Rapid Diagnostic Testing: Changing the Game for Antimicrobial Stewardship

The clinical microbiology laboratory is experiencing a technological revolution in pathogen identification, antimicrobial susceptibility testing (AST), and resistance detection, which includes the advancement of rapid diagnostic tests (RDTs) for infectious diseases. These tests can provide accurate results in significantly less time than traditional methods.¹

For most suspected infectious cases, an empiric antimicrobial agent will be started while waiting for the microbiological result to confirm or rule out an infectious disease. Administering antimicrobial agents to people who do not need them can lead to severe consequences, such as the development of antimicrobial resistance and unnecessary adverse events to the patient. On the other hand, antimicrobial therapy may be delayed for a serious infection while waiting for the microbiological result in which starting the therapy rapidly is considered a critical factor for improving clinical outcomes in these cases (eg, high-grade bacteremia, central nervous system [CNS] infections).^2-4

To improve health outcomes, more emphasis has been placed on expediting the process of reporting the microbiological results to clinicians so they can make the appropriate decision based on microbiological findings. Implementing new methods that can help identify and test for antimicrobial susceptibility and resistance rapidly and accurately has become increasingly important. Fortunately, RDTs can help providers achieve these goals.

RDTs can rapidly identify many microorganisms (bacteria, fungi, viruses) and detect antimicrobial resistance associated with these pathogens. RDTs use different techniques, such as polymerase chain reaction (PCR), magnetic resonance, mass spectrometry, microarray technology, next-generation sequencing, and others.^5-8

In the last few years, RDTs have expanded their capability to detect many pathogens simultaneously instead of a single pathogen at a time. As a result, RDTs commonly are available as a panel for different indications (eg, respiratory panel, bacteremia panel). In this review, we mainly focus on the role of RDTs in the following areas: respiratory infections, bacteremia, and resistance detection.

Clinical Importance of RDTs for Respiratory Tract Infections

Respiratory tract infections (RTIs) are a major health concern worldwide. RTIs remain one of the most common infections globally and can lead to significant morbidity and mortality.^9-11 The incidence and prevalence of RTIs vary based on certain factors such as socioeconomic status, population density (ie, more dense population associated with a higher risk), and geographic location (ie, areas with a dry climate and cold winters can promote the spread of respiratory infections).^9,10 However, many other patient-specific factors can increase the risk for RTIs, including age (ie, young and old), chronic lung disease, tobacco smoking, and immunocompromised conditions.^12-14 RTIs are divided into upper (eg, influenza, common cold, pharyngitis) and lower (eg, pneumonia, bronchitis, bronchiolitis).¹⁴

The etiology of RTIs typically have been determined through traditional culture methods, which involves the isolation and growth of bacterial colonies on agar plates and subsequent testing for antimicrobial susceptibility. This process usually takes 24 to 48 hours for microorganism identification and another 24 to 48 hours for susceptibility testing.¹⁵ During this period, patients receive empiric antimicrobials that may be suboptimal or overly broad-spectrum, which can lead to the development of antimicrobial resistance and adverse events (eg, Clostridioides difficile infections).^3,4 For this reason, the emergence of novel molecular testing methods, such as multiplex PCR assays, can play a significant role in improving antibiotic use and reducing harmful effects of treatment. A list of selected common RDTs for RTIs is provided in Table 1 for ease of comparison.

Upper respiratory tract infections (URTIs), which are primarily caused by viruses including rhinovirus, coronavirus, and respiratory syncytial virus, are among the top diagnoses in the outpatient setting and account for an estimated 10 million outpatient visits per year in the United States.¹⁶ Because the etiology is usually viral, extreme caution should be taken to avoid inappropriate use of antibiotics.¹⁷ It is estimated that fewer than 10% of URTIs are attributed to bacteria.¹⁸ However, the World Health Organization (WHO) estimated that the number of antibiotic prescriptions for URTIs increased significantly between 1992 and 2006.¹⁹ The rising trend may be attributable, in part, to slow turnaround times for results associated with traditional diagnostic methods. However, the increased availability of RDTs may help to improve the appropriateness of antibiotics for URTIs.

Common rapid diagnostic testing panels designed for URTIs include VERIGENE Respiratory Pathogens Flex Test, NxTAG Respiratory Pathogen Panel, BioFire Respiratory 2.1 (RP2.1) Panel, SPOTFIRE Respiratory Panel, ePlex Respiratory Pathogen Panel, QIAstat-Dx Respiratory Panel, and the BioCode Respiratory Pathogens Panel (Table 1). These revolutionary tests can support antimicrobial stewardship program (ASP) measures by decreasing the use of antibiotics for URTIs.²⁰ RDTs can detect common viral pathogens associated with URTIs such as influenza A and B; parainfluenza 1, 2, 3, and 4; RSV A and B; adenovirus; rhinovirus; and metapneumovirus. However, panels differ in the technology used, result turnaround time, and their ability to detect additional pathogens (eg, atypical bacteria, Bordetella species, coronavirus, SARS-CoV-2).^21-25 For example, the ePlex Respiratory Pathogen Panel and NxTAG Respiratory Pathogen Panel cannot identify Bordetella species.^26,27 Moreover, the ePlex Respiratory Pathogen Panel cannot identify atypical bacteria (Mycoplasma pneumoniae, Chlamydia pneumoniae), while the NxTAG Respiratory Pathogen Panel is the only product listed that can identify bocavirus.^28–31 The BioFire RP2.1 Panel was the first FDA-cleared panel that could identify SARS-CoV-2. The RDTs also differ on the speed of sample processing and result reporting. The result turnaround time can range from 15 minutes (ie, SPOTFIRE Respiratory Panel) to 5 to 7 hours (ie, NxTAG Respiratory Pathogen Panel). Table 1 shows a list and comparison of selected common RDTs designed for RTIs.

