Guidance

Surveillance protocol for the Clostridioides difficile ribotyping network (CDRN)

Published 25 July 2025

Background and purpose

The Clostridioides difficile ribotyping network (CDRN) for England and Northern Ireland was established in 2007. Its aim was to support NHS trusts by determining the scale and spread of high virulence and highly transmissible ribotypes in acute trusts in England. The CDRN provides timely access to Clostridioides difficile culture, ribotyping and DNA fingerprinting services, according to standardised submission criteria. National CDRN services are centralised at the UK Health Security Agency (UKHSA) Laboratory Leeds (Reference Laboratory, Leeds Teaching Hospitals Trust). Whole genome sequencing (WGS) techniques have since been developed allowing phylogenetic analysis with sufficient discriminatory power to assist local interpretation of transmission and outbreak patterns and guide subsequent interventions.  

Since the inception of the CDRN, there has been considerable heterogeneity in its use across England with a wide range in the proportions of CDI (Clostridium difficile infection) cases submitted between regions and between trusts. To provide a more structured sample of cases, and enhance the information available per sample, UKHSA has introduced active surveillance of C. difficile strains circulating in England and implemented WGS. This new ‘sentinel’ scheme using WGS is designed to identify newly circulating strains more rapidly than current surveillance does, and will study isolates from specific, named trusts. For all trusts, whether within the sentinel scheme or not, the CDRN will maintain the on-request service for polymerase chain reaction (PCR)-ribotyping or multi-locus variable number tandem repeat analysis (MLVA).   

Structure of the surveillance

Sentinel site surveillance

The active surveillance using WGS will start by using isolates from 20 ‘sentinel sites’. The sentinel sites were identified by epidemiological modelling work (1, 2), which calculated that sampling from these sentinel sites will identify a circulating novel strain of C. difficile 27% faster than if using 20 randomly chosen sentinel sites. These NHS trusts combine to intercept a significant proportion of the routes of C. difficile dissemination, when colonised or infected patients admitted to multiple facilities act as inter-hospital vectors. Thus, studying isolates from patients in these facilities gives the greatest likelihood of detecting emerging ribotypes.  

The sentinel surveillance scheme will provide the sentinel sites with C. difficile PCR-ribotyping data and WGS-inferred relatedness data (core-genome multi-locus sequence typing (cgMLST) distances) for their sentinel scheme samples but at no cost to the sentinel sites. 

For sentinel sites, the local diagnostic laboratory is expected to submit the first 10 C. difficile toxin-immunoassay positive stool specimens (detected by two-stage testing, ref SMI) per month to the CDRN for WGS. The counting of 10 isolates should start from the first of the month. Where there are fewer than 10 such isolates in a month, the laboratory team is expected to submit all of them.  

Please note that sentinel sites will not be able to request WGS of C. difficile isolates beyond the first 10 per month submitted for surveillance. However, samples beyond the first 10 can continue to be submitted via the on-request service for PCR-ribotyping or MLVA as for non-sentinel sites.  

Sentinel samples will be submitted to the CDRN in the usual way via the Electronic Requesting System.  

PCR-ribotyping will be performed on sentinel surveillance samples and results will be made available in the same way as for the CDRN standard service (trusts will be emailed automatically through the ERS to inform them that results are ready to view, within 14 days of sample receipt).  

Sentinel sites will receive a quarterly report summarising the genetic relatedness of C. difficile isolates originating from specimens referred by their centre.

Selecting sentinel surveillance samples

Specimens should be selected for submission using the following criteria:

All CDI cases reported on the data capture system (DCS) by the sentinel trust (including cases where the diagnostic test requester was outside the trust but for which the acute trust reports, for example cases from GP specimen, mental health and community trusts and so on) are eligible for inclusion in the sentinel surveillance submissions. The sentinel surveillance thus aligns with the modelling work to select sentinel trusts which are based on all reported CDI cases to the mandatory CDI surveillance DCS, including all apportionment categories.

