Laboratory advances for infection control

Laboratory advances for infection control

How do the contributions of laboratory advances impact infection control?

Background

Two major roles of the clinical microbiology laboratory are to detect and accurately identify organisms from clinical specimens and to provide, where appropriate, accurate antimicrobial susceptibility results. These major laboratory roles form the foundation for evaluating any recent advance in testing by the clinical microbiology laboratory and their respective impact on infection control.

Laboratory advances that impact infection control have been made primarily in three major areas, those being antibiotic susceptibility testing, detection and identification of infectious disease pathogens, and finally, genotyping methods. Through implementation of these laboratory advances, the laboratory is better able to be more effective in supporting the infection control program. Some of these advances have directly impacted protocols that fill the previously mentioned major laboratory roles. It is beyond the scope of this chapter to review every single advance.

Antimicrobial susceptibilities

New antimicrobial resistances continue to emerge, many of which have appeared in nosocomial pathogens. A challenge with many of these newly emerging resistance mechanisms for laboratories is the failure of automated systems used for identification and antimicrobial susceptibility testing (AST) to detect some of these mechanisms in clinical isolates; particular challenges are those isolates displaying hetero- or inducible resistance mechanisms or newly emerging resistances. Expansion of laboratory capabilities to better detect and monitor for new and emerging patterns of antimicrobial resistance acquired by microorganisms not detected by conventional methods (manual or automated instrumentation).

Delineation of the specific drug mechanism is essential for patient management, particularly as organisms exhibit multiple drug resistance often leaving few treatment options; avoiding the exclusion of potentially useful drugs from consideration is of paramount importance. For example, an Enterobacter species with derepression of AmpC appears to correlate with increased MICs for ertapenem but not imipenem, thus, imipenem could still be useful for therapy. However, imipenem would not be indicated for therapy caused by an Enterobacter species expressing a Klebsiella pneumoniae carbapenamase (KPC). This has become an incredible challenge for laboratories to accurately identify the specific drug resistance mechanisms due to the ever-increasing number of beta lactamase enzymes being described along with an increasingly evident corresponding diversity, both biochemically and genetically.

Molecular assays

Development and introduction of molecular assays that have excellent sensitivity and specificity with a significant decreased turn-around time for results to directly detect -resistant organisms such as methicillin-resistant Staphylococcus aureus(MRSA), and multiply or extensively drug-resistant Mycobacterium tuberculosis, and Klebsiella pneumoniae. Of note, a commercially available molecular assay, the GeneXPERT Carb-R (Cepheid, Sunnyvale CA), has recently become available. This rapid PCR assay detects and differentiates the most prevalent carbapenemase gene families (KPC, NDM, VIM, IMP-1 and OXA-48, and also covering OXA-181 and OXA- 232). With results available in about an hour, epidemiological measures to control the spread of these organisms in the hospital environment could be enhanced.

Phenotypic methods

Development and introduction of phenotypic methods to more easily and better detect drug resistance such as the use of chromogenic agars for direct detection of extended spectrum beta lactamases (ESBLs), MRSA, gram-negative organisms with reduced susceptibility to carbapenems, and vancomycin-resistant enterococci (VRE). Chromogenic agars are rapid, culture-based tests that can provide an alternative to molecular testing in terms of cost, need for technical expertise, and equipment. Direct identification of the organism from primary culture is possible within 24 to 48 hr.

Detection and identification of pathogens

The last decade has witnessed the introduction of diagnostic assays using non-molecular- or molecular-based methods into clinical microbiology laboratory practice that have significantly decreased the turn-around time (TAT) for results and/or expanded the spectrum of etiologic agents of infectious disease that can be detected and/or identified. A brief summary of some of these advances are described below:

Non-molecular based approaches

• Immunochemical assays

  Enzyme immunoassays (EIAs). These assays are able to directly detect antigens (glutamate dehydrogenase [GDH] or toxins (A and B) of Clostridium difficile in stool; influenza A, influenza B, and respiratory syncytial virus (RSV) in nasal aspirates, nasal washes or nasopharyngeal swabs; and Campylobacter jejuni and Camplylobacter coli in stool. Most EIA results are available in about 4 hours.

  Immunochromatographic and immunoblot assays. E.g., rapid qualitative immunochromatographic assay to directly detect soluble antigen of Legionella pneumophila serogroup 1 in urine, rapid HIV-1; results of these assays are available in approximately 15 minutes.

• Miscellaneous

  Chromogenic agars. In addition to chromogenic agars previously mentioned with respect to direct detection of drug-resistant nosocomial pathogens, other chromogenic agar are available that detect Salmonella spp. directly from stool specimens and differentiate the most common Candida spp.

  Viral cell culture. R-Mix Culture and D³ Direct Immunofluorescent staining (Quidel; San Diego, CA): the combination of hybrid shell vial cell cultures combined with specific virus monoclonal antibodies significantly decrease TAT for a variety of viruses such as enterovirus, varicella virus, upper respiratory tract viruses. Most results are available within 1 to 3 days.

  Rapid direct immunofluorescent antibody staining. Introduction of the D³ Fast Point L-DFA Respiratory Virus Kit (Quidel) can directly detect eight major respiratory viral pathogens in a clinical specimen in a single 15-minute incubation step.