Despite the variation among these products regarding the turnaround time, all can provide results hours to days faster than traditional diagnostic methods. In addition to the fast turnaround time, these RDT panels show high sensitivity and specificity compared with traditional laboratory-developed, real-time PCR assays. For example, the BioFire RP2.1 Panel showed a sensitivity of 97.1% and a specificity of 99.3%, and the NxTAG Respiratory Pathogen Panel showed a sensitivity of 95.2% and a specificity of 99.6%.^23,32 A retrospective study by Nijhuis et al compared the ePlex Respiratory Pathogen Panel with laboratory-developed, real-time PCR assays for the detection of 464 respiratory pathogens from 323 positive clinical specimens. Results showed that the ePlex panel had 97.4% agreement with the real-time PCR assay. Furthermore, the panel detected 17 more pathogens than the laboratory-developed, real-time PCR assays.³³ Another study compared the performance of the ePlex Respiratory Pathogen Panel with the BioFire RP2.1 for the detection of viral and bacterial respiratory tract pathogens in nasopharyngeal swabs. The results demonstrated overall agreement between both panels (>95%) for all targets. The overall positive percent agreement (PPA) of the ePlex Respiratory Pathogen Panel compared with the BioFire RP2.1 for viruses ranged from 85.1% (95% CI, 80.2%-88.9%) to 95.1% (95% CI, 89.0%-97.9%), while the negative percent agreement (NPA) ranged from 99.5% (95% CI, 99.1%-99.7%) to 99.8% (95% CI, 99.5%-99.9%).³⁴

All panels mentioned above are FDA-cleared for use with nasopharyngeal swab specimens. They are considered qualitative tests that use multiplex nucleic acid or PCR technology, except for the BioCode Respiratory Pathogens Panel, which uses a bioinformatics approach involving DNA or RNA sequencing techniques. Finally, given their frequent viral etiology, URTI treatment could benefit from RDT implementation to rule out bacterial cause, minimize the overuse of antimicrobial agents, and optimize patient care.

In the United States, pneumonia has been a major contributor to morbidity and mortality, causing about 63,000 deaths and 1.2 million hospitalizations yearly.^35,36 The American Thoracic Society/Infectious Diseases Society of America (ATS/IDSA) recommendations for empiric antimicrobial therapy in community-acquired pneumonia (CAP) are based on selecting agents targeted against the major treatable respiratory bacterial pathogens.³⁷ Unfortunately, overuse of antibiotics is common in lower respiratory tract infections (LRTIs) because of the difficulty in distinguishing between bacterial and viral etiologies due to similar manifestations.³⁸ Antibiotic therapy may be safely withheld in patients with isolated viral pneumonia if these infections can be identified easily from those with concomitant bacterial etiology.³⁹

Multiple syndromic molecular testing panels for LRTIs, including the BioFire FilmArray Pneumonia Panel and Curetis Unyvero Lower Respiratory Tract (LRT) Panel, offer increased sensitivity over traditional cultures and provide the presence of resistance markers in as little as 1 to 5 hours from specimen collection and testing. The BioFire panel offers detection of 8 viruses, 8 resistance genes, 3 atypical bacteria using qualitative targets, and 15 bacterial targets with semiquantitative analysis that can help in evaluating colonization versus infection. The Curetis panel includes the detection of 29 bacterial pathogens and 19 resistance genes. Both panels can be used with multiple specimen types (sputum, endotracheal aspirates, bronchoalveolar lavage fluid). Although semiquantitative analysis may improve the clinical specificity, neither molecular testing panels nor culture separate airway colonizers from invasive pathogens. The possibility of a downstream effect of paradoxically increasing antimicrobial use should be a notable concern. These panels may be most useful when patients have new or worsening lung infiltrates, are moderately to severely ill, have received empiric antibiotics before obtaining cultures, and/or there is concern about multidrug-resistant bacteria or a polymicrobial infection.⁴⁰

The BioFire FilmArray Pneumonia Panel showed PPA and NPA of 98.1% and 96.2%, respectively, for identifying bacterial targets on bronchoalveolar lavage specimens compared with culture.⁴¹ Similarly, a high overall agreement of 99.2% (95% CI, 98.4%-99.6%) for viral detection was seen between the FilmArray panel and culture.⁴² The excellent sensitivity of the pneumonia panel may be useful to rule out bacterial coinfections and avoid inappropriate prescribing of antibiotics, but positive results should be interpreted with caution. In a multinational study of 52 laboratories, the panel identified significantly more positive specimens (76.1%) than standard-of-care (SOC) testing (56.03%) (P=0.0001) and more potential pathogens than SOC (P=0.0001) independent of specimen type, with the most significant discrepancies for fastidious pathogens.⁴³ Lower SOC bacteria detection may be explained by local reporting guidelines and testing of specimens from patients on antibiotics.⁴³

In summary, the panel carries interpretation challenges, including understanding the increased detection rates, colonization, and infection differentiation.⁴³ A recent study prospectively examined sputum specimens in 70 patients with pneumonia, and the potential for antibiotic optimization was observed in 56 of 70 patients.⁴⁴ Nine bacteria in 9 patients were not covered by empiric therapy, and 70 antibiotics in 49 patients could have been discontinued.⁴⁴ Likewise, a retrospective multicenter study observed antimicrobial de-escalation in 63 of 159 (40%) and escalation in 35 (22%) hospitalized pneumonia patients based on FilmArray panel results.⁴⁵ This reinforces the panel’s potential to reduce unnecessary antimicrobial exposure and increase the appropriateness of empiric antibiotic therapy.

The Curetis Unyvero LRT Panel also has reported robust diagnostic accuracy. In an evaluation by Collins et al, the PPA and NPA for the bacterial targets were 96.5% and 99.6%, respectively.⁴⁶ Also, Klein et al found high overall PPA and NPA with culture, but 21.7% of specimens had additional potential pathogens identified by the panel.⁴⁷ Moreover, the Curetis panel had the potential to facilitate earlier initiation of effective antibiotic therapy in 20 of 95 patients (21%) and de-escalation in 37 patients (39%) with ventilator-associated pneumonia in a prospective study.⁴⁸ Pickens et al predicted antibiotic de-escalation from unnecessary methicillin-resistant Staphylococcus aureus (MRSA) and Pseudomonas aeruginosa coverage in 65.9% (405/615) of patients.⁴⁹ Challenges with the interpretation of newer RDTs are of concern but may be mitigated by antimicrobial stewardship program (ASP) interventions, requiring further exploration.

