Area of technical uncertainty for susceptibility testing of amoxicillin/clavulanate against Escherichia coli: analysis of automated system, Etest and disk diffusion methods compared to the broth microdilution reference

Bacteriuria and phenotypic antimicrobial susceptibility testing in 45 min by point-of-care Sysmex PA-100 System: first clinical evaluation

PURPOSE

This study compared the results of the new Sysmex PA-100 AST System, a point-of-care analyser, with routine microbiology for the detection of urinary tract infections (UTI) and performance of antimicrobial susceptibility tests (AST) directly from urine.

METHODS

Native urine samples from 278 female patients with suspected uncomplicated UTI were tested in the Sysmex PA-100 and with reference methods of routine microbiology: urine culture for bacteriuria and disc diffusion for AST.

RESULTS

The analyser delivered bacteriuria results in 15 min and AST results within 45 min. Sensitivity and specificity for detection of microbiologically confirmed bacteriuria were 84.0% (89/106; 95% CI: 75.6-90.4%) and 99.4% (155/156; 95% CI: 96.5-100%), respectively, for bacterial species within the analyser specifications. These are Escherichia coli, Klebsiella pneumoniae, Proteus mirabilis, Enterococcus faecalis and Staphylococcus saprophyticus, which are common species causing uncomplicated UTI. Overall categorical agreement (OCA) for AST results for the five antimicrobials tested in the Sysmex PA-100 (amoxicillin/clavulanic acid, ciprofloxacin, fosfomycin, nitrofurantoin and trimethoprim) ranged from 85.4% (70/82; 95%CI: 75.9-92.2%) for ciprofloxacin to 96.4% (81/84; 95% CI: 89.9-99.3%) for trimethoprim. The Sysmex PA-100 provided an optimal treatment recommendation in 218/278 cases (78.4%), against 162/278 (58.3%) of clinical decisions.

CONCLUSION

This first clinical evaluation of the Sysmex PA-100 in a near-patient setting demonstrated that the analyser delivers phenotypic AST results within 45 min, which could enable rapid initiation of the correct targeted treatment with no further adjustment needed. The Sysmex PA-100 has the potential to significantly reduce ineffective or unnecessary antibiotic prescription in patients with UTI symptoms.

Ten simple rules for the sharing of bacterial genotype-Phenotype data on antimicrobial resistance

The increasing availability of high-throughput sequencing (frequently termed next-generation sequencing (NGS)) data has created opportunities to gain deeper insights into the mechanisms of a number of diseases and is already impacting many areas of medicine and public health. The area of infectious diseases stands somewhat apart from other human diseases insofar as the relevant genomic data comes from the microbes rather than their human hosts. A particular concern about the threat of antimicrobial resistance (AMR) has driven the collection and reporting of large-scale datasets containing information from microbial genomes together with antimicrobial susceptibility test (AST) results. Unfortunately, the lack of clear standards or guiding principles for the reporting of such data is hampering the field's advancement. We therefore present our recommendations for the publication and sharing of genotype and phenotype data on AMR, in the form of 10 simple rules. The adoption of these recommendations will enhance AMR data interoperability and help enable its large-scale analyses using computational biology tools, including mathematical modelling and machine learning. We hope that these rules can shed light on often overlooked but nonetheless very necessary aspects of AMR data sharing and enhance the field's ability to address the problems of understanding AMR mechanisms, tracking their emergence and spread in populations, and predicting microbial susceptibility to antimicrobials for diagnostic purposes.

Machine Learning for Antimicrobial Resistance Prediction: Current Practice, Limitations, and Clinical Perspective

Antimicrobial resistance (AMR) is a global health crisis that poses a great threat to modern medicine. Effective prevention strategies are urgently required to slow the emergence and further dissemination of AMR. Given the availability of data sets encompassing hundreds or thousands of pathogen genomes, machine learning (ML) is increasingly being used to predict resistance to different antibiotics in pathogens based on gene content and genome composition. A key objective of this work is to advocate for the incorporation of ML into front-line settings but also highlight the further refinements that are necessary to safely and confidently incorporate these methods. The question of what to predict is not trivial given the existence of different quantitative and qualitative laboratory measures of AMR. ML models typically treat genes as independent predictors, with no consideration of structural and functional linkages; they also may not be accurate when new mutational variants of known AMR genes emerge. Finally, to have the technology trusted by end users in public health settings, ML models need to be transparent and explainable to ensure that the basis for prediction is clear. We strongly advocate that the next set of AMR-ML studies should focus on the refinement of these limitations to be able to bridge the gap to diagnostic implementation.