Molecular assays

Employing molecular-based assays, culture confirmation or direct detection of microorganisms is more rapid, sensitive, and specific than non-molecular based methods. In addition, as a result of using these various methods, the spectrum of agents the laboratory is able to detect and identify has greatly expanded. Without molecular-based assays, targets would be missed in testing for surveillance and outbreak investigations. As more nosocomial pathogens are identified, there will be a need to expand surveillance and further evaluate new technologies and automation beyond currently-available diagnostics to address detection of a broad range of potential pathogens.

• Peptide nucleic acid (PNA) FISH. Several PNA FISH assays for use on blood cultures that signal positive and have a particular morphotype present have become commercially available (AdvanDx, Woburn Mass). Results are available within 1.5 to 2.5 hr. Examples of single-, two-, or three-color PNA FISH assays that are FDA cleared include the identification of Escherichia coli and Pseudomonas aeruginosa, or E. coli, K. pneumoniae and/or P. aeruginosa on smears from positive cultures containing gram-negative rods, as well as S. aureus alone, coagulase-negative staphylococci and S. aureus, Enterococcus faecalis and other enterococci on smears from positive blood cultures containing gram-positive cocci.

Finally, detection of Candida albicans alone or C. albicans and Candida glabrata, and a three-color assay that discriminates between Candida spp. that are likely to be susceptible to fluconazole with C. albicans and Candida parapsilosis highlighted in green, Candida tropicalis in yellow and those that either may not be or are innately resistant to fluconazole in red, C. glabrata and Candida krusei from positive blood cultures containing yeast.

• Real time PCR technologies. The introduction of ’real time’ amplification assays has decreased TAT from conventional PCR methods for detecting and differentiating nosocomial pathogens. Most of these assays use PCR as the base chemistry, but many other forms of nucleic amplification are also available in a real-time format. With real-time PCR, small, automated, commercially available instruments combine target nucleic acid amplification with qualitative or quantitative measurement of amplified product using fluorescently labeled probes as the hybrids form (detection of amplicon real time). Risk of amplicon contamination is significantly decreased since amplification and detection is in a closed system, thereby decreasing the possibility of false-positive results.

  Monoplex PCR assays are usually qualitative, namely an expected organism is present or absent in a clinical specimen. These assays are most advantageous for those infectious agents that are either fastidious, non-culturable, and/or slow growing such as Mycobacterium tuberculosis, Coxiella burnetti or Legionella pneumophila, and C. difficile.

  Multiparametric detection of disease-related determinants has also been introduced, in which amplification and subsequent identification of multiple microorganisms is accomplished. An example of this type of format is the diagnosis of bloodstream infections in which 40 different bacterial fungal pathogens are identified from whole blood. Another such format is the diagnosis of respiratory tract pathogens. Examples of two commercially available assays include one platform in which influenza A, influenza A subtype H1, influenza A subtype H3, influenza B, rhinovirus, human metapneumovirus, parainfluenza viruses 1 through 3, adenovirus, respiratory syncytial virus (RSV) are directly detected and identified from clinical specimens and is completed in approximately 6 hours; the other platform detects 3 viruses, influenza A, influenza B, and RSV, and is completed in approximately 2 hours.

  Loop-mediated isothermal amplification (LAMP) is a novel amplification method for direct detection of microorganisms in clinical specimens. This technology amplifies nucleic acid under isothermal conditions in the vicinity of 65^oC using DNA polymerase with auto-cycling strand displacement activity. LAMP has numerous attributes that makes this technology attractive to clinical microbiology laboratories and they are, no thermal cycler is required and assays evaluated to date have shown ease of use, flexibility in terms of single or batch processing (10 samples can be run in an hour), reproducibility, high sensitivity and specificity, and less cost than other amplification formats.

An assay using LAMP technology to detect C. difficile directly in stool is now commercially available both inside and outside the U.S. (Illumigene, Meridian Biosciences, Corp, Cincinnati, OH) and a LAMP assay developed by Eiken Chemical Co., Ltd., Tochigi, Japan, is currently under evaluation in resource-limited countries for the direct detection of M. tuberculosis in pulmonary specimens.

• Array-based technologies

  Amplification followed by reverse hybridization (line probe assays). Commercially available NAA assays manufactured by Innogenetics NV Zwijndrecht (Belgium) and Hain Lifescience GmbH (Nehren, Germany) employ amplification followed by reverse hybridization of amplicon(s) to immobilized, membrane-bound probes; these assays are referred to as line probe assays (LPAs). Initially, these LPAs were able to identify clinical isolates of M. tuberculosis complex as well as the more commonly isolated nontuberculous mycobacteria or detect direct directly these organisms in clinical specimens. Subsequently, LPAs have been introduced by these companies that are able to identify M. tuberculosis complex along with resistance to isoniazid (INH) and/or rifampin in clinical isolates or directly from sputum. In addition, a number of LPAs for VRE, MRSA, C. difficile toxin A/B, ESBLs, and metallo-beta lactamases are also available.

  Microarrays. This method relies upon the hybridization of a fluorescent-labeled nucleic acid target, usually an amplicon, to large sets of oligonucleotides synthesized at precise locations on a miniaturized glass substrate or “chip”. These arrays can vary from low density arrays carrying a few hundred to a thousand probes, to high density arrays containing tens of thousands to millions of probes. The use of microarrays in diagnostic applications for microbiology is rapidly expanding with commercially available microarrays now used to identify a wide range of pathogens (bacteria, parasites, viruses, fungi) from a variety of clinical specimens, virulence markers, and/or drug resistance genes; these tools have the potential to accomplish these tasks without the need for culture.