Surveillance Screening for MRSA

Nasal screening methods for MRSA have evolved beyond use for infection prevention and control practice to have clinical utility for routine use in de-escalations of MRSA therapy, predominantly in patients with suspected or confirmed pneumonia. Robust evidence has reflected more than 95% negative predictive value (NPV) for using the test to rule out MRSA pneumonia.⁵⁰ Therefore, the ATS/IDSA CAP guidelines endorsed the routine use of MRSA nasal PCR screening to de-escalate MRSA coverage.³⁷ ASP implementation of this approach has been associated with a median decrease of 2.1 days of vancomycin (P<0.01).⁵¹ Other implementation results of nasal screening in suspected or confirmed pneumonia among ICU patients have been associated with $108 per patient in cost avoidance based on the cost of surveillance testing, vancomycin, and vancomycin therapeutic drug monitoring levels.⁵² Reviews of implementation considerations suggest fidelity of the nasal testing for 7 days after results and lack of impact of vancomycin exposure in affecting testing results.^53,54 Systematic reviews and meta-analyses have also supported the use of the screen for NPV beyond pneumonia, such as in skin and soft tissue infections.⁵⁵ A national study from the Veterans Affairs health system has supported this concept in the largest cohort, including clinical cultures (N=561,325).⁵⁶ These data showed a high overall NPV for all infection types (96.5%) and among specific infections, including bloodstream infections (BSIs) (96.5%), intraabdominal infections (98.6%), RTIs (96.1%), wound cultures (93.1%), and urinary tract infections (UTIs) (99.2%). Table 3 is a list of selected RDTs used for resistance testing, including MRSA screening.

Clinical Importance of RDTs For Bloodstream Infections

BSIs refer to the presence of bacterial (bacteremia) or fungal (fungemia) microorganisms in the bloodstream. The blood is considered a sterile environment, and BSIs can result from infections in other parts of the body (ie, pneumonia, UTI, skin and soft tissue infections, etc), catheter use, and contaminated surgical procedures, or simply through breaks in the skin.^57,58 BSIs are serious conditions associated with high morbidity and mortality globally.^59,60 Blood culture remains the gold standard for BSI diagnosis, but it has limitations such as slow turnaround time and reduced sensitivity. Some studies estimate the sensitivity of blood culture to be as low as 50%.^61,62

However, new and advanced methods like RDTs can improve the sensitivity of pathogen detection and shorten result reporting time significantly for BSI.⁶³

Molecular RDTs have transformed the approach to managing BSIs and address potential contaminants in blood cultures, such as coagulase-negative staphylococci. RDTs can provide crucial, actionable insights at significantly earlier stages of the treatment process compared with traditional microbiological culture methods. The implementation of molecular RDTs has been associated with a reduction in time to effective therapy, hospital length of stay (LOS), and mortality when accompanied by ASP interventions.⁶⁴ Similar to the clinical impacts observed, a cost-effectiveness analysis also has reflected the benefits of molecular RDTs in BSIs.⁶⁵ These data also highlight the strong synergy of stewardship and RDTs, which have an 80% chance of cost-effectiveness with an ASP, but only 41.1% without. Alternatively, automated rapid phenotypic testing systems such as the Accelerate PhenoTest system (Accelerate Diagnostics) can yield organism identification, minimum inhibitory concentration (MIC), and susceptibility interpretation with a turnaround time of approximately 7 hours after positive blood culture.⁶⁶

RDTs developed for BSIs use different diagnostic technologies, including PCR-based technologies (eg, BioFire FilmArray BCID, GenMark ePlex BCID), nanoparticle probe technology (eg, VERIGENE BC-GP and BC-GN), matrix-assisted laser desorption/ionization time of flight (MALDI-TOF) (eg, BioMérieux, BD Bruker), magnetic resonance technology (eg, the T2Candida and T2Bacteria panels, T2 Biosystems), or morphokinetic cellular analysis technique (eg, Accelerate PhenoTest BC Kit). Although all of these RDTs can process specimens rapidly and provide results in just a few hours, the T2 panels are FDA-approved diagnostic panels that can process samples directly from whole blood without waiting for positive blood culture, further expediting result reporting. This is particularly important in patients for whom rapid administration of directed therapy is crucial due to severe infection (eg, septic shock).⁶⁷ However, T2 panels can only identify a small number of pathogens—6 bacteria and 5 fungi—compared with larger panels like the ePlex panel, which detects 20 gram-positive and 21 gram-negative bacterial pathogens. Among RDTs that can only process positive blood culture samples, the largest difference in pathogen detection time is 1.5 hours. Although the BioFire BCID2 panel, which can detect 15 gram-negative bacteria, 11 gram-positive bacteria, and 7 yeast with the fastest turnaround time (1 hour), showed improved outcomes in identifying gram-negative bacteria causing BSIs compared with the VERIGENE panel (which detects 9 gram-negative bacteria, 12 gram-positive bacteria) and the BioFire original BCID panel (which detects 11 gram-negative bacteria, 8 gram-positive bacteria, 5 yeast).⁶⁸

Some RDTs may include pathogens in their detection list but are not FDA-cleared for that specific pathogen identification. For example, the VERIGENE Bloodstream Infections Testing Panel is not FDA-cleared for Serratia marcescens or Micrococcus species. Furthermore, some tests may not be able to distinguish between certain bacteria. For example, the VERIGENE Bloodstream Infections Testing Panel cannot distinguish Escherichia coli from Shigella species. Interestingly, only the BioFire BCID2 panel and ePlex BCID Panels can identify Stenotrophomonas maltophilia.

