A recent Lancet Microbe study investigated the accuracy of antibiotic resistance determination from Enterococcus faecium genomes for diagnostic purposes.
Study: Antibiotic resistance determination using Enterococcus faecium whole-genome sequences: a diagnostic accuracy study using genotypic and phenotypic data. Image Credit: Summer 1810/Shutterstock.com
Background
The World Health Organization (WHO) has announced that antimicrobial resistance (AMR) is one of human's leading global public threats. The absence of effective antibiotics would fail to prevent infection and would increase the risk of major surgery or cancer chemotherapy.
Among all antibiotic-resistant strains, Enterococcus faecium, particularly vancomycin-resistant E faecium, presents a significant concern as it accounts for more than 200,000 deaths per year globally.
This microbial strain has been identified to be the leading cause of hospital-acquired infection in the surgical site, bloodstream, or urinary tract. E faecium infection particularly increases mortality risks in immunocompromised and critically ill patients. Several studies have shown that E faecium resists many common antibiotics like ampicillin and vancomycin.
Disc diffusion or broth microdilution methods are commonly used for antibiotic susceptibility testing (AST).
This test is crucial to determine the correct antibiotics to treat a bacterial infection. Since antibiotic resistance is genetically encoded, the whole genome sequencing method has been used as an alternative strategy to detect AMR.
The accuracy of this test depends on the availability of updated databases of genetic determinants of AMR. Bioinformatic tools are used to interpret genetic data.
A whole-genome sequencing strategy has been previously used for AMR of Mycobacterium tuberculosis, Streptococcus pneumoniae, and Staphylococcus aureus.
About the study
The main objective of the current study is to evaluate the accuracy of AMR determination from E faecium genomes.
A total of 4,382 E faecium isolates were assembled with whole genome sequences and were correlated with available AMR phenotypes. This strategy identified the concordance between genotypic and phenotypic AMR in E faecium.
E faecium used in this study was obtained between 2000 and 2018 from eleven countries, including Australia, New Zealand, Germany, the Netherlands, the UK, Austria, Denmark, the USA, and Pakistan.
Bacterial genomes were analyzed to detect the presence of genes and mutations that predict AMR in E faecium. The accuracy of genotypic predictions was measured using phenotypic resistance as the reference standard.
The ARIBA (Antimicrobial Resistance Identification By Assembly) software was used to determine antibiotic resistance genes and mutations from whole-genome sequences.
Study findings
To develop a database of genetic determinants of antibiotic resistance in E faecium, a total of 316 genetic determinants were curated that comprised 103 single mutations, 100 mutations in combination, 82 single acquired genes, and 27 multiple acquired genes against 17 antibiotics.
A single genetic mutation or determinant can alter the susceptibility of a pathogen to multiple antibiotics.
The database developed in his study contained genetic determinants that cause resistance to 12 different antibiotics in E faecium.
Notably, this database could accurately predict antibiotic resistance against ampicillin, quinupristin–dalfopristin, ciprofloxacin, vancomycin, linezolid, and teicoplanin.
However, genotypic predictions for tetracyclines and aminoglycosides need further improvement. Furthermore, there is a need to improve sensitivity for tigecycline and daptomycin.
Previous studies have indicated that mutations in penicillin-binding protein 5 (Pbp5) significantly contribute to ampicillin resistance in E faecium.
Resistance to glycopeptides could be determined with high sensitivity and specificity based on the presence of different functional variants of the van A and van B operons. Operon variants that lose specific genes are found in phenotypically resistant isolates.
More research must be conducted in the future to explore the contribution of rare van operons, such as van C, van D, van E, van G, van L, van M, and van N, for glycopeptide resistance in E faecium.
Compared to previously developed AMR databases, the currently developed one exhibited more accurate genotypic predictions for many antibiotics.
The current study observed that identifying non-functional van operon variants, non-conferring resistance mutations, truncated variants of genes, and incorrectly encoded gene–resistance relationships confers low specificity for determining antibiotic resistance.
Therefore, for the prediction of antibiotic resistance, both the presence and integrity of the genetic determinant are crucial.
In line with available database predictions, the currently curated database exhibited low specificity in predicting resistance to tetracycline, while a high-level resistance was observed for aminoglycosides.
Phenotypic re-testing indicated the presence of silenced AMR-associated genes. This phenomenon in E faecium must be explored in the future.
Also, future studies must fully characterize the genetic basis of resistance to these antibiotics.
Conclusions
The current study revealed that the newly curated database exhibited equivalent or higher accuracy in predicting AMR based on the E faecium genome than the existing databases.
These genotypic predictions have been implemented on Pathogenwatch, which is a web-based tool that helps determine AMR from the genomes of many pathogens.
The increase in popularity of the whole-genome sequencing method in clinical and public health microbiology laboratories will aid in adapting this strategy for diagnosing and surveilling AMR in E faecium.
New long COVID index highlights five symptom subtypes