Of note, a microarray containing oligonucleotides representing different drug resistance mutations (e.g., the hundreds of different ESBL variants in tem, oxa, shv, and ctx-m genes) will then allow for more drug resistant determinants to be assigned to the same organism at one time, i.e., genotyped, and the obtained information can be used for epidemiological surveillance. Verigene™ Respiratory Pathogens FLEX Test (Luminex; Austin, TX), a microarray-based detection system utilizes reverse transcription, PCR, and microarray hybridization to detect gene sequences of the following organism types and subtypes: adenovirus, human metapneumovirus, influenza A, influenza A (subtype H1), influenza A (subtype H3), influenza B, parainfluenza 1-4, respiratory syncytial virus A and B, rhinovirus, and Bordetella parapertussis/bronchiseptica, Bordetella holmesii, and Bordetella pertussis. The system consists of a random access, multifunction test processor, reader and a single-use, self-contained test unit or cartridge. Of interest, one only pays for the targets used. See Figure 1 for a Verigene test cartridge. Similar approaches by this manufacturer and others have assays to detect and identify gastrointestinal pathogens, central nervous system pathogens, bloodstream organisms including gram-negative and gram-positive organisms and various antibiotic resistance genes.

Figure 1.

• Sequence-based technologies

  Amplification using ’universal’ or broad-based primers (e.g., rpoB gene, 16S rRNA gene sequences) in conjunction with Sanger-based, automated sequencing technology coupled with curated, commercially available sequence databases has not only decreased the time for pathogen identification, but has allowed for more accurate identification of microorganisms, in particular bacteria and fungi.

  Amplification in conjunction with pyrosequencing. Pyrosequencing is a method of DNA sequencing based on the “sequencing by synthesis” principle. It differs from Sanger sequencing, in that it relies on the detection of pyrophosphate release on nucleotide incorporation. To date, identification of mycobacteria using pyrosequencing has been accomplished using vendor-developed primers targeting a hypervariable region within the Mycobacterium 16S rRNA gene (Biotage, Uppsala, Sweden). In one recent study, pyrosequencing was able to identify 98% of the mycobacterial isolates tested including a number of different Nocardia species as well as when compared to Sanger sequencing.

• Mass spectrometry. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has now been introduced for identification of bacteria. Essentially, a spectral profile of abundant bacterial proteins are generated, the majority of which are likely ribosomal. The profile is then referred to a compiled database in which the protein profiles of various bacteria can be compared and differentiated. Since this technology does not require lengthy biochemical reactions, it is faster than traditional methods for identification, requires less expertise, and costs are significantly reduced compared to commonly used methods such as sequencing.

• Next generation sequencing (NGS). NGS, also referred to as deep, high-throughput sequencing have gone well beyond the traditional dideoxynucleoside chain termination method. There are several platforms commercially available that differ in their sequencing chemistries, read length and throughput capabilities; over the last several years, investment costs and costs for performing NGS have decreased significantly with remarkably rapid throughput. There are a number of applications of NGS such as identification of microbial pathogens in which conventional diagnostics have been unsuccessful, characterization and surveillance of pathogens, taxonomy, outbreak investigation and genotyping (discussed below), and analysis of microbial communities in human diseases to identify correlative or causative relationships (metagenomics) such as inflammatory bowel disease, cystic fibrosis, etc.

Epidemiological genotyping

Although restriction fragment length polymorphism (RFLP) remains the gold standard, alternative methods, many that are PCR-based, have been developed because of its slowness and expense. Multilocus sequence typing (MLST) and repetitive-element based PCR (rep-PCR) are such examples of PCR-based methods. Although these assays are faster, these methods are also costly and labor-intensive. The choice of typing method depends on the indication for typing, the proportion of isolates that are typable and the reproducibility, discriminatory power, ease of use and interpretation, and cost. The rapidly expanding number of sequenced microbial genomes has allowed for the development of a variety of molecular typing approaches that focus on either single or multiple chromosomal loci. These new approaches using sequence-based molecular epidemiology are attractive since they offer reproducible typing profiles that are amenable to standardization, uniform interpretation, and database cataloging since the databases would be based on quaternary data, i.e., A, T, G and C.

Single locus sequence typing

Single locus sequence typing (SLST) is based on the revelation that sequence data for specific loci such as genes for virulence or drug resistance from different strains of the same species show variability in a specific gene, such as single-nucleotide polymorphisms and areas with repetitive sequences. At this time, this approach holds much promise for throughout, ease of use, and interpretation, however, more research and evaluations must be undertaken.

Electrospray ionization-mass spectrometry (ESI-MS)

This technology, which couples PCR to mass spectrometry, offers advantages over routine single-target and multiplex PCR in that it is a full bioinformatics sequence analysis system. This approach is rapid and has high-throughput. Published studies indicate this method’s suitability for not only detection of microorganisms in clinical specimens but for surveillance and infection control interventions as well.

NGS

Studies have demonstrated the use of NGS technology for hospital infection control surveillance programs and community outbreak investigations with respect to environmental sources and food-borne pathogens as well as human-to-human infections. This advance in large part is due to the ability of NGS platforms to generate the large volume of data needed for single nucleotide polymorphisms (SNPs) or variant analysis. For example, by performing whole genome sequencing (WGS) using NGS, an outbreak of hemolytic-uremic syndrome caused by an unusual strain of Escherchia coli was rapidly identified. In addition, other outbreak investigations using WGS for tuberculosis, extended-spectrum-beta-lactamase-producing E. coli, neonatal MRSA, and Legionella have been reported.