Table 2 provides a direct comparison among common RDTs for BSIs. New RDTs are in development to further reduce time to results to less than 1 hour. For example, the FAST-ID BSI Panel (Qvella) can identify more than 90% of pathogens causing BSI directly from a whole blood sample. The use of non–culture-based methods in rapid diagnostic testing will likely lead to decreased time to appropriate therapy and may improve health outcomes, specifically in critically ill patients. Moreover, in addition to using RDTs for detection and early initiation of the appropriate antimicrobial agents, they can be used for early discontinuation of empirical therapy. For example, the use of the T2Candida panel resulted in a reduction in empiric antifungal days of therapy in ICU patients compared with the beta-D-glucan biomarker.⁶⁹

However, RDTs that can decrease the turnaround time for biomarker detection (eg, beta-D-glucan, galactomannan) could also help to rule out infection and facilitate the early discontinuation of empirical antimicrobial therapy. For example, Fungitell STAT is an FDA-cleared assay (Associates of Cape Cod), that can provide rapid screening for invasive candidiasis through the detection of serum beta-D-glucan in about 1 hour. Beta-D-glucan testing has demonstrated utility in some circumstances; for instance, this test can be helpful for the detection of intraabdominal candidiasis in ICU patients where blood culture sensitivity is low.^70,71

A retrospective study by Walker et al evaluated the clinical impact of VERIGENE BC-GN panel implementation for the detection of gram-negative bacteria in positive blood cultures in hospitalized patients by comparing the 6-month period before and the 6-month period after implementing the panel, and demonstrated that the BC-GN panel identified 87% of gram-negative cultures and was accurate in 98% of these cultures compared with the conventional culture method. The BC-GN panel was associated with faster pathogen identification than the traditional culture method (10.9 vs 37.9 hours; P<0.001). Furthermore, LOS and 30-day mortality were significantly lower after implementing the BC-GN panel (P<0.05).⁷² Another retrospective study compared the turnaround time and accuracy of organism and resistance gene detection for VERIGENE Bloodstream Infections Testing Panels and the BioFire Blood Culture Identification (original panel) with the traditional culture method for 173 patients with positive blood cultures. This study found that both panels accurately identified pathogens much faster than traditional cultures by 27.95 and 29.17 hours for the VERIGENE and BioFire panels, respectively.⁷³ One study compared 3 RDT panels (VERIGENE BC-GN, BioFire BCID, and BioFire BCID2) using the Desirability of Outcome Ranking for the Management of Antimicrobial Therapy (DOOR-MAT) to assess potential downstream antimicrobial prescribing decisions resulting from the panel’s different organism and resistance detection. They found that the mean DOOR-MAT score was higher, where a higher score reflects better outcome, for BioFire BCID2 than VERIGENE BC-GN (P<0.7) and BioFire BCID (P<0.0001). The mean DOOR-MAT scores were 83.8 (SD, 25.7), 59.9 (SD, 33.7), and 89.7 (SD, 24.7) for VERIGENE BC-GN, BioFire BCID, and BioFire BCID2, respectively.⁶⁸

A study by Huang et al compared the ePlex BCID panels to detect BSI pathogens with traditional culture for 216 patients with positive blood culture and found the PPA was 96% and 94% for gram-positive and gram-negative bacteria, respectively.⁷⁴ Another recent study evaluated the turnaround time of the ePlex BCID-GN panel compared with MALDI-TOF for gram-negative bacteria identification. The study showed that the ePlex BCID-GN panel identified gram-negative bacteria 26.7 hours earlier on average than MALDI-TOF.⁷⁵ A recent prospective, multicenter study compared the T2Candida panel with beta-D-glucan testing for the diagnosis of intraabdominal candidiasis in critically ill patients using blood samples, and demonstrated that T2Candida performance was comparable with beta-D-glucan. Of the 134 patients evaluated, 13 had proven intraabdominal candidiasis and 2 had concurrent candidemia. The sensitivity, specificity, positive predictive value (PPV), and NPV, respectively, were 46%, 97%, 61%, and 94% for T2Candida and 85%, 83%, 36%, and 98% for beta-D-glucan. All positive T2Candida results were consistent with the culture results at the species level, except for 1 case of dual infection.⁷⁶ Furthermore, a study compared the turnaround time, sensitivity, and specificity of the T2Bacteria panel with the standard blood culture method in patients with BSI. This study found that the mean time to species identification using T2Bacteria was 3.61 to 7.70 hours, depending on the number of samples tested, compared with the mean time from the start of blood culture incubation to positivity and species identification of 38.5 hours (SD, 32.8 hours) and 71.7 hours (SD, 39.3 hours), respectively, for the traditional method.⁷⁷

A unique feature of the Accelerate PhenoTest system is that it can provide both organism identification within 2 hours and antimicrobial susceptibility testing within 7 hours from positive blood culture. Several studies have explored the Accelerate PhenoTest system coupled with ASP intervention. Results of 448 patients with gram-negative BSIs in a randomized controlled trial with prospective audit and feedback in both arms reflected significantly faster antibiotic changes (median decrease of approximately 25 hours for gram-negative antibiotics; P<0.001) with the Accelerate PhenoTest system compared with culture. Antibiotic escalation also was significantly faster with the Accelerate PhenoTest system relative to culture-based methods for antimicrobial-resistant BSIs (median decrease of approximately 43 hours; P=0.01). There were no differences between arms in patient outcomes, including LOS and mortality.⁷⁸

Role of RDTs in Detecting Antimicrobial Resistance

Bacteria can develop resistance against antimicrobial agents through various mechanisms such as enzymatic inactivation (eg, TEM, SHV, CTX-M, NDM, IMP, VIM, AmpC, OXA) altering membrane permeability (eg, porin loss), efflux pump overexpression, target site modification, and bypass of inhibition (ie, bacteria avoid using the targeted enzyme).^79,80 Antimicrobial resistance is associated with increased healthcare costs, morbidity, and mortality.⁸¹ Traditionally, AST is used to measure antimicrobial resistance; however, the broth microdilution (BMD) method is the gold standard. The BMD method determines the minimum inhibitory concentration (MIC) of the bacterial strain, which helps guide treatment decisions. Furthermore, the BMD method has an advantage over the Kirby-Bauer disk diffusion method in that the quantitative MIC values allow assessment of the degree of resistance or susceptibility.⁸²⁼⁸⁴ Although the BMD method is the gold standard, the Kirby-Bauer disk diffusion method has also been commonly used for AST in clinical microbiological laboratories since its development in 1940.^85,86 This method involves culturing bacteria on an agar plate and exposing them to paper disks impregnated with different antibiotics. After incubation, the zones of inhibition are measured to determine whether bacteria are susceptible or resistant to antibiotics. However, these traditional methods can take a long time for the results to be available (about 2-3 days from positive culture).^87,88 Fortunately, the availability of RDTs can significantly shorten the AST results from days to hours. Currently, there are several methods used by RDTs to detect antimicrobial resistance and susceptibility. Below, we discuss the advantages and disadvantages of some of these methods.