Assessing the Risk of Resistance to Cationic Biocides incorporating Realism-based and Biophysical Approaches

The control of microorganisms is a key objective in disease prevention and in medical, industrial, domestic, and food-production environments. Whilst the effectiveness of biocides in these contexts is well-evidenced, debate continues about the resistance risks associated with their use. This has driven an increased regulatory burden, which in turn could result in a reduction of both the deployment of current biocides and the development of new compounds and formulas. Efforts to balance risk and benefit are therefore of critical importance and should be underpinned by realistic methods and a multi-disciplinary approach, and through objective and critical analyses of the literature. The current literature on this topic can be difficult to navigate. Much of the evidence for potential issues of resistance generation by biocides is based on either correlation analysis of isolated bacteria, where reports of treatment failure are generally uncommon, or laboratory studies that do not necessarily represent real biocide applications. This is complicated by inconsistencies in the definition of the term resistance. Similar uncertainties also apply to cross-resistance between biocides and antibiotics. Risk assessment studies that can better inform practice are required. The resulting knowledge can be utilised by multiple stakeholders including those tasked with new product development, regulatory authorities, clinical practitioners, and the public. This review considers current evidence for resistance and cross-resistance and outlines efforts to increase realism in risk assessment. This is done in the background of the discussion of the mode of application of biocides and the demonstrable benefits as well as the potential risks.

Impact of the introduction of EUCAST's concept of "area of technical uncertainty"

On the first of January 2019, the European Committee on Antimicrobial Susceptibility Testing, EUCAST, introduced the concept of "area of technical uncertainty" (ATU). The aim was to report on the incidence of ATU test results in a selection of common bacterial species and the subsequent impact on antimicrobial resistance categorization and workload. A retrospective analysis of clinical samples collected from February 2019 until November 2019 was performed. Susceptibility to amoxicillin-clavulanic acid and piperacillin-tazobactam in Enterobacterales (Escherichia spp., Klebsiella spp., Proteus spp.), piperacillin-tazobactam in Pseudomonas aeruginosa, and amoxicillin-clavulanic acid and cefuroxime in Haemophilus influenzae was studied. Disk diffusion antibiotic susceptibility testing was read and interpreted by ADAGIO 93400 automated system (Bio-Rad, France). In case of an inhibition zone in the ATU, strains were retested using gradient minimal inhibitory concentration method (Etest, BioMérieux, France). Overall, 14,164 isolate-antibiotic combinations were tested in 7922 isolates, resulting in 1204 (8.5%) disk zone diameters in the ATU region. Retesting of ATUs with Etest resulted in a category change from S to R for amoxicillin-clavulanic acid in 63/498 (12.7%) of Escherichia spp., 2/58 (3.4%) of Klebsiella spp., 2/37 (5.4%) of Proteus spp., and 6/125 (4.8%) of Haemophilus influenzae. For piperacillin-tazobactam, a category change from S to R was found in 33/92 (35.9%) of Pseudomonas aeruginosa. We conclude that ATU testing has a substantial impact on the correct interpretation of antimicrobial resistance, at the expense of turn-around time and with the cost of additional workload.

Broth Microdilution and Gradient Diffusion Strips vs. Reference Agar Dilution Method: First Evaluation for Clostridiales Species Antimicrobial Susceptibility Testing

Antimicrobial susceptibility testing of anaerobes is challenging. Because MIC determination is recommended by both CLSI and EUCAST, commercial broth microdilution and diffusion strip tests have been developed. The reliability of broth microdilution methods has not been assessed yet using the agar dilution reference method. In this work, we evaluated two broth microdilution kits (MICRONAUT-S Anaerobes^® MIC and Sensititre Anaerobe MIC^®) and one gradient diffusion strip method (Liofilchem^®) for antimicrobial susceptibility testing of 47 Clostridiales isolates (Clostridium, Clostridioides and Hungatella species) using the agar dilution method as a reference. The evaluation focused on comparing six antimicrobial molecules available in both microdilution kits. Analytical performances were evaluated according to the Food and Drug Administration (FDA) recommendations. Essential agreements (EA) and categorical agreements (CA) varied greatly according to the molecule and the evaluated method. Vancomycin had values of essential and categorical agreements above 90% for the three methods. The CA fulfilled the FDA criteria for three major molecules in the treatment of Gram-positive anaerobic infections (metronidazole, piperacillin/tazobactam and vancomycin). The highest rate of error was observed for clindamycin. Multicenter studies are needed to further validate these results.