What elements of laboratory advances for infection control are necessary for prevention and control?

Clinical microbiology laboratory director

It is incumbent upon the clinical microbiology laboratory director to be well-versed in:

• Emerging new pathogens and corresponding methods for laboratory diagnosis

• The trends in frequency of known pathogens, including those well recognized as nosocomial pathogens

• Emerging patterns of antimicrobial resistance and the current literature regarding the ability of automated systems to detect emerging resistance including the need for any additional testing to detect new resistance

• Both the advantages and limitations of all laboratory methods including newly introduced methods as well as conventional methods. Actively involved in the infection control program.

• A thorough understanding of the:

  Assay parameters (including procedural limitations) for interpretation of laboratory results

  Target organism’s basic microbiology and pathogenesis

• Provide easy access to high-quality and timely data, including annual antibiograms

• Provide guidance and support on how to use its resources for epidemiologic purposes

• Monitor laboratory results for unusual findings: e.g., clusters of pathogens that may indicate an outbreak, emergence of multidrug resistant organisms, a new drug resistant pattern, and the isolation of highly infectious, unusual, or virulent pathogens

• Independently evaluate the automated AST systems on an ongoing basis. Such oversight and evaluations have encouraged the updating of product software and periodic issuance of manufacturer notices recommending alternative testing methods for certain organism-antimicrobial combinations

Good laboratory practices

Good laboratory practices are as follows:

• Quality assurance and control procedures are in place

• Participation in external and internal quality assessment

• Selection of appropriate genotyping methods to support infection control activities based on typability, consistent and reproducible isolate profiles that are highly amenable to standardization and database cataloguing, ease of performance, and interpretation since various methods can be too specific and the discriminatory power of different methods varies for each organism group

• Validation of laboratory assays, in particular with respect to molecular tests that are now widely used in many clinical laboratories

• Full compliance with the Clinical Laboratory Improvement Amendments (CLIA) of 1988; these CLIA federal regulatory standards were passed by the 100th Congress in 1988 and extensively revised in January 2003, such that the accuracy, reliability and appropriateness of clinical laboratory test results are ensured

• Follow performance standards for antimicrobial susceptibility testing as set forth by the Clinical Laboratory Standards Institute (CLSI) and be well versed and current regarding their latest recommendations for detecting emerging resistance

• Ongoing process of evaluation of laboratory practices, methods and results

Communication between the clinical microbiologist/laboratory and infection control

Good communication of the clinical microbiologist/laboratory with infection control is required.

• Updates of any laboratory changes in method, reporting, etc., that might impact infection control

• Notification of infection control of any unusual laboratory results germane to infection control

• Participation in appropriate committees pertaining to infection control

• Despite the significant advantages and strengths that certain phenotypic and molecular methods offer to infection control in terms of rapid and sensitive detection of infectious agents and resistant organisms, there are limitations and caveats to these assays that must be understood by the clinical microbiologist. It is imperative that this individual communicates this information to infection control practitioners such that they are aware of the issues, and as such are able to interpret laboratory data within the context of infection control practices.

What are the consequences of ignoring laboratory advances for infection control?

Any failure of the laboratory or laboratory director to adhere to any of the elements previously described, in particular, good laboratory practices, could result in an incorrect microbial identification, incorrect susceptibility test results, or incorrect interpretation of laboratory data. Any such failure on behalf of the laboratory could, in turn, potentially severely compromise the goals of the infection control program.

Moreover, limitations intrinsic to laboratory testing can also potentially undermine infection control efforts in controlling the spread of a particular agent and in the generation of reliable surveillance data. Assays with limited sensitivities will miss patients who, in turn, could be ineffectively isolated, thereby leading to further spread of the organism.

Similarly, assays lacking in specificity generate more false-positive cases in patients, who may not only receive inappropriate therapy, but these same patients could potentially be cohorted with patients with the disease, thereby putting these patients at increased risk. Finally, the caveats associated with laboratory assays, i.e., sensitivity and specificity, could undermine the reliability of surveillance data and their conclusions. Further complicating laboratory assays with limited sensitivity or specificity rests with the design of research studies evaluating the assay’s performance. Often, the standard to which the new assay is compared is often in itself flawed such that true positives and true negative results are incorrect. Therefore, selection and use of a new laboratory assay with inaccurate results only leads to additional problems ’downstream’ for the patient and infection control.

Summary of current controversies regarding laboratory advances for infection control.

New laboratory advances

Controversy and concerns abound with many of the recent laboratory advances for those microorganisms that play a significant role as nosocomial pathogens. As it is beyond the scope of this document to summarize these controversies for all potential nosocomial pathogens, the focus will be on two, Clostridium difficile and MRSA. In addition, a similar review and summary of the controversies and systematic studies surrounding the laboratory diagnoses of Mycobacterium tuberculosis has been recently completed. By reviewing these findings for these organisms, problems inherent to published studies and interpretation of data in terms of laboratory advances and the possible consequences upon infection control programs will become more transparent. To accomplish this task, the key conclusions from published studies and the current controversies will be summarized for each, and finally, the respective arguments for each side outlined.