Molecular Testing for Genotypic Detection

The RDTs that use this method search for resistance-associated mutations or genes, usually by using various technologies like PCR-based technology, nanoparticle probe-based technology, and others. Although molecular testing can rapidly detect antimicrobial resistance markers, it cannot confirm antimicrobial susceptibility. Another disadvantage of this method is that it can only find the resistance markers for which the test was designed to search.⁸⁹ RDTs that use molecular testing can be further classified based on the technology used: nanoparticle probe-based technology (eg, VERIGENE Bloodstream Infections Testing panels), PCR-based technology (eg, BioFire FilmArray Pneumonia [PN] Panel, BioFire BCID2 Panel, BioFire Joint Infection Panel, ePlex Blood Culture Identification panels, and Xpert CARBA-R), and PCR technology combined with microarray analysis (eg, Unyvero Lower Respiratory Tract Panel).

Phenotypic Testing

This method assesses the ability of microorganisms to grow in the presence of antimicrobial agents. The advantage of this method over molecular testing is that it can predict both susceptibility and resistance. For example, the Next Generation Phenotyping (NGP) System developed by Selux Diagnostics and the Accelerate PhenoTest BC Kit can provide AST results in about 5 to 7 hours, which is markedly faster than BMD. However, NGP panels can only provide AST, whereas the Accelerate PhenoTest BC Kit can provide both AST and pathogen identification. Phenotypic methods can be beneficial for detecting antimicrobial resistance in organisms that usually develop resistance through multiple mechanisms (eg, gram-negative bacteria).

Mass Spectrometry

This method can identify organisms and indirectly detect resistance patterns based on the organism’s unique protein profile. MALDI-TOF mass spectrometry is the typical example of this method. A disadvantage of this method is the relatively slow turnaround time.^90,91

DNA-Sequencing Technologies

Next-generation sequencing (NGS) can provide detailed information about pathogen resistance by analyzing the microorganism’s whole genome. Limitations of NGS include high cost and slower turnaround time, as well as its inability to confirm antimicrobial susceptibility.^92,93 For example, the Karius test uses NGS to identify more than 100 pathogens based on the fragments of genomic DNA these pathogens leave in blood. The Karius test aims to provide next-day results. Although it has not been cleared by the FDA, it is certified under the Clinical Laboratory Improvement Amendments (CLIA). Similarly, Clear Labs has developed Clear Dx Surveillance, a fully automated whole-genome sequencing (WGS) platform that offers complete characterization of bacterial isolates in about 27 hours. At this time, Clear Dx Surveillance is for research use only.⁹⁴

Immunochromatographic Assays

This lateral flow test (LFT) uses antibodies to detect resistance-associated specific antigens. Advantages of this test include simplicity (extensive training and specialized equipment not required) and fast turnaround time.⁹⁵ One example of a RDT that uses this method is the NG-Test CARBA 5.⁹⁶

Bacteriophage Technology

An example of this method is the cobas vivoDx MRSA test (Roche Diagnostics), which uses bacteriophage technology to detect MRSA in nasal swab specimens. This test uses S. aureus-specific phages to insert a plasmid encoding a bioluminescent enzyme. When the proper substrate is added, S. aureus produces light detected by the cobas system. To differentiate between methicillin-susceptible S. aureus and MRSA, the nasal swab sample is incubated in cefoxitin. What makes this method unique is the novel technology that may have further applications for identifying pathogens in the future. However, it has a slow turnaround time (about 5 hours) compared with other MRSA screening methods. Similar to genotypic and NGS methods, this test cannot provide information about susceptibility.⁹⁷

Evidence of RDTs for Resistance

Comparative studies evaluating the different RDTs for resistance detection are scarce. One study with a sample size of 133 positive blood cultures of Enterobacterales compared Xpert CARBA-R to NG-Test CARBA 5 for rapid detection of carbapenemase genes or carbapenemases, respectively, in positive blood cultures and found that the sensitivity and specificity of both tests to be 100% when compared with PCR sequencing.⁹⁸ However, the Xpert Carba-R does not require a second incubation period, which significantly reduces setup time from 16 hours to about 5 minutes. Similarly, Page et al. evaluated the impact of the Xpert MRSA/SA blood culture test on the management of 22 positive blood cultures in obstetric patients. The study compared patients’ outcomes before Xpert MRSA/SA was implemented and after implementing Xpert MRSA/SA where results demonstrated a reduction in the median antimicrobial duration (55.5 to 43.5 hours, P=0.46) and LOS (66.5 to 56.0 hours, P=0.15) after RDT implementation.⁹⁹ These findings suggest that RDTs for resistance detection can help decrease the frequent use of nephrotoxic antimicrobial agents like vancomycin, which is commonly used in practice as part of an empirical therapy for patients with septic shock.¹⁰⁰ A study by Dortet et al. compared the carbapenemase-producing Enterobacteriaceae rapid detection performance of 3 MALDI-TOF-based techniques (MBT STAR-Carba IVD kit, Microflex LT Biotyper, and VITEK MS Plus) with RAPIDEC CARBA NP. A total of 175 isolates were tested, and MALDI-TOF-based techniques demonstrated sensitivities ranging from 95% to 100% and specificities from 98.2% to 100% compared with 99.2% and 100%, respectively, for the RAPIDEC CARBA NP.¹⁰¹ A multicenter retrospective observational study compared clinical and antimicrobial stewardship metrics before and after implementation of Accelerate PhenoTest BC Kit (AXDX) in patients with BSIs where 854 patients (435 pre-AXDX, 419 post-AXDX) were included. This study found that the median time to optimal therapy was 17.2 hours shorter in the post-AXDX arm (23.7 hours) compared with the pre-AXDX arm (40.9 hours; P<0.0001), and LOS was shorter in the post-AXDX arm (5.4 vs 6.4 days; P=0.03) among patients with gram-negative bacteremia.¹⁰² Yoo et al accurately detected resistance genes using the BioFire pneumonia panel, showing concordant results for the resistant organisms identified by culture. However, the genetic marker of antimicrobial resistance, particularly the CTX-M and carbapenemase genes, could not be definitively linked to the microorganisms detected.¹⁰³ Thus, culture testing is essential to confirm susceptibility or resistance, as current culture methods with AST must be maintained while pursuing consistency in technological advances.⁴⁶

