Research and analysis

Emergence and outbreak of Clostridioides difficile ribotype 955 in England

Published 27 November 2025

Applies to England

This report presents the results of investigations carried out by UK Health Security Agency (UKHSA), working with stakeholders and partners, to investigate an outbreak of Clostridioides difficile infection (CDI) of ribotype 955 (RT955) mainly affecting the NHS Midlands region.

C. difficile is a bacterium, which causes a spectrum of disease in humans, from self-limiting mild diarrhoea to colitis that can result in death. Contracted via the faeco-oral route, C. difficile spores can be cross-transmitted from the environment to patients where infection prevention and control measures are suboptimal. RT955 is genetically similar to C. difficile ribotype 027 (RT027) (1, 2), which is associated with severe illness and higher mortality rates due to the hyper-production of the toxins which cause CDI. At the beginning of this century, CDI due to RT027 was responsible for widely reported healthcare-associated outbreaks including in the UK, Canada and the USA (3).

To mid-October 2025, the RT955 outbreak in England comprised 69 cases of CDI due to RT955, in 65 patients (Figure 1). There have been no further cases identified since April 2025. Fifteen NHS trusts have identified patients with CDI due to RT955. The outbreak appeared to be concentrated in the Midlands region (57 cases, 82.6% of cases), mainly in association with 2 large hospital clusters, and a third smaller cluster developing during 2024 in a hospital that had previously had 1 case (Figure 1). However, sporadic cases in the East of England, London, the South West, the North East and Yorkshire, and the Midlands NHS regions have been identified with no apparent links to the Midlands hospitals’ clusters or to the other NHS trusts with sporadic cases.

There was a slight male predominance (55.4%) amongst patients with C. difficile RT955 and most patients with CDI due to RT955 were aged 65 years or over (67.7%) (Figure 2). Most cases (72.5%) were of hospital-onset healthcare-associated (HOHA) attribution (as per UKHSA mandatory surveillance protocol definitions enhanced using Hospital Episode Statistics Admitted Patient Care (HES APC) to determine exposure to other acute trusts) (4), followed by community-onset healthcare-associated (COHA) (21.7%), with only 5.8% of cases being of community-onset of indeterminate association (COIA). There were no known community-onset, community-associated (COCA) cases (Figure 3). The overall 30-day all-cause case-fatality rate (CFR) among C. difficile RT955 cases (23.2%) was higher than the England 30-day all-cause CFR reported for all C. difficile cases over the same period of 2021 to 2024 (13.5%). The CFR for RT955 is comparable to that for HOHA CDI (19.6%), which is consistent with the high prevalence of HOHA attribution among RT955 cases (72.5%).

According to C. difficile Ribotyping Network (CDRN) Reference Laboratory data for England, C. difficile RT955 had been encountered in a single isolate in 2019 prior to the outbreak reported here. CDRN figures for 2018 to 2023 indicate that approximately 1 in every 2 reported cases of CDI are referred to CDRN for typing. Data available during this period shows only minor regional differences in the number of samples submitted to the service, relative to those submitted in previous periods. Submissions from the East Midlands, the North East and the North West UKHSA regions have been proportionally lower than from other regions (5).

Six C. difficile RT955 cases have been identified in Wales, 1 of which, diagnosed in February 2024, had previously been an inpatient on an outbreak ward in the Midlands. Two of 3 cases diagnosed in May and June 2022 had links to English hospitals outside the Midlands region (in the South West and North West regions), which had not had CDI RT955 cases; and 2 cases in care home residents in Wales, in July and September 2024, were linked to each other but had no known links to English hospitals.

There have been no known cases in Northern Ireland, and no known cases in Scotland since 2012.

Internationally, cases have most recently been reported in central and eastern Europe (1,2).

Figure 1. Distribution of C. difficile RT955 cases in England by time (September 2021 to July 2025)

Figure 1 footnote: In this graph, each colour represents a different NHS trust. Four of the 69 cases are relapses (that is: samples from the same person, over 28 days apart; denoted with an *).

Figure 2. Age-sex distribution of C. difficile RT955 patients in England (September 2021 to July 2025)

Figure 3. Distribution of C. difficile RT955 cases in England by time and prior NHS trust exposure (September 2021 to July 2025)

Figure 3 footnote: HOHA = hospital-onset, healthcare-associated; COHA = community-onset, community-associated; COIA = community-onset of indeterminate association (as per UKHSA mandatory surveillance protocol definitions enhanced using Hospital Episode Statistics Admitted Patient Care (HES APC) to determine exposure to other acute trusts) (4).