Before each organism is discussed, some general comments regarding new laboratory advances are prerequisite:

• Definitive conclusions. First, drawing definitive conclusions regarding the performance sensitivity, specificity, predictive values and clinical utility of the newer laboratory assays from the myriad of studies is often a daunting task. It is a given that the most significant advantage of these newer advances is their rapid TAT for the direct detection, which may have important implications for patient management and infection control. However, assessing the performance of these tests is problematic based on the current literature. Although new diagnostic tests have been developed and marketed, and despite numerous studies having been published evaluating their performance, there still remain numerous questions as to the value of these assays for the diagnosis of nosocomial pathogens and how best to use these assays.

Systematic reviews and meta-analysis are critical for evidence-based clinical practice; however, these studies are only as good as the quality of the studies that are included in the review and analysis. Through systematic reviews and meta-analysis, it is apparent that primary research on diagnostics is often not sufficiently methodically rigorous. For example, concerns to this effect were raised regarding laboratory diagnostic studies for M. tuberculosis, as well as for Clostridium difficile. As noted by Walsh and McNerney, the value of published studies has been frequently compromised by inadequate study design, poor execution, interpretation of results, as well as assay-specific issues. Results from one study to the next are often difficult to compare because of the different bias in study design, as well as in analysis.

• Lack of definitive reference test. An additional significant problem in the assessment of the new advances in diagnostic assays for nosocomial pathogens lies with the lack of a definitive reference test. Historically, the standard reference test for most pathogens has been culture or other more conventional laboratory assays. However, the ’gold standard’ may fail to detect the pathogen that can be picked up by some of the newer assays, in particular, nucleic acid amplification (NAA) assays; this has been a major issue with incorrectly classifying patients with tuberculosis as false-positives. In contrast, depending on which reference assay is used for C. difficile infection, the interpretation of results of the new assay being evaluated may be completely opposite from one another depending on which ’gold standard’ is used. Also, as is the case for MRSA, M. tuberculosis and C. difficile, it may well be that the true accuracy of commercial NAA assays may be actually higher than reported when using an imperfect gold standard.

• NAA assays. Initially, most NAA assays were evaluated as screening rather than diagnostic tests, and sensitivities, specificities, positive predictive values (PPV), and negative predictive values (NPV) were determined. However, to fully appreciate the accuracy of a test, the prevalence of disease in the test setting must be considered, in other words, the actual accuracy of an NAA assay depends on how common the disease is in the population being tested. Thus, the clinical value of these assays depends largely on their PPVs and NPVs, and these vary considerably with the pretest probability. However, few studies have evaluated NAA assays in the context of clinical risk assessment and pretest probability.

• Molecular assays. Finally, there are common concerns for all molecular assays for the diagnosis of infectious diseases, including inhibition of amplification from various clinical specimen types, false-positive and false-negative results, ease of performance, sampling error, cost and cost-effectiveness, and clinical utility. As studies are published that correlate clinical findings with results of molecular assays, it has become exceedingly evident that although molecular assays enhance diagnostic capabilities, their results must be clearly interpreted within the clinical context and performance of the laboratory assay. In addition, caution must be exercised with any molecular result in terms of phenotype versus genotype. An absolute prerequisite before performing any molecular assay is a thorough understanding of all parameters of the assay. Similar issues also apply to other non-molecular based assays.

Methicillin-resistant Staphylococcus aureus

MRSA surveillance is essential to limit the transmission and prevent future outbreaks. While rapid diagnostics testing has been implemented successfully within healthcare settings, controversy still exists from the laboratory’s perspective as to which method(s) is optimum for direct detection of MRSA.

Background

Clinically relevant methicillin resistance in S. aureus isolates is the result of the acquisition of an alternative penicillin binding protein (PBP2a) encoded by the mecA gene, which has a low affinity for most beta-lactam antibiotics. The mecA gene is carried on a mobile genetic element, staphylococcal chromosomal cassette, SCC_mec. Of significance, the ability of S. aureus to accommodate SCC_mec and/or to functionally integrate PBP2a differs from strain to strain, resulting in a wide range of resistance levels. Isolates with this heterogeneous type of methicillin resistance have subpopulations of cells with different levels of methicillin resistance. Thus, a major problem in detecting MRSA facing clinical laboratories is the variable phenotypic expression of the mecA gene-dependent methicillin resistance.

A wide range of culture methods including chromogenic agars and, more recently, commercially available molecular methods have been used for MRSA screening. Since two FDA-approved, commercially available methods have been more widely evaluated; discussion will focus on the BD GeneOhm Methicillin-Resistant Staphylococcus aureus ACP Assay Amplification Kit (BD GeneOhm MRSA Assay; BD Diagnostics, Sparks, MD), Cepheid’s GeneXpert MRSA. Both assays target similar sequences in MRSA by using multiple forward primers amplifying within the SCC_mec but near its insertion site, a reverse primer at the 3’ end of the orfX junction, and probes to detect the interface of the two genes. This design of primers and probes thereby make these assays able to distinguish between MRSA and mixtures of methicillin-susceptible S. aureus (MSSA) and methicillin-resistant coagulase-negative staphylococci. With respect to detection of the various SCC_mec types, the BD GeneOhm MRSA Assay (most current version) and the GeneXpert MRSA detect SCC_mectypes I through VIII. Finally, TATs for the BD GeneOhm MRSA Assay and GeneXpert MRSA are approximately 2h and 72min, respectively. Another FDA approved assay, the BD Max Staph SR assay (BD Diagnostics) is a fully automated, qualitative assay for direct detection of not only MRSA but S. aureus (MSSA) from nasal swabs. This assay targets the mec right-extremity junction (MREJ), the nuc gene (encodes thermostable S. aureus nuclease), genes for methicillin resistance (mecA/mecC) such that MRSA, MSSA, and methicillin-resistant coagulase-negative staphylococci. This assay has high sensitivity and specificity to detect and identify MRSA strains in ESwab-collected nares samples, moreover in one study the assay detected at least one isolate carrying the mecC gene.