Furthermore, the surveillance of gram-negative bacterial resistance using rectal swab testing may became standard in some settings. However, the clinical utility of these tests in directing therapy has yet to show significant promise. Streck ARM-D resistance detection kits are examples of RDTs that can play a beneficial role for gram-negative resistance surveillance. These kits can detect multiple beta-lactamase-associated resistance such as AmpC, OXA, CTX-M, NIM, IMP, TEM, SHV, and others. However, these kits are not FDA-cleared yet, but have received clearance in Europe to be used for diagnostic purposes.^104,105 Table 3 is a summary and comparison of selected common RDTs designed for resistance detection.

Role of RDTs for Other Infections

Although the primary focus of this article is RDTs for BSIs, RTIs, and resistance detection, numerous RDTs are designed for other samples (cerebrospinal fluid [CSF], synovial fluid, stool, and more). For example, the rapid identification of causative pathogens for meningitis is crucial because meningitis is associated with high mortality, particularly when Streptococcus pneumoniae is implicated, for which 30-day mortality is estimated to be 7.2% compared with 3.8% for all other causative pathogens.¹⁰⁶ As a result, bioMérieux developed the BioFire FilmArray Meningitis/Encephalitis (ME) Panel that received FDA clearance for the identification of 14 common bacterial, viral, and fungal causes of CNS infections from CSF samples. A meta-analysis study included 3,764 patients and found that the sensitivity and specificity of the BioFire FilmArray ME panel was 90% (95% CI, 86%-93%) and 97% (95% CI, 94%-99%), respectively.¹⁰⁷

Another application for RDTs includes screening for bacterial colonization, such as Group B Streptococcus (GBS) during pregnancy. Most women colonized with GBS are asymptomatic and can transmit the bacteria to their newborns during delivery. Approximately 12% to 27% of pregnant women are colonized with GBS in their rectum or vagina. GBS infections are considered the leading cause of sepsis and meningitis in newborns during their first week of life.^108,109 Luminex developed the ARIES GBS assay that can detect GBS from vaginal-rectal specimen swabs obtained from pregnant women. This assay provides results in about 18 to 24 hours due to the need for overnight incubation in Lim-broth enrichment, whereas, traditional culture methods usually take about 48 hours. A study by Hernandez et al included 688 specimens and showed that the ARIES GBS assay identified GBS with 96.1% sensitivity (95% CI, 91.2%-98.7%) and 91.4% specificity (95% CI, 88.8%-93.6%).¹¹⁰ There are other FDA-cleared RDTs for gastrointestinal tract infections (eg, VERIGENE Enteric Pathogens Test, BioFire FilmArray Gastrointestinal Panel), joint infections (eg, BioFire Joint Infection Panel), Clostridioides difficile infection (eg, VERIGENE Clostridium difficile Test, ARIES C. difficile Assay, Xpert C. difficile BT, cobas C.diff Test), and other infections. Table 4 presents a summary and comparison of selected RDTs designed for other infections.

Conclusion

The rapid advancement of RDTs can play a crucial role in timely identification of causative pathogens, testing for susceptibility, and detecting resistance markers. RDTs can shorten the time between specimen collection and results compared with traditional culture methods. Specifically, non-culture-based methods eliminate the need for long incubation periods, which can significantly decrease the result turnaround time. Implementing these tests can result in quickly initiating the appropriate therapy and avoiding the use of inappropriate antimicrobial agents. RDTs can also play a key role in discontinuing unnecessary empiric antimicrobial therapy. RDTs showed promising results in terms of accuracy and turnaround time. The focus now should be on finding cost-effective methods that facilitate implementing these RDTs in clinical microbiology laboratories.

