Changing Diagnostic Methods and Increased Detection of Verotoxigenic Escherichia coli, Ireland

From Species to Genes: A New Diagnostic Paradigm

Molecular diagnostics has the potential to revolutionise the field of clinical microbiology. Microbial identification and nomenclature have, for too long, been restricted to phenotypic characterisation. However, this species-level view fails to wholly account for genetic heterogeneity, a result of lateral gene transfer, mediated primarily by mobile genetic elements. This genetic promiscuity has helped to drive virulence development, stress adaptation, and antimicrobial resistance in several important bacterial pathogens, complicating their detection and frustrating our ability to control them. We argue that, as clinical microbiologists at the front line, we must embrace the molecular technologies that allow us to focus specifically on the genetic elements that cause disease rather than the bacterial species that express them. This review focuses on the evolution of microbial taxonomy since the introduction of molecular sequencing, the role of mobile genetic elements in antimicrobial resistance, the current and emerging assays in clinical laboratories, and the comparison of phenotypic versus genotypic analyses. In essence, it is time now to refocus from species to genes as part of a new diagnostic paradigm.

Shiga toxin-producing Escherichia coli diagnosed by Stx PCR: assessing the public health risk of non-O157 strains

BACKGROUND

The implementation by diagnostic laboratories in England of polymerase chain reaction (PCR) to screen faecal specimens for Shiga toxin-producing Escherichia coli (STEC) has resulted in a significant increase in notifications mainly due to non-O157 strains. The purpose of this study was to develop an approach to public health risk assessment that prioritizes follow-up to cases caused by haemolytic uraemic syndrome (HUS) associated E. coli (HUSEC) strains and minimizes unnecessary actions.

METHODS

Epidemiological and microbiological data were prospectively collected from 1 November 2013 to 31 March 2017 and used to compare three risk assessment approaches.

RESULTS

A history of HUS/bloody diarrhoea/age under 6 years and faecal specimens positive for stx-predicted HUSEC with a diagnostic accuracy of 84% (95% CI; 81-88%). STEC isolated by Gastrointestinal Bacteria Reference Unit (GBRU) and stx2 and eae positive predicted HUSEC with a diagnostic accuracy of 99% (95% CI; 98-100%). Risk assessment combining these two tests predicts the most efficient use of resources, predicting that 18% (97/552) of cases would be eligible for follow-up at some stage, 16% (86/552) following local stx PCR results, 1% (7/552) following GBRU results of stx2 and eae status and 0.7% (4/552) following whole-genome sequencing. Follow-up could be stopped in 78% (76/97) of these cases, 97% (74/76) following second stage risk assessment.

CONCLUSIONS

This three-stage risk assessment approach prioritizes follow-up to HUSEC and minimizes unnecessary public health actions. We developed it into the algorithm for public health actions included in the updated PHE Guidance for management of STEC published in August 2018.

Optimization of Notification Criteria for Shiga Toxin-Producing Escherichia coli Surveillance, the Netherlands

We describe the consequences of 2 major changes in notification criteria for Shiga toxin–producing Escherichia coli surveillance in the Netherlands. The change to reporting acute, more severe infections appears to be a good compromise between workload, redundancy, and public health relevance, provided isolates remain available for typing and sequencing.

Do changes in STEC diagnostics mislead interpretation of disease surveillance data in Switzerland? Time trends in positivity, 2007 to 2016