Key conclusions regarding the laboratory advances for detection of MRSA

• The cost-effectiveness of rapid MRSA diagnostics is a concern and has not been adequately addressed in studies, particularly in light of the fact that molecular methods cost approximately five to six times more per test than agar-based methods.

• Particularly in low prevalence settings, false-positive MRSA detection using direct detection without culture may lead to unnecessary efforts to eliminate a patient’s putative MRSA colonization.

• Preliminary results with MALDI-TOF MS to differentiate MRSA from MSSA show lack of reproducibility and sensitivity (results must be culture confirmed); therefore, more optimization and standardization is required.

• Immunological tests for identification of clinical isolates of MRSA from culture using the slide MRSA (bioMerieux), and PBP2’ latex test (Oxoid) have high sensitivity (> 97%) and greater than 99% specificity and the TAT is less than 15 minutes.

• Non-chromogenic, selective media such as mannitol salt agar supplemented with oxacillin have limited sensitivity and specificity and as such, broth enrichment is necessary.

• Chromogenic, cefoxitin-based selective agar media are more sensitive, specific, and have a faster TAT to detect MRSA for screening than non-chromogenic media.

• Extended incubation of chromogenic agars beyond the manufacturer’s suggested length of incubation might increase detection of MRSA but with a corresponding decrease in specificity.

• Although delaying results an additional 18 h, the use of an enrichment broth prior to inoculating chromogenic media significantly increases the detection rate of MRSA in nasal specimens from 13 to about 40%, depending on the commercial manufacturer of the chromogenic agar.

• Chromogenic media show uniformly high specificities but sensitivities tend to vary widely both among media and among studies. (Of note, earlier publications showed better sensitivity because culture results were not compared to MRSA real-time PCR assays’ results but only to other culture-based methods.)

• Molecular assays provide a rapid method for the identification of persons who are not colonized with MRSA. The three FDA-cleared molecular assays, BD GeneOhm MRSA Assay, BD Max Staph SR assay, and the GenXpert MRSA, have comparable assay performance.

• The impact of the newer, rapid laboratory methods on infection control for detection of MRSA in screening specimens remains unclear. To date, data from recent studies regarding this subject have shown a significant heterogeneity that is related to different study designs, study populations, and hospital settings. Further contributing to this heterogeneity is the variability in the design of studies evaluating the performance of culture-based and molecular-based assays: differences such as the reference method(s), test protocol, definition of a positive result, diverse MRSA prevalence, local strain dominance, length of incubation for culture-based methods can all contribute to the heterogeneity among studies. A systematic review and meta-analysis of 10 studies using the primary outcomes of MRSA acquisition rate, and incidence of MRSA bacteremia and surgical site infections (SSIs) suggest that in a hospital setting in which culture screening is applied, there is no evidence to support the use of a rapid test for MRSA to significantly decrease MRSA acquisition rates and SSIs.

Controversies and unanswered questions regarding laboratory methods for screening of MRSA in clinical specimens

• Qualifying statement. Unfortunately, published reports on methods propose a variety of approaches such as different gold standards or comparators when evaluating method performance, discrepancy resolution, and possible bias in results due to the prevalence of a particular local strain, all of which confound the determination of which technique is most effective.

• Screening additional sites such as the throat, groin, axilla wounds for MRSA

  Point: Increases the sensitivity of screens. In one study, screening of throat swabs increased the sensitivity of detection among carriers by almost 26%.

  Counterpoint: Increases screening costs for MRSA by sampling multiple sites and studies have demonstrated that the nares is the most sensitive site for detecting MRSA.

• A positive PCR result must be confirmed with culture to exclude false-positive results.

  Point: The PPV of the BD GeneOhm MRSA Assay and GenXpert MRSA are similar among several studies, ranging from 65.9% to 77% indicating a possible issue in specificity. Spontaneous excision of the SCCmeccassette can occur, creating what is referred to as either mecA dropouts or SCCmec variants in which the genetic target is detected by the PCR but the organism lacks a functional or incomplete mecA gene. Following discrepancy analysis, two studies reported that many false-positives associated with the real-time PCR assays were in fact such dropouts. For example, Farley et al., reported that 22 of 23 false-positive MRSA isolates were found by separate mecA gene PCR to be missing the mecA gene. Of concern is a recent study in 2016 by Proulx et al about reversion from methicillin susceptibility to methicillin resistance in S. aureus during treatment of bacteremia and resulting therapeutic failure. This work underscores the concept that resistant phenotypes are unstable as these investigators showed a modest but unknown portion of strains could revert from highly susceptible to highly resistant during the course of therapy. Unfortunately, current laboratory methods are unable to detect this phenomenon and cannot discriminate between reversion to resistance and superinfection.

