains, about per This has been associated with a in the use of which are used for the identification of only to of that are not identified by the MALDI-TOF MS or that are only identified score or of MALDI-TOF MS result with some characteristics of the the need for and used to identify strains not identified by routine has been by since the of MALDI-TOF MS in laboratory. Thus, MALDI-TOF MS has largely the need of many identification However, despite the accuracy and low cost of MALDI-TOF MS, clinical will use some cost-effective phenotypic such as or as a identification tool or to MALDI-TOF MS regarding other MALDI-TOF MS applications, such as typing and carbapenemase mass spectrometry represents a and faster However, MALDI-TOF MS will not current to limitations of the technique for some of these Susan Poutanen: I believe MALDI-TOF MS will be used in the microbiology laboratory, and I believe it will be competitive and will traditional at least in some areas of microbiology laboratories. In it has in some laboratories. laboratories may not be able to purchasing a MALDI-TOF MS instrument and I that many if not most laboratories will be using this technology in some Jens Jørgen Christensen: MALDI-TOF MS has the potential to phenotypic identification for most bacterial strains examined in clinical microbiology laboratories. The great advantages of being able to for the of bacteria from specimens and of antibiotic-susceptibility testing the in a very Markus Kostrzew: MALDI-TOF MS has testing in many laboratories. There are of laboratories using the technology for identification, partially as the one identification system. A combination of MALDI-TOF MS and may most tests in the future. What do you for quality and standardization when implementing MALDI-TOF MS for microbial Gilbert Greub: negative and positive controls should be to each controls will the of which may be to or the of Indeed, despite of the MALDI about of plate as with will that generate mass spectra to some species. In addition to negative controls, a calibration positive should be on a at least a that will the amount of bacterial on the plate might also be considered in the but it to be also is an external quality that will of the of different diagnostic laboratories. Susan Poutanen: of MALDI-TOF MS should be as would be done for any new traditional identification system. The and of in the published by the for in and by Susan is an to use as initial should be with a such as validation of the ability of the MALDI-TOF MS instrument to identify organisms should be with the use of quality organisms and external quality Jens Jørgen Christensen: of quality have to be for MALDI-TOF MS to ensure quality of This that bacteria from different must be and the results On a laboratories must ensure that the quality of interpretation and reporting to clinicians is This and, in many laboratories, establishing a group of can and provide guidance when needed. Additionally, the quality testing has to on the reporting of results to Although it may be to report suggestions or detect new it is to be in as much confusion can result from identifications and Markus Kostrzew: One should use the quality controls by the controls should be in In a controls should be to the overall can be controlled by negative controls with the use of MALDI-TOF MS in microbiology with current (e.g., ease of use, of ease of Gilbert Greub: MALDI-TOF MS is a for bacterial identification to or better than most of the of Despite the relatively high cost of the instrument and the need for MALDI-TOF MS in microbiology is very being much than phenotypic such as and to both the low and the very low technician time of mass this technology may be in the laboratory, since the instrument is relatively to use and most results are to However, interpretation of MALDI-TOF MS results are very and the should at least if the obtained results to a bacterial species that is expected (1) the growth characteristics or (2) the colony and (3) the or nature of the colony, when The interpretation also the of a species in a as well as the in score values between the best and A of the database content and the algorithm the score of will also the quality of the interpretation of MALDI-TOF MS Susan Poutanen: to traditional the use of MALDI-TOF MS is more is to use, requires is to and is more time and work flow than traditional it that this technology traditional as the routine identification system used in laboratories, at least in some However, this not that traditional are on the instrument and database being used, there may be organism identifications for which MALDI-TOF MS has If this is after a verification of the instrument or is reported in the as a there may be traditional testing that will need to be MALDI-TOF MS to different organisms that may be potentially by the MALDI-TOF MS In addition, there may be work flow identification currently in place in some areas of the laboratory (e.g., the are more to continue than all to MALDI-TOF MS. as a system, should the MALDI-TOF MS instrument some laboratories may to traditional identification available and to use, the cost associated with purchasing a MALDI-TOF MS instrument as a Jens Jørgen Christensen: The relatively low cost for with cost of US dollars) per identification, MALDI-TOF MS competitive compared to existing identification However, is since it is For the routine clinical microbiology laboratory, the ease of use for the it a although it is not without its pitfalls. It is to time in and in interpretation of results and reporting to The need for and must not be Markus Kostrzew: MALDI-TOF MS is to since this technology represents a from the used in traditional microbiology laboratories, is the technology is in interpretation in most The positive economic impact was one of the major for the in the laboratories within laboratories with identifications per see a cost MALDI-TOF MS is to microbial infections from and prior culture of the microbial is required. MALDI-TOF MS is in its ability to identify infections. Do you feel that MALDI-TOF MS will be able to be used for these If what are If Gilbert Greub: most laboratories use MALDI-TOF MS on positive blood culture The analytical sensitivity of MALDI-TOF MS, however, its direct use on clinical samples such as time from of blood to of blood culture an fast amplification or requires further in terms of analytical sensitivity of the MS infections are not identified with current algorithms. However, with improved identification algorithms of of the best identification of bacteria present in infections will be further improved in the future. algorithms into not only the of but also will also the of MALDI-TOF MS. Susan Poutanen: There are data to that direct from at least in specimens associated with significant growth of cultures, is with MALDI-TOF MS without the need for prior which the use of MALDI-TOF MS from and a make this application a requires more work to if should be done to each as a to or other that may with work would also need to be done to the analytical sensitivity of MALDI-TOF MS in organisms in low numbers or in If these direct direct identification from specimens is a potential other which is by Jens Jørgen Christensen: MALDI-TOF MS has been applied to cultured bacteria. However, results have also been when on of clinical In many different to of blood were to the species level. Direct testing of samples have also been with There may be a potential for although data are in sample and analytical may these in the near future. Markus Kostrzew: The direct identification from blood is not since the limit of MALDI-TOF MS, which is working without is For other such as cerebrospinal the direct identification of pathogens might possible when the and the analytical sensitivity of the are For a algorithm is This must be further validated and if The of that can be will be to 3 different microorganisms in a since specific will to the of an organism in a will its to a of in a Do you have any comments or Gilbert Greub: In the the of some or other will likely be possible by mass However, most are not to the relatively of mass 000 detected with current MALDI-TOF MS Susan Poutanen: the same time that MALDI-TOF MS is an technology being to microbiology laboratories, spectrometry is technology that is also being While each has its advantages and them may not be with the of make it an time to be in the field of clinical microbiology. In a relatively of these will likely lead to substantial in traditional microbiology laboratories as we them Jens Jørgen Christensen: One is the use of mass spectrometry for identification, which can be a to the identification Additionally, although the has been on the identification of bacteria, identification for invasive infections are being and in the future we may be able to add these microorganisms to the of the potential in susceptibility testing Markus Kostrzew: MALDI-TOF MS is used for species identification in the clinical microbiology laboratory. developments that there is a very for its to such as of strains, and of MALDI-TOF MS may a technology in the microbiology laboratory.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.