BackgroundLaboratory-confirmed cases of Shiga toxin-producing Escherichia coli (STEC) have been notifiable to the National Notification System for Infectious Diseases in Switzerland since 1999. Since 2015, a large increase in case numbers has been observed. Around the same time, syndromic multiplex PCR started to replace other diagnostic methods in standard laboratory practice for gastrointestinal pathogen testing, suggesting that the increase in notified cases is due to a change in test practices and numbers.AimThis study examined the impact of changes in diagnostic methods, in particular the introduction of multiplex PCR panels, on routine STEC surveillance data in Switzerland.MethodsWe analysed routine laboratory data from 11 laboratories, which reported 61.9% of all STEC cases from 2007 to 2016 to calculate the positivity, i.e. the rate of the number of positive STEC tests divided by the total number of tests performed.ResultsThe introduction of multiplex PCR had a strong impact on STEC test frequency and identified cases, with the number of tests performed increasing sevenfold from 2007 to 2016. Still, age- and sex-standardised positivity increased from 0.8% in 2007 to 1.7% in 2016.ConclusionIncreasing positivity suggests that the increase in case notifications cannot be attributed to an increase in test numbers alone. Therefore, we cannot exclude a real epidemiological trend for the observed increase. Modernising the notification system to address current gaps in information availability, e.g. diagnostic methods, and improved triangulation of clinical presentation, diagnostic and serotype information are needed to deal with emerging disease and technological advances.

Isolate independent molecular typing improves the yield of O typing of infections due to Shiga toxin producing Escherichia coli

Molecular screening has increased detection of Shiga-toxin producing Escherichia coli (STEC). However, it is difficult to isolate the organism for epidemiological typing. We applied a molecular method for direct detection of nine O types from 110 stx positive faeces samples and compared the results with conventional isolate based methods. Using conventional methods 55/110 (50%) samples were O typed. Using the molecular method, 72/110 (65%) were O typed, including 23/38 (61%) culture negative samples. Combining both techniques typed 88/110 (80%) of samples. Molecular typing increased detection of O128 (2-25%, p<0.001), O26 (11-16%) O45 (0-6%) and O103 (1-6%) infections. Molecular typing of STEC direct from faecal samples improved O type yield; risk of bias in epidemiological and surveillance activities may be reduced by inclusion of culture independent typing methods.

Virulence Profiling and Molecular Typing of Shiga Toxin-Producing E. coli (STEC) from Human Sources in Brazil

Since no recent data characterizing Shiga toxin-producing E. coli (STEC) from human infections in Brazil are available, the present study aimed to investigate serotypes, stx genotypes, and accessory virulence genes, and also to perform pulsed-field gel electrophoresis (PFGE) and multi-locus sequence typing (MLST) of 43 STEC strains recovered from 2007 to 2017. Twenty-one distinct serotypes were found, with serotype O111:H8 being the most common. However, serotypes less frequently reported in human diseases were also found and included a hybrid STEC/ETEC O100:H25 clone. The majority of the strains carried stx1a as the sole stx genotype and were positive for the eae gene. Regarding the occurrence of 28 additional virulence genes associated with plasmids and pathogenicity islands, a diversity of profiles was found especially among the eae-harboring strains, which had combinations of markers composed of up to 12 distinct genes. Although PFGE analysis demonstrated genetic diversity between serotypes such as O157:H7, O111:H8, O26:H11, O118:H16, and O123:H2, high genetic relatedness was found for strains of serotypes O24:H4 and O145:H34. MLST allowed the identification of 17 distinct sequence types (STs) with ST 16 and 21 being the most common ones. Thirty-five percent of the strains studied were not typeable by the currently used MLST approach, suggesting new STs. Although STEC O111:H8 remains the leading serotype in Brazil, a diversity of other serotypes, some carrying virulence genes and belonging to STs incriminated as causing severe disease, were found in this study. Further studies are needed to determine whether they have any epidemiological relevance.

The epidemiology of Shiga toxin-producing Escherichia coli infections in the South East of England: November 2013-March 2017 and significance for clinical and public health

PURPOSE

This study describes the epidemiology of Shiga toxin-producing Escherichia coli (STEC) infections in a population in the South East of England.