  Counterpoint: First, such a strategy would lead to excessive costs and other studies have demonstrated that mecA dropouts are not a prevalent problem. A debate is ongoing regarding the relevance of detecting PCR-positive samples with corresponding negative cultures; besides the phenomenon of possible SCC_mec excisions, low bacterial density below culture’s limit of detection (LoD), or if MRSA DNA is present from non-viable organisms have been offered as other possible explanations.

• Unanswered question: What is the optimum laboratory approach to use to screen patients for nasal carriage of MRSA?

  It must be first noted that the laboratory is only one element in a broad range of measures taken to drive down MRSA infection rates. Thus, depending on the study protocol and the underlying infection control practices, the assessment of the effectiveness of a particular laboratory approach for MRSA screening as to its efficiency for infection control, remains difficult and leads to inconsistent data; this in turn leads to controversy as to the best laboratory methods or approach to use. Second, all laboratory tests have limitations and culture-based methods and amplification-based assays are no exceptions to this rule.

  The significant heterogeneity among published studies due to the large number of reasons previously discussed add to the challenge when assessing which laboratory methods would provide optimal support. Suffice, new approaches to detection of MRSA in clinical specimens continue to be explored such as phage amplification combined with MALDI-TOF, MALDI-TOF with the Alere PBP2 Culture Colony Test, LAMP in combination with a lateral flow dipstick test, and shotgun metagenome sequencing.

  A number of strategies as to the optimum laboratory approach to screen patients for MRSA have been proposed, however, not one is suitable for all laboratories. Interpretation of studies performed within a single population is a challenge and conclusions from these studies cannot necessarily be applied to other populations as local endemic MRSA strains may predominate and thereby skew performance data. Of paramount importance is the requirement for local assessment of the laboratory strategy prior to implementation with interaction with infection control being an essential component. Most laboratory methods provide adequate support; however, recent data demonstrate that broth-enriched culture methods and molecular assays offer the highest assay sensitivity; the answer to which of these approaches is best, remains with the laboratory and infection control.

Clostridium difficile

The incidence of Clostridium difficile has markedly increased, most notably associated with the epidemic spread of PCR ribotype 027 (NAP1). Accurate laboratory diagnosis of Clostridium difficile infection (CDI) is crucial to ensure patients receive appropriate treatment and that correct infection control measures are implemented.

Background

Infection with C. difficile can cause asymptomatic colonization or a spectrum of clinical manifestations ranging from mild diarrhea to severe colitis, the latter often resulting in serious complications, such as pseudomembranous colitis, toxic megacolon, and sepsis. Strains of C. difficile can be toxigenic or nontoxigenic; however, only toxigenic strains produce disease. C. difficile toxins include toxin A (TcdA), an enterotoxin with some cytotoxic effects, and toxin B (TcdB), a potent cytotoxin that also inhibits bowel motility (in some newer strains, the binary toxin, CDT, an actin-specific ADP-ribosyltransferase, is also produced but its role in the pathogenesis of CDI remains controversial); TcdA and TcdB are regarded as the primary virulence factors for C. difficile.

The genes encoding the primary toxins, TcdA and TcdB, are encoded on a large, well-defined chromosomal region called the pathogenicity locus (PaLoc); strains that are non-toxigenic have the PaLoc replaced by a non-coding sequence. C. difficile shows marked variability in the PaLoc region. C. difficile strains can be differentiated using several typing methods including PCR-ribotyping, PFGE, restriction enzyme analysis, and multiple locus variable number tandem-repeats analysis (MLVA).

Although it was first believed that C. difficile strains producing both toxins (A+B+) caused disease, and nontoxinogenic strains (A^–B^–) producing neither toxin did not cause disease, it has become apparent that this organism is a very heterogeneous species and that two types of variant isolates can be differentiated: toxin-gene-variants, defined as variant toxinotypes and characterized by changes in toxin genes TcdA and TcdB, and toxin-production-variants, characterized by producing only TcdB or by the production of binary toxin CDT Currently, there are 27 variant toxinotypes. These variations in PaLoc can not only affect toxin production but can also result in production of toxins with altered properties such as substrate specificity and cytopathic effect; because of their unusual pattern of toxin production, the TcdA^–TcdB⁺ strains were the first variants to be described. Of note, point mutations detected by RFLPs of amplified PCR products are significantly more common in the TcdB gene than in the TcdA gene.

In terms of laboratory diagnostic methods, historically, the cell culture cytotoxicity neutralization assay (CCCNA) was considered the gold standard for the laboratory detection of C. difficile. However, when using optimal culture conditions in combination with a sensitive and specific method for toxin detection, toxigenic culture (TC) is sometimes regarded as the most sensitive method of toxigenic C. difficile detection; again, results are delayed possibly for more than 72h with this approach. As a result of these labor-intensive assays coupled with their long TATs, rapid kits to directly detect bothTcdA and TcdB were developed and made commercially available in a variety of formats including EIA and lateral-flow assays.

Subsequently, an additional assay was introduced that detects the common antigen of C. difficile, glutamate dehydrogenase (GDH); although this assay detects the presence of C. difficile directly in stool, it cannot differentiate toxigenic and nontoxigenic strains. Finally, four molecular-based assays have been FDA-cleared for the direct detection of C. difficile in stool specimens: the BD GeneOhm C. diff Assay, Cepheid’s GeneXpert C. difficile, Prodesse’s proGastro Cd Assay (Hologics, Inc., Marborough, MA) and Illumigene C. difficile Assay (Meridian Bioscience, Cincinnati, OH). Results with these various amplification platforms are available in 1 to 4 hours. In terms of target(s) detected, BD’s and Prodesse’s assays detect the TcdB gene; Cepheid’s the TcdB gene, binary toxin and the TcdC deletion to allow for the presumptive identification of the 027/NAP/B1; and Meridian’s Illumigene detects a conserved region of the TcdA gene.