Hospital network diagram of patient movements

To produce a hospital network plot, specimen and admission dates were collected from the Mandatory Surveillance of Bacteraemia and CDI Data Capture System (DCS) and enhanced by the HES APC dataset to include discharge dates and capture admissions to other acute NHS trusts not recorded on the DCS. Specifically, the HES APC data added all inpatient admissions in all acute NHS trusts in England within 3 months (84 days) prior (6) to and after the CDI RT955 positive specimens were collected for patients with CDI due to C. difficile RT955. Emergency Department and outpatient attendances are not captured by HES APC. These patient admission histories were used to construct a hospital network diagram of patient movements. Admissions to acute NHS trusts in England during which a C. difficile RT955 positive sample was collected and, where applicable, admissions to acute NHS trusts in England known to have had cases of C. difficile RT955 within 84 days before and after the patient’s positive specimen have been plotted. The resulting network plot (Figure 4) demonstrates the relatedness of patient movements and of cases, particularly within the Midlands region, but also shows the apparently sporadic nature of some of the cases. Figure 1 and Figure 4 demonstrate that since June 2023 all patients with CDI due to RT955 in England have been identified within the Midlands region or have previously been inpatients at an outbreak hospital in the Midlands (n=2). This may be due to ascertainment bias because the diagnostic laboratories in the Midlands are aware of the RT955 outbreak within their region. The 2 hospitals outside the Midlands region, mentioned here, referred specimens to the CDRN because they were aware of the RT955 outbreak. In general, though, samples from community-onset cases are less likely to be sent for ribotyping, which also introduces bias.       

Figure 4. Hospital network diagram of C. difficile RT955 cases in England (September 2021 to July 2025)

Explanatory note for figure 4: Nodes (circles and triangles) represent each acute NHS trust in England that has diagnosed a case of CDI due to C. difficile RT955 between September 2021 and July 2025. Each NHS trust is anonymised by labelling by region and an arbitrary letter (for example, Midlands B). The colours of the nodes in Figure 4 match those in Figure 1 and represent the same NHS trusts in England. Using the dates of the specimens and admissions:

• triangles indicate NHS trusts where RT955 positive specimens were only collected in the community or within the first 48 hours of hospital admission thus designating the infections as community-onset
• circles indicate NHS trusts where at least 1 patient had a positive RT955 specimen that was collected after 48 hours of admission, designating the infection as HOHA

However, these NHS trusts may also include patients with community-onset infection. Node size is proportional to the number of patients with RT955 at that NHS trust. A patient can appear in the patient count of multiple NHS trusts, but only once per NHS trust. Lines represent consecutive admissions for a patient to different NHS trusts:

• solid black lines indicate that the NHS trust change occurred after the patient’s positive RT955 specimen date, with arrows representing direction of travel
• dashed red lines indicate that the NHS trust change occurred prior to the patient’s positive RT955 specimen date, also with directional arrows
• the numbers on the lines represent a specific patient (for example, patient 53, patient 4)

Typing tools and characterisation of cases

Multiple-locus variable number tandem-repeat analysis (MLVA) on 47 isolates representative of all 15 NHS trusts known to have had cases (September 2021 to March 2025) has continued to suggest that all isolates, except one, have originated from expansion of a single clone. Whole genome sequencing (WGS) of all 48 isolates identified to the end of November 2023 (representative of all 15 trusts known to have had cases), determined that they were from clade 2, which notably includes RT027.

RT027 is associated with a deletion in the tcdC gene responsible for downregulating the production of the toxins responsible for CDI. This deletion creates the potential for hyper-production of toxins, and RT027 is associated with severe disease and higher morbidity and mortality. WGS established that the RT955 isolates had toxin A, toxin B and binary toxin genes and a deletion in the negative regulator for toxin production.

The GeneXpert® system (Cepheid) returns a ‘presumptive 027’ result when toxin A, toxin B, binary toxin genes and the tcdC gene deletion are detected in combination. In fact, such results are compatible with the presence of RT955, which could be a valuable presumptive identification tool for diagnostic laboratories. However, C. difficile-positive samples very near to the limit of detection may result in a false negative ‘presumptive 027’ prediction. Therefore, a negative ‘presumptive 027’ Cepheid GeneXpert® assay does not rule out a RT955 case and re-testing and referral to CDRN is advised if the index of suspicion remains high.

Case-case study

The first UKHSA Briefing Note reporting on this outbreak (issued 29 November 2023) stated that C. difficile RT955 appears to transmit readily, may present with severe disease or as a recurrence, and that it has caused significant mortality. In the 2 large Midlands hospitals’ clusters, gastroenterology patients were particularly affected, which complicated identification of CDI because many were receiving laxatives.