Each sentinel site will submit specimens using the following criteria:

• the first 10 C. difficile-positive faecal specimens received at their laboratory per month
• deduplicated (no duplicate specimens from a patient within 28 days)
• C. difficile-positive – confirmed by local two-stage testing, including toxin immunoassay
• irrespective of prior trust apportionment category (for example, HOHA, COHA, COIA or COCA)
• irrespective of patient location at time of CDI-positive specimen (for example, NHS acute trust inpatient, A&E, GP or other community specimen)
• belonging to the sentinel trust only (where ‘sentinel trust’ is the same as the ‘reporting trust’ on the mandatory CDI surveillance DCS) and not from other trusts that testing laboratories may provide services to)

Please note:

• a maximum of 10 eligible specimens should be submitted per month
• where there are fewer than 10 such specimens in a month, all should be submitted
• if fewer than 10 specimens were submitted in the previous month, do not make up the shortfall in the following month
• please use the first day of each month as the point to start re-counting submissions
• if an eligible specimen cannot be found, or if insufficient remains and so on, please move on to the next eligible specimen

To define the ‘Sentinel Trust’ please use the same local methods employed in DCS submissions that would define the reporting trust for the mandatory surveillance of CDI. The ‘Sentinel Trust’ is the same as the DCS reporting trust. We understand some laboratories provide services to multiple hospital trusts; for such laboratories only specimens associated with episodes reported by the named ‘Sentinel Trust’ on the DCS are eligible for submission. We recommend that such laboratories liaise with their colleagues who submit DCS data in order to define their ‘Sentinel Trust’ (for example, a lookup list of GP practices for the Sentinel Trust or the hospitals included in the Sentinel Trust) to enable the laboratory team to identify the specimens that require referral for sentinel surveillance based on information supplied on the request at time of test request.

Non-sentinel site service 

Hospitals are still able to refer non-sentinel samples for PCR-ribotyping, using currently published criteria (clusters of CDI or increases in CDI frequency, severity, mortality or relapse rate). To ensure that CDRN services are maintained, and resources are available to support a local NHS problem such as a confirmed outbreak assigned to a specific ribotype or enhanced incident monitoring, all ribotyping submissions (that sit outside of the prescribed sentinel surveillance scheme) became chargeable in May 2024. These are charged at the rate published in the current UKHSA Microbiology Services Specialist Testing Price List.

After discussion with the service lead, hospital sites will remain able to request MLVA for enhanced sample relatedness on clusters of isolates where ribotyping data is insufficient – for example, when a hospital or trust with a high rate of CDI is identified with local commissioners, a hospital or trust is failing to meet its C. difficile target trajectory despite implementation and audit of control measures, or during a declared outbreak of CDI as agreed with the local health protection unit.

C. difficile isolates undergoing MLVA investigation will also be subject to internal WGS-based analysis. This will represent service development and will not be reported to users. WGS-based analyses will subsequently replace MLVA testing, following a period of internal validation.   

Submitting requests via the Electronic Reporting System (ERS) 

Data on samples for submission to the CDRN will be entered on to the Electronic Reporting System. ERS will pass data to MOLIS (Modular Open Laboratory Information System), the UKHSA Laboratory Information Management System (LIMS), for managing the information flows as the laboratory work is undertaken. ERS will store a MOLIS sample identifier along with the ERS identifier to allow linking to results. Requests for enhanced strain relatedness (MLVA) will continue via discussion with the service lead – this option is not available via ERS.       

Referral criteria for using CDRN ribotyping:

• increased frequency of cases or high baseline rates of CDI
• increased severity or complications of cases of CDI
• increased mortality associated with CDI
• increased recurrence rate of CDI

Table 1. Definitions of data items on ERS (patient clinical questionnaire)

Data item	Definition
Gender	Patient’s sex at birth
Number of antibiotics in 30 days before onset of this episode of C. difficile infection	The number of different antibiotics received by the patient prior to the date of diagnosis of the current episode of C. difficile infection
Was the patient admitted to an intensive care unit (ICU) due to C. difficile infection?	The patient has been admitted to a level 3 critical care unit following the date of diagnosis for the episode of C. difficile infection from which the sample originates
Did the patient die within 30 days of onset of symptoms?	Patient died of any cause within 30 days of date of diagnosis of CDI
If yes, was C. difficile a cause or contributory factor in death?	Based on clinical judgement, the CDI contributed to, or directly caused, the patient’s death
Did the patient have toxic megacolon?	There is radiological or sigmoidoscopic evidence of toxic megacolon following the diagnosis of CDI
Any surgical procedures as a consequence of the C. difficile infection?	The patient has undergone surgery to treat sequalae of CDI following the diagnosis of the current episode of infection

Results

WGS results

Data from WGS will be stored on an analytic database and will be used to detect sequences that (i) determine sequence type (ST), (ii) encode toxin genes, (iii) confer resistance to antibiotics and (iv) facilitate measurement of relatedness using core genome multi-locus sequence typing (cgMLST) and phylogenetic analysis.