Again, major challenges are faced by laboratories as to the optimum approach to the laboratory diagnosis of CDI in which the most accurate results are obtained in a timely and cost-effective manner in these times of diminishing resources.

Key conclusions regarding the laboratory advances for detection of Clostridium difficile

• All EIA methods that detect antigens of TcdA and TcdB have been shown to be less than optimal, stand-alone, diagnostic tests for CDI but remain the most commonly used laboratory assays because of their ease of performance, cost, and rapid TAT compared to CCCNA and TC. Unacceptably low PPVs may lead to unnecessary treatment and hinder infection control prevention and outbreak control, and relatively low sensitivity, lead to lack of both appropriate treatment and institution of infection control measures.

• Three or more loose stools per day for at least 1 to 2 days, identifies subjects as at high risk for CDI. Appropriate use of testing for CDI is required, i.e., not submitting formed stool, because colonization with C. difficile is common in hospitalized patients.

• There is little value of repeat testing for C. difficile by enzyme immunoassay or by PCR within a 7-day window.

• Given the different targets and combinations that can be used to detect C. difficile (the organism, GDH, toxins, and toxin genes), the measured incidence of infection will vary according to the laboratory method used.

• If a two-step algorithm is adopted that includes GDH testing as the first negative screening step due to its high NPV, there is no consensus as to the best assay for the second step: EIAs for TcdA and TcdB, CCCNA, or PCR.

• The optimal approach to the detection of CDI from the laboratory perspective remains controversial.

Controversies and unanswered questions regarding laboratory methods for screening of C. difficile in clinical specimens

• Qualifying statement: As previously mentioned, the value of the published studies has been frequently compromised by inadequate study design as well as assay-specific issues. Results from one study to the next are often difficult to compare because of the different bias in study design, as well as in analysis.

• There is an accepted reference assay (’gold standard’) to which results of any new assay are compared to.

  Point: CCCNA has been long considered the reference standard for the laboratory confirmation of CDI with the ability to detect as low as 1 pg. Although studies show that TC detects more positive samples than CCCNA, studies have also demonstrated that a variable proportion of cases are TC negative, but positive by CCCNA. The target for TC is C. difficile organisms that have the potential to produce toxin, while the target for CCCNA is the presence of C. difficile toxins. If a stool specimen is CCCNA-negative but TC-positive, the clinical relevance of detecting C. difficile with the capacity to produce toxin but no actual detectable toxin is not entirely clear; in other words, TC has uncertain specificity for CDI. This is of particular significance since toxin-producing C. difficile is cultured from about 2% of the general population and 7-25% of hospitalized patients, thus, a positive TC may occur in absence of CDI. A recent large study (Planche et al.;12,420 fecal samples) found that detection of free toxin in feces using the cell cytotoxicity assay was associated with increased mortality and concluded that CDI is confirmed not by TC but by CCCNA. These investigators described a new diagnostic category of ‘potential C difficile excretor’ (TC positive but CCCNA negative) which would be used to characterize patients with diarrhea that is probably not due to C difficile infection, but who can cause cross-infection.

  Counterpoint: TC is now accepted as the true gold standard or reference method, since it is more sensitive than CCCNA.

  Counterpoint: Methods of CCCNA vary such as the cell lines and antisera employed for toxin detection, time of specimen storage, and dilution of stool specimens. Similarly, methods of TC vary widely as to whether enrichment procedures are used (heat or alcohol shock) and the composition of the selective media for C. difficile such as cycloserine-cefoxitin fructose agar. In light of the current reference methods, i.e., CCCNA and TC, each method produces different results for diagnosing CDI and as such, more and larger studies to determine the optimal reference standard laboratory method for the diagnosis of CDI are required that include relevant clinical data.

• The development and introduction of a two-step approach to the laboratory diagnosis of CDI using GDH EIAs in conjunction with EIAs for toxins A and B that directly detect antigens in stool.

  Point: Increases the sensitivity for the laboratory detection of CDI.

  Counterpoint: Samples with a positive GDH but a negative confirmatory toxin test may require a third test to resolve discrepant results.

• Unanswered questions regarding laboratory methods for screening for C. difficile in clinical specimens

  Since any amplification assay can only detect the presence of the toxin gene and not that of the toxin, what is the relevance of a positive result? A body of literature suggests that amplification assays lack clinical specificity and as such, inflate CDI rates.

  Will the new commercially available, FDA-cleared amplification assays used as a stand-alone test for CDI prove to be not only sensitive, specific and cost-effective compared to other methods?

  Will the C. difficile strain type, i.e., PCR ribotype, impact the assay or algorithm performance? A result study by Tenover et al., demonstrated that TcdA and TcdB EIAs lacked sensitivity in the detection of C. difficile strains of ribotypes other than 027.

Overview of the important clinical trials, meta-analyses, case control studies, case series, and individual case reports related to laboratory advances for infection control

Methicillin-resistant Staphylococcus aureus

See Table I for a description of the results of a meta-analysis for direct detection of MRSA in nasal swabs.