References

1. Messacar K, Parker SK, Todd JK, et al. J Clin Microbiol. 2017;55(3):715-723.
2. Baron EJ, Miller JM, Weinstein MP, et al. Clin Infect Dis. 2013;57(4):e22-e121.
3. Calderaro A, Buttrini M, Farina B, et al. Microorganisms. 2022;10(9):1856.
4. Kuti EL, Patel AA, Coleman CI. J Crit Care. 2008;23(1):91-100.
5. Buchan BW, Ginocchio CC, Manii R, et al. PLoS Med. 2013;10(7):e1001478.
6. Mylonakis E, Clancy CJ, Ostrosky-Zeichner L, et al. Clin Infect Dis. 2015;60(6):892-899.
7. Naccache SN, Federman S, Veeraraghavan N, et al. Genome Res. 2014;24(7):1180-1192.
8. Seng P, Drancourt M, Gouriet F, et al. Clin Infect Dis. 2009;49(4):543-551.
9. Jin X, Ren J, Li R, et al. EClinicalMedicine. 2021;37:100986.
10. Safiri S, Mahmoodpoor A, Kolahi AA, et al. Front Public Health. 2023;10:1028525.
11. GBD 2016. Lancet Infect Dis. 2018;18(11):1191-1210.
12. Almirall J, Bolíbar I, Balanzó X, et al. Eur Respir J. 1999;13(2):349-355.
13. Tazinya AA, Halle-Ekane GE, Mbuagbaw LT, et al. BMC Pulm Med. 2018;18(1):7.
14. Rogan M. International Encyclopedia of Public Health 2017. doi:10.1016/B978-0-12-803678-5.00383-0
15. MacVane SH, Oppermann N, Humphries RM. J Clin Microbiol. 2020;58(11):e01468-20.
16. Grief SN. Prim Care. 2013;40(3):757-770.
17. Saleri N, Ryan ET. Travel Medicine 2017. doi:10.1016/B978-0-323-54696-6.00059-8
18. Mandell GL, Bennett JE, Dolin R, et al. Arch Patho Lab Med. 1997;121(8):908.
19. World Health Organization. Accessed August 20, 2023. https://apps.who.int/ iris/ handle/ 10665/ 70032
20. Apisarnthanarak A, Bin Kim H, Moore LSP, et al. Clin Infect Dis. 2021;74(11):2067-2076.
21. VERIGENE Respiratory Pathogens Flex Panel. Package Insert. Accessed September 5, 2023. http://www.nanosphere.us/ sites/ default/ files/ support-docs/ nanosphere_rpflex_ivd_revc_final.pdf
22. Luminex Corporation. Published online 2020. Accessed September 5, 2023. https://www.luminexcorp.com/ respiratory-pathogens-flex-test/ ?wpdmdl=44093
23. Leber AL, Everhart K, Daly JA, et al. J Clin Microbiol. 2018;56(6):e01945-17.
24. Leber AL, Lisby JG, Hansen G, et al. J Clin Microbiol. 2020;58(5):e00155-20.
25. Visseaux B, Le Hingrat Q, Collin G, et al. J Clin Microbiol. 2020;58(8):e00630-20.
26. van Asten SAV, Boers SA, de Groot JDF, et al. BMC Microbiol. 2021;21:236.
27. Jarrett J, Uhteg K, Forman MS, et al. J Clin Virol. 2021;135:104737.
28. NxTAG Respiratory Pathogen Panel. Package Insert. Accessed September 5, 2023. https://www.biomed-global.com/ wp-content/ uploads/ 2021/ 08/ NxTAG-RPP-CAN-IVD-Catalogue-1.pdf
29. Chen JHK, Yip CCY, Chan JFW, et al. J Clin Microbiol. 2020;58(8):e00936-20.
30. Beckmann C, Hirsch HH. J Med Virol. 2016;88(8):1319-1324.
31. Popowitch EB, Kaplan S, Wu Z, et al. Microbiol Spectr. 10(4):e01248-22.
32. Chan KH, To KKW, Li PTW, et al. Adv Virol. 2017;2017:1324276.
33. Nijhuis RHT, Guerendiain D, Claas ECJ, et al. J Clin Microbiol. 2017;55(6):1938-1945.
34. Babady NE, England MR, Jurcic Smith KL, et al. J Clin Microbiol. 2018;56(2):e01658-17.
35. Kung HC, Hoyert DL, Xu J, et al. Deaths: final data for 2005. Natl Vital Stat Rep. 2008;56(10):1-120.
36. Ramirez JA, Wiemken TL, Peyrani P, et al. Clin Infect Dis. 2017;65(11):1806-1812.
37. Metlay JP, Waterer GW, Long AC, et al. Am J Respir Crit Care Med. 2019;200(7):e45-e67.
38. Huang DT, Yealy DM, Filbin MR, et al. N Engl J Med. 2018;379(3):236-249.
39. Ruuskanen O, Lahti E, Jennings LC, et al. Lancet. 2011;377(9773):1264-1275.
40. Hanson KE, Azar MM, Banerjee R, et al. Clin Infect Dis. 2020;71(10):2744-2751.
41. Buchan BW, Windham S, Balada-Llasat JM, et al. J Clin Microbiol. 2020;58(7):e00135-20.
42. Webber DM, Wallace MA, Burnham CA, et al. J Clin Microbiol. 2020;58(7):e00343-20.
43. Ginocchio CC, Garcia-Mondragon C, Mauerhofer B, et al. Eur J Clin Microbiol Infect Dis. 2021;40(8):1609-1622.
44. Zacharioudakis IM, Zervou FN, Dubrovskaya Y, et al. Int J Infect Dis. 2021;104:354-360.
45. Monard C, Pehlivan J, Auger G, et al. Crit Care. 2020;24:434.
46. Collins ME, Popowitch EB, Miller MB. J Clin Microbiol. 2020;58(5):e02013-19.
47. Klein M, Bacher J, Barth S, et al. J Clin Microbiol. 2021;59(3):e02497-20.
48. Peiffer-Smadja N, Bouadma L, Mathy V, et al. Crit Care. 2020;24(1):366.
49. Pickens C, Wunderink RG, Qi C, et al. Diagn Microbiol Infect Dis. 2020;98(4):115179.
50. Parente DM, Cunha CB, Mylonakis E, et al. Clin Infect Dis. 2018;67(1):1-7.
51. Willis C, Allen B, Tucker C, et al. Am J Health Syst Pharm. 2017;74(21):1765-1773.
52. Smith MN, Erdman MJ, Ferreira JA, et al. J Crit Care. 2017;38:168-171.
53. Smith MN, Brotherton AL, Lusardi K, et al. Ann Pharmacother. 2019;53(6):627-638.
54. Carr AL, Daley MJ, Givens Merkel K, et al. Pharmacotherapy. 2018;38(12):1216-1228.
55. Butler-Laporte G, De L’étoile-Morel S, Cheng MP, et al. J Infect. 2018;77(6):489-495.
56. Mergenhagen KA, Starr KE, Wattengel BA, et al. Clin Infect Dis. 2020;71(5):1142-1148.
57. Tan CCS, Ko KKK, Chen H, et al. Nat Microbiol. 2023;8(5):973-985.