A subsequent case-case study at these 2 NHS trusts matched 34 C. difficile RT955 cases by NHS trust and sample date with non-RT955, non-RT027 ribotyped cases. The study identified a statistically significant (p<0.05, Fisher’s exact test) association of C. difficile RT955 with alcohol dependency (p=0.003) and prior quinolone use (within 28-days p=0.021 and within 6 months p=0.001). Severity data could not be fully assessed due to the small case numbers and data limitations.

Three patients with C. difficile RT955 were admitted to ICU versus zero patients in the non-RT955 cohort (p=0.2). One patient in the CDI RT955 group developed toxic megacolon compared to zero patients in the non-RT955 group and 2 patients in the CDI RT955 group received intravenous immunoglobulin compared to 1 patient in the non-RT955 case group, though this difference did not reach statistical significance. Statistically significant associations were not found for other variables including 30-day and 60-day all-cause mortality, C. difficile-attributable mortality, disease relapse or recurrence (as per NICE definitions), laxative or PPI use.

The study adopted a pragmatic design and included the cases from these 2 NHS trusts until the end of 2023. The comparator arm included cases that had been sent to the CDRN reference laboratory for ribotyping and so was potentially biased towards more severe cases. In addition, cases were not matched for other gastroenterology patient/ward non-RT955, non-RT027 cases. Further, as patient numbers in this study are small (n=34 per arm), increasing the potential for confounding, results should be interpreted with caution.

Hospital outbreaks response and laboratory investigations

The 2 large hospital outbreaks in the Midlands were difficult to bring under control and required additional infection prevention and control measures: in particular, full ward closure and patient decant to facilitate use of hydrogen peroxide vapour.

Once the association between a Cepheid GeneXpert® ‘presumptive 027’ result and RT955 was identified, one trust implemented a change such that all glutamate dehydrogenase (GDH) positive samples were tested by Cepheid GeneXpert®, which enabled enhanced IPC efforts as the outbreak continued. The other trust tested all patients who had a GDH and toxin immunoassay positive combination with epidemiological links to an outbreak ward as well as GDH positive but toxin immunoassay negative specimens with Cepheid GeneXpert®. C. difficile RT955 positive patients were nursed for the duration of their stay with contact precautions if readmitted to hospital.

The recording of a webinar on CDI infection prevention and control, including a presentation on C. difficile RT955 by one of the affected NHS trusts, is available (the presentation starts at approximately 1 hour, 28 minutes, 32 seconds, into the recording).

UKHSA Porton Down has assessed the efficacy of commonly used sporicidal disinfectants against C. difficile RT955. Tests were performed to the method outlined in EN 17126:2018 for the assessment of sporicidal activity of chemical disinfectants in suspension. A peracetic acid-based and a range of chlorine-based disinfectant agents were tested against spore suspensions of C. difficile RT955 in the presence of either light or heavy loadings of organic matter. The light and heavy organic soiling conditions were used to simulate a clean environment and an environment which is heavily contaminated with organic matter respectively. The results obtained do not indicate that C. difficile RT955 has an increased tolerance to the investigated disinfectant agents in suspension.

These investigations only determined the susceptibility of C. difficile RT955 spores to the chemical formulations tested. The tests do not account for variance in cleaning practice or product use. Products must be prepared and used as described by the manufacturer to achieve effective sporicidal action. Importantly, some disinfectant products only show sporicidal activity in the presence of light organic soiling. When selecting a disinfectant agent, the product microbial efficacy data should be reviewed to establish whether the product has shown sporicidal activity against C. difficile in clean (light organic soiling) or dirty (heavy organic soiling) conditions. Good cleaning should always precede disinfection.

Susceptibility testing of 17 samples of C. difficile RT955 (13 isolates after deduplication), interpreted using EUCAST breakpoints (7), determined geometric mean minimum inhibitory concentrations (MIC) of selected antibiotics (Table 1) (8). Treatment of CDI with oral metronidazole is not recommended by NICE and it is no longer a first line treatment. Using breakpoints from the ClosER study (9), the geometric MIC means were determined for tigecycline, rifampicin, moxifloxacin, imipenem, and clindamycin (10). In a recent study of antimicrobial susceptibility of common or epidemiologically significant C. difficile ribotypes, the RT955 isolate was resistant to the greatest number of agents (10).

Table 1. Geometric mean of minimum inhibitory concentrations of selected antibiotics against Clostridioides difficile RT 955

	MTZ	VANC	FDX	RIF	MXF	CLIN	IMI	TIG
Ribotype 955 geometric mean MIC	3	0.7	0.3	19	23	104	15	0.1
EUCAST breakpoint mg/L	≤2	≤2	<1	≤0.004	≤2	≤2	≤4	<0.25

MTZ = metronidazole; VANC = vancomycin; FDX = fidaxomicin; RIF = rifampicin; MXF = moxifloxacin; CLIN = clindamycin; IMI = imipenem; TIG = tigecycline.