METHODS

From 1 November 2013 to 31 March 2017 participating diagnostic laboratories reported Shiga toxin gene (stx) positive real-time PCR results to local public health teams. Stx positive faecal samples/isolates were referred to the Gastrointestinal Bacteria Reference Unit (GBRU) for confirmation by culture and typing by whole genome sequencing (WGS). Key clinical information was collected by public health teams.Results/Key findings. Altogether, 548 faecal specimens (420 were non-travel associated) were stx positive locally, 535 were submitted to the GBRU. STEC were isolated from 42 %, confirmed by stx PCR in 21 % and 37 % were PCR negative. The most common non-travel associated STEC serogroups were O157, O26, O146 and O91. The annualized incidence of confirmed STEC infections (PCR or culture) was 5.8 per 100 000. The ratio of O157 to non-O157 STEC serogroups was 1:7. The annualized incidence of non-O157 haemolytic uraemic syndrome-associated Escherichia coli (HUSEC) strains was 0.4 per 100 000. Bloody diarrhoea was reported by 58 % of cases infected with E. coli O157, 33 % of cases infected with non-O157 HUSEC strains and 12 % of other lower risk non-O157 strains. Overall, 76 % of non-O157 HUSEC isolates possessed the eae virulence gene.

CONCLUSIONS

HUSEC including serogroup O157 were uncommon and more likely to cause bloody diarrhoea than other STEC. The routine use of stx PCR testing can influence clinical management. Understanding the local epidemiology facilitates a proportionate public health response to STEC, based on clinical and microbiological characteristics including stx subtype(s).

xTAG Luminex multiplex assay for rapid screening of verocytotoxin-producing Escherichia coli strains

The O26 verocytotoxin-producing Escherichia coli (VTEC)-associated outbreak of hemolytic uremic syndrome (HUS) cases in Romania during 2016 showed the need to improve the current methodology of non-O157 VTEC detection and surveillance. An in-house assay based on xTAG Luminex technology was optimized to identify seven of the most relevant diarrheagenic E.coli serogroups (O-specific wzx genes), two convenient VTEC virulence markers (eaeA and ehxA genes), and a species-specific control gene (uidA). Twenty-nine strains previously characterized in terms of serogroup and virulence genes were tested with the optimized protocol and the results were as expected. The ratio of sample signal to background varied from 66.7 (ehxA) to 7.6 (uidA) for positive samples, with a cut-off of 3. Sensitivity varied depending on the target to be amplified from approximately 102 genomic copies to approximately 104 genomic copies per reaction, respectively. The current approach seems an affordable alternative to commercially available assays that can be further exploited to improve existing autochthonous strategies to prevent future VTEC outbreaks.

Towards understanding clinical campylobacter infection and its transmission: time for a different approach?

Campylobacter spp. are among the most commonly diagnosed causes of human infection. Methods for detection of the 29 campylobacter species have mainly focused on cultivation of the thermophilic species. More than 99% of clinical campylobacter isolates notified in the UK in the recent past have been from faecal samples and associated with gastroenteritis. Campylobacter enteritis notifications in temperate zones show a seasonal increase during the summer months with a sharp decrease in the winter months, a pattern which remains incompletely understood. The striking seasonality in the expression of many human genes, some concerned with inflammation and immunity, suggests a need for further study of the host regarding the temporal distribution of many human infections, including campylobacteriosis. A tendency for campylobacter to enter a non-cultivable state under adverse conditions effects a reduction in the number of isolations. A Polymerase Chain Reaction (PCR)-based screening approach for the presence of the Campylobacter genus and followed by speciation has provided some insight into the limitations of cultivation for campylobacter, also allowing the discovery of new species. The increased sensitivity of the PCR-based approach over culture-based methods may make it difficult for the laboratory to differentiate asymptomatic campylobacter carriage from clinical campylobacter infection in non-sterile body sites. Campylobacter infection depends on a combination of host factors, and on acquisition of a suitably virulent strain with a tropism for human epithelium. The possibility of persistence of campylobacter in a viable but non-culturable latent form in the human body may also require further investigation. The scope of this review includes a discussion of current methods for diagnosing acute campylobacter infection and for detecting campylobacter in water and foodstuffs. The review also questions the prevailing view that poultry is the most common source of campylobacteriosis.