58. Viscoli C. Virulence. 2016;7(3):248-251.
59. Santella B, Folliero V, Pirofalo GM, et al. Antibiotics (Basel). 2020;9(12):851.
60. Timsit JF, Ruppé E, Barbier F, et al. Intensive Care Med. 2020;46(2):266-284.
61. Murray PR, Masur H. Crit Care Med. 2012;40(12):3277-3282.
62. Brun-Buisson C, Doyon F, Carlet J, et al. JAMA. 1995;274(12):968-974.
63. Dunbar SA, Gardner C, Das S. Front Cell Infect Microbiol. 2022;12:859935.
64. Timbrook TT, Morton JB, McConeghy KW, et al. Clin Infect Dis. 2017;64(1):15-23.
65. Pliakos EE, Andreatos N, Shehadeh F, et al. Clin Microbiol Rev. 2018;31(3):e00095-17.
66. Brosh-Nissimov T, Tzur A, Grupel D, et al. Ann Clin Microbiol Antimicrob. 2023;22:62.
67. Kumar A. Curr Infect Dis Rep. 2010;12(5):336-344.
68. Claeys KC, Schlaffer K, Smith R, et al. Clin Infect Dis. 2021;73(6):1103-1106.
69. Gill CM, Kenney RM, Hencken L, et al. Diagn Microbiol Infect Dis. 2019;95(2):162-165.
70. Bassetti M, Righi E, Ansaldi F, et al. Intensive Care Med. 2015;41(9):1601-1610.
71. Lamoth F, Akan H, Andes D, et al. Clin Infect Dis. 2021;72(suppl 2):s102-s108.
72. Walker T, Dumadag S, Lee CJ, et al. J Clin Microbiol. 2016;54(7):1789-1796.
73. Ward C, Stocker K, Begum J, et al. Eur J Clin Microbiol Infect Dis. 2015;34(3):487-496.
74. Huang TD, Melnik E, Bogaerts P, et al. J Clin Microbiol. 2019;57(2):e01597-18.
75. McCarty TP, Cumagun P, Meeder J, et al. Microbiol Spectr. 11(1):e04092-22.
76. Krifors A, Ullberg M, Castegren M, et al. J Fungi (Basel). 2022;8(1):86.
77. Nguyen MH, Clancy CJ, Pasculle AW, et al. Ann Intern Med. 2019;170(12):845-852.
78. Banerjee R, Komarow L, Virk A, et al. Clin Infect Dis. 2021;73(1):e39-e46.
79. Kakoullis L, Papachristodoulou E, Chra P, et al. Antibiotics (Basel). 2021;10(4):415.
80. Sawa T, Kooguchi K, Moriyama K. J Intensive Care. 2020;8:13.
81. Dadgostar P. Infect Drug Resist. 2019;12:3903-3910.
82. Clinical and Laboratory Standards Institute (CLSI). M07. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. 11th ed. Clinical and Laboratory Standards Institute; 2018.
83. Golus J, Sawicki R, Widelski J, et al. J Appl Microbiol. 2016;121(5):1291-1299.
84. Kowalska-Krochmal B, Dudek-Wicher R. Pathogens. 2021;10(2):165.
85. The European Committee on Antimicrobial Susceptibility Testing (EUCAST): Clinical breakpoints and dosing of antibiotics. Accessed September 6, 2023. https://www.eucast.org/ clinical_breakpoints
86. Clinical and Laboratory Standards Institute (CLSI). M02. Performance Standards for Antimicrobial Disk Susceptibility Tests. 13th ed. Clinical and Laboratory Standards Institute; 2018.
87. Smith KP, Kirby JE. Clin Lab Med. 2019;39(3):333-344.
88. Khan ZA, Siddiqui MF. Diagnostics (Basel). 2019;9(2):49.
89. Burnham CAD, Leeds J, Nordmann P, et al. Nat Rev Microbiol. 2017;15(11):697-703.
90. Idelevich EA, Becker K. J Clin Microbiol. 2021;59(12):e0181419.
91. Vasala A, Hytönen VP, Laitinen OH. Front Cell Infect Microbiol. 2020;10:308.
92. Motro Y, Moran-Gilad J. Biomol Detect Quantif. 2017;14:1-6.
93. Qin D. Cancer Biol Med. 2019;16(1):4-10.
94. Hogan CA, Yang S, Garner OB, et al. Clin Infect Dis. 2021;72(2):239-245.
95. Boutal H, Moguet C, Pommiès L, et al. Diagnostics (Basel). 2022;12(7):1744.
96. Kon H, Abramov S, Frenk S, et al. Ann Clin Microbiol Antimicrob. 2021;20:61.
97. Office of the Commissioner. FDA authorizes marketing of diagnostic test that uses novel technology to detect MRSA bacteria. Accessed August 27, 2023. https://bit.ly/ 3sUv776-IDSE
98. Xu Y, Song W, Huang P, et al. Infect Drug Resist. 2022;15:7719-7725.
99. Page A, O’Rourke S, Brennan M, et al. Ir J Med Sci. 2017;186(4):995-998.
100. Guarino M, Perna B, Cesaro AE, et al. J Clin Med. 2023;12(9):3188.
101. Dortet L, Tandé D, de Briel D, et al. J Antimicrob Chemother. 2018;73(9):2352-2359. doi:10.1093/jac/dky209
102. Bhalodi AA, MacVane SH, Ford B, et al. Clin Infect Dis. 2022;75(2):269-277.
103. Yoo IY, Huh K, Shim HJ, et al. Int J Infect Dis. 2020;95:326-331.
104. Lindblom A, Karami N, Magnusson T, et al. Eur J Clin Microbiol Infect Dis. 2018;37(8):1491-1497.
105. Rottier WC, Bamberg YRP, Dorigo-Zetsma JW, et al. Clin Infect Dis. 2015;60(11):1622-1630.
106. Kiyani M, Hodges SE, Adil SM, et al. Neurol Clin Pract. 2021;11(2):117-126.
107. Tansarli GS, Chapin KC. Clin Microbiol Infect. 2020;26(3):281-290.
108. World Health Organization. State of the art of new vaccine research and development. Accessed September 4, 2023. http://apps.who.int/ iris/ bitstream/ 10665/ 69348/ 1/ WHO_IVB_06.01_eng.pdf
109. Centers for Disease Control and Prevention. Accessed September 4, 2023. https://www.cdc.gov/ groupbstrep/ about/ fast-facts.html
110. Hernandez DR, Wolk DM, Walker KL, et al. J Clin Microbiol. 2018;56(8):e01768-17.

Copyright © 2023 McMahon Publishing, 545 West 45th Street, New York, NY 10036. Printed in the USA. All rights reserved, including the right of reproduction, in whole or in part, in any form.