At service launch, only a PCR-ribotype will be reported to hospital users within a specified turn-around time. Analysis of WGS data will be performed on the national sentinel data regularly and WGS-inferred relatedness summaries will be issued to individual sites quarterly.

In line with the UKHSA Pathogen Genomics Strategy, WGS data sets will be published to an International Nucleotide Sequence Database Collaboration (INSDC) partner for access by the wider science community three months after sample receipt. If potential outbreaks are detected ahead of the reporting schedule, referring sites will be approached separately. 

How you will receive results

PCR-ribotyping results (from both sentinel and on-request routes) and enhanced C. difficile isolate relatedness reports (MLVA) will be returned to end users as described for existing CDRN routine services. CDRN ERS will notify end users by email that PCR-ribotype results are available to view. The CDRN laboratory will directly notify end users that enhanced sample relatedness reports (MLVA) are available. The quarterly WGS-inferred pair-wise relatedness summaries will be sent via e-mail as an attached spreadsheet to key contacts at the sentinel site.

Detecting highly related isolates in whole genome sequencing results

The cgMLST profiles that will be generated from each sentinel sample allow pair-wise comparisons of relatedness, by counting the number of loci (short regions common to both genomes) at which the nucleotide sequence differs. The technique as implemented at CDRN was validated on historical C. difficile WGS data sets of characterised outbreaks. A ‘distance’ of 20 or fewer differing loci between sample pairs captured a very high proportion that were designated as outbreak-linked, according to the reference data set (true positives) (3). Sample pairs in the quarterly WGS-inferred relatedness summaries for a particular site, whose cgMLST profiles differ by fewer than 20 loci, should be investigated as part of a potential outbreak.

If a particular C. difficile strain emerged among hospitals under surveillance, it would manifest as an increasing proportion of sentinel samples, within the last 3 months, being related by fewer than 20 loci. Thus the CDRN surveillance protocol is to monitor for such clusters monthly. In addition, previously characterised genes encoding virulence factors will be screened for in surveillance sample WGS data.

Data sharing

Data on isolates referred to the CDRN laboratory will be shared within UKHSA for the purposes of the surveillance of C. difficile infection, including national surveillance and the investigation of local outbreaks of disease. Whole genome sequences per isolate will be deposited in BioProject PRJEB72420 via the European Nucleotide Archive 3 months after sample processing, in line with UKHSA policy on sharing WGS data. The CDRN service will share data internally as required to meet its public health remit.

Legal basis for collecting data

Legal permission to collect data for the CDRN without patient consent is granted by section 251 of the NHS Act 2006.

Contact details

For service and laboratory support, phone 0113 392 7134 or email leedsth-tr.clostridiumdifficile-ribotypingnetwork@nhs.net, or phone Dr Warren Fawley on 0113 392 8663 or email warren.fawley@nhs.net.

References

1. Ciccolini M and others. ‘Efficient surveillance for healthcare-associated infections spreading between hospitals’ Proceedings of the National Academy of Sciences of the United States of America 2014: volume 111, issue 6, pages 2,271 to 2,216

2. Pople D and others. ‘Network analysis of English hospital patients to select sentinel sites for efficient rapid identification of novel Clostridium difficile strains’ in ‘Congress of the European Society of Clinical Microbiology and Infectious Diseases 2024’, Barcelona, Spain

3. Eyre DW and others. ‘Comparison of multilocus variable-number tandem-repeat analysis and whole-genome sequencing for investigation of Clostridium difficile transmission’ Journal of Clinical Microbiology 2013: volume 51, issue 12, pages 4,141 to 4,149