Recommendations

NHS staff should consider the possibility of CDI with RT955 in all cases of CDI, including community-onset cases, and especially those associated with severe disease or with links to affected NHS trusts and regions. Such patients’ samples can be confirmed locally as ‘presumptive 027’ positive using Cepheid GeneXpert®, if available, or submitted to the CDRN laboratory for ribotyping to improve the ascertainment of this ribotype. NHS trusts wishing to refer for free ribotyping as potential RT955 cases, or where there are exceptional circumstances, should discuss directly with the CDRN laboratory or their Consultant in Public Health Infection (CPHI) or, if in London, the Consultant Public Health Microbiologist (CPHM).

Approximately half of C. difficile cases are referred to the CDRN for ribotyping, but referral is not uniform across NHS trusts. The referral criteria to CDRN are:

• increased frequency of cases
• high baseline rates of C. difficile infection
• failure to meet a trajectory or target for CDI
• increased severity or complications in cases of C. difficile infection
• increased mortality associated with C. difficile infection
• increased recurrence rate of C. difficile infection

UKHSA recommends adherence to these criteria as best clinical practice. Further DNA fingerprinting using MLVA can be carried out on request if the criteria for enhanced fingerprinting are met.

Further information

Relevant guidance is available as follows:

• the National Infection Prevention Control Manual (NIPCM)
• the guidance Clostridioides difficile infection: how to deal with the problem
• Health and Social Care Act 2008 code of practice on the prevention and control of infections and related guidance
• UKHSA guidance on good antimicrobial stewardship
• NICE recommendations on Antimicrobial stewardship: systems and processes for effective antimicrobial medicine use
• UKHSA (2025). UK Standards for Microbiology Investigations: ‘Investigation of faecal specimens for Clostridioides difficile (Bacteriology B10)’

Acknowledgements

UKHSA would like to thank colleagues for their help and support in this investigation, particularly: clinical and IPC teams from affected NHS trusts, public health colleagues from Northern Ireland and Scotland and Wales, colleagues at the University of Leeds, and colleagues at NHS England.

References

1. Zikova J, Szarek K, Kabała M, Wultańska D, Frankowska N, Iwanicki A, and others (2025). ‘Newly emerging metronidazole-resistant Clostridioides difficile PCR ribotype 955 identified in Poland, 2021 to 2023 but not in Czechia, 2012 to 2023 and Slovakia, 2015 to 2023’. Eurosurveillance: volume 30, number 21

2. Stojanovic P, Kraakman M, Notermans D W, Groot J, Harmanus C, van Prehn J, and others (the Dutch National Expertise Centre for Clostridioides difficile group infections) (2025). ‘Emerging Clostridioides difficile strains belonging to PCR ribotype 955 in Serbia are distinct from metronidazole resistant RT955 outbreak isolates from the UK’. Emerging Microbes and Infections: volume 14, issue 1

3. He M, Miyajima F, Roberts P, Ellison L, Pickard D J, Martin MJ, and others (2013). ‘Emergence and global spread of epidemic healthcare-associated Clostridium difficile’. Nature Genetics: volume 45, number 1

4. UKHSA (2025). Quality and methodology information (QMI) for healthcare-associated infections (HCAI) reports

5. UKHSA (2024). Clostridioides difficile ribotyping network (CDRN) for England and Northern Ireland. Report 2018 to 2023

6. UKHSA. Mandatory enhanced MRSA, MSSA and Gram-negative bacteraemia, and Clostridioides difficile infection surveillance protocol

7. European Committee on Antimicrobial Susceptibility Testing (2025). ‘Breakpoint tables for interpretation of MICs and zone diameters’.

8. Puleston R, Roulston K, Morgan K, Hopkins S, Wilcox M, Fawley W, andothers (2024). Emergence of new concerning ribotype of Clostridioides difficile (955). Congress of the European Society of Clinical Microbiology and Infectious Diseases 2024 (Barcelona)

9. Freeman J ,Vernon J, Morris K, Nicholson S, Todhunter S, Longshaw C and Wilcox M H and the Pan-European Longitudinal Surveillance of Antibiotic Resistance among Prevalent Clostridium difficile Ribotypes’ Study Group (2015). ‘Pan-European longitudinal surveillance of antibiotic resistance among prevalent Clostridium difficile ribotypes’. Clinincal Microbiology and Infection:  volume 21, number 248

10. Clark E, Kaur K, Fawley W, Jeyaratnam D, Brown C, Hope R, Wilcox M, Freeman J (2025). Antimicrobial resistance in commonly-isolated Clostridioides difficile PCR-ribotypes (2025). Congress of the European Society of Clinical Microbiology and Infectious Diseases 2025